[{"content":"A fresh global report on renewable electricity and batteries, covering industry history, value chains, market structure and operating challenges.  Version v2 · Information cutoff: 3 October 2026\nGlobal market. Renewable electricity and rechargeable batteries for vehicles and stationary storage. Public-material cutoff: 3 October 2026. Sources read: 6 October 2026.\nEvidence records were established before drafting. Bracketed identifiers link facts and figures to the independent Supporting Evidence appendix, ordered by first appearance. Original disclosures, official statistics, public research and third-party estimates are distinguished; economic synthesis is author analysis. Publication dates limit eligibility; event dates, statistical periods and reading dates are recorded separately. Exact release days are not invented where only a publication month or year is established. No research originals are redistributed.\nFigures are original redraws; source documents are linked to their publishers.\nPart 1 Industry Story Shortly after 3 p.m. on 16 January 2025, a fire began at Vistra’s 300 MW Phase I battery storage facility in Moss Landing, California. Monterey County’s emergency report recorded approximately 1,200 people evacuated and both directions of Highway 1 closed at Salinas Road. An installation built to support the electricity system had become a local emergency. [C01]\nThe consequences continued after the initial emergency. EPA’s September 2025 community plan described the destruction of the Moss Landing 300 building and federal oversight of Vistra’s battery removal. Vistra’s annual filing recorded an approximately USD 400 million asset write-off in the first quarter of 2025. Its next mid-year filing estimated removal and cleanup at USD 175 million as of 30 June 2026: USD 90 million incurred and USD 85 million still accrued for future costs. [C02][C03][C04]\nThose figures describe different things. The write-off removes an asset’s book value; the cleanup estimate concerns remediation obligations at a specified date. The June filing also disclosed fully collected insurance claims by February 2026 under business-interruption and property-loss policies with combined limits of USD 500 million, net of deductibles. Adding the write-off and cleanup estimate does not produce a verified final net loss. The cited accounting disclosures do not establish the damaged facility’s battery supplier. This one incident does not measure worldwide storage safety. [C03][C04]\nThe commercial lesson is broader than the accident. A battery maker sells equipment. A project owner must keep that equipment useful, available and safe over years of operation. Electricity users buy a service whose value depends on when power is available, not simply how much hardware was purchased. DEX’s central question is therefore how the renewable energy and battery value chain turns expanding manufacturing and deployment into dependable service and sustainable cash flow. That is an analytical question, not a finding about the cause of this fire.\nThis report follows two connected businesses: renewable electricity, with a focus on solar and wind, and rechargeable batteries for vehicles and stationary storage, with a focus on lithium-ion systems. Broader renewable-capacity figures retain their original definition, including technologies beyond solar and wind. Generation capacity, electricity output, battery energy, company revenue and future orders each answer a different question. They cannot be added into a single credible ‘industry size’ without eliminating overlap and matching the underlying boundaries.\nThe history explains how these products acquired scale. The market landscape shows who makes them, where demand is growing and what companies actually report as profit and orders. The challenges then return to the operating asset: performance, integration, financing, material supply, accountability and safety. The public-material cutoff is 3 October 2026; subsequent outcomes are outside this review.\nPart 2 Industry History Our reading of this history is that the industry grew by solving successive commercial problems. Promising physics needed repeatable products; those products needed evidence from operation; projects needed customers and payment arrangements. Each step widened the business from equipment production toward a chain connecting materials, manufacturing, construction, electricity networks and services. The following stages explain how those relationships took shape.\nStage 1 Discoveries become workable systems DOE dates Bell Laboratories’ photovoltaic cell to 1954 and describes early applications in remote telephone equipment and communications spacecraft. [H01] These were specific uses with specific customers. The business lesson we draw is that a device had to function where it was installed. Applying that lesson to a broader electricity market would require consistent manufacturing, durable equipment and installation capabilities alongside the ability to convert sunlight.\nWind followed an engineering path. The DOE–NASA Mod-Series, running from the late 1970s through the 1990s, tested turbine architectures and components. DOE’s account acknowledges that the program did not itself launch the commercial industry. [H02] We see a larger coordination problem in this history: blades, towers, drivetrains and grid operation have to work together. Component suppliers consequently influence lifetime reliability and the business conditions under which equipment can be operated.\nBattery development likewise depended on a complete system. The Royal Swedish Academy traces a rechargeable intercalation demonstration to 1976, a cobalt-oxide cathode breakthrough to 1979/1980 and Yoshino’s carbon-anode development to 1985. [H03] The early lithium-metal design and the later carbon-anode ion-transfer cell were different configurations. [H03] This progression suggests why electrode materials, electrolytes and safety testing belong together in a battery value chain: improving one element must preserve compatibility with the others.\nStage 2 Products earn trust through use The Academy places commercial lithium-ion battery release in 1991. [H03] Wind’s route to utility acceptance also involved demonstration: DOE and EPRI formed their Wind Turbine Verification Program in 1993, evaluating precommercial turbines through utility-hosted projects and collecting operating data. [H04] For business, we see commercialization adding a need to verify technology in use. Customers and financiers need a basis for judging maintenance, dependable output and operating risk before they can make long-term commitments.\nLFP illustrates the distance between a material result and a useful electrode. The publicly reproduced abstract of Padhi, Nanjundaswamy and Goodenough’s 1997 paper reports reversible lithium extraction and insertion in LiFePO4/FePO4, together with rate-related capacity limits. [H05] This is a research milestone, with commercial production still outside the evidence established here. The implication we draw is that discovering a workable chemistry opens further product-engineering tasks.\nA 2002 Nature Materials paper’s public abstract explicitly addressed LFP’s low electronic conductivity and reported a material-engineering approach. [H06] We use that record to identify a constraint researchers were trying to solve, without extending the abstract to every modern production process. The implication for today’s chain is analytical: value can sit in formulation and processing knowledge as well as in the material itself. A known chemical composition still leaves considerable room for product engineering.\nStage 3 Policy turns equipment into projects China adopted its Renewable Energy Law on 28 February 2005, with commencement on 1 January 2006. The 2009-amended text links targets and guaranteed purchase with grid development, storage and cost compensation. [H07] Those provisions are identified here as the amended framework. In our view, this framework shows how a power project acquires a commercial structure: the producer, network and payment mechanism need defined relationships. Equipment demand then depends partly on how those relationships work in practice.\nEurope’s 2009 renewable-energy directive set binding national targets consistent with at least 20% renewable energy in Community gross final energy consumption by 2020. Its grid provisions also addressed storage facilities. [H08] The percentage was a target for total final energy, covering more than electricity. We see this policy stage as a connection between demand creation and system integration. Networks and flexibility became relevant to delivering projects, even when the generating equipment was already technically workable.\nThe cost boundary widened too. DOE’s SunShot announcement on 4 February 2011 targeted roughly a 75% reduction in total PV system costs, including manufacturing, installation and permitting work. [H09] This was an announced objective. The business shift we identify is its system perspective: a cheaper cell addresses only one part of delivering electricity. Construction methods and administrative processes can also affect the cost of turning equipment into an operating asset.\nStage 4 Storage gains a route to market On 15 February 2018, FERC issued Order 841, requiring regional transmission organizations and independent system operators to establish participation models recognizing storage characteristics. Technically capable resources were to be eligible for capacity, energy and ancillary services. [H10] This concerned U.S. wholesale markets under FERC’s jurisdiction and covered storage technologies beyond batteries. [H10] The historical change was a rule for participation; actual implementation and a project’s financial outcome still require their own evidence.\nIn our view, market access adds another layer to the battery business. A grid project needs to turn physical capability into a service that can be scheduled, measured, settled and paid for. Integrators and operators consequently have to connect equipment choices with the services a particular market permits. The investment question expands from the quality of a manufactured cell to the quality of an operating arrangement, including how the asset earns revenue over time. Contracts and operating commitments therefore become part of our assessment of a storage business. Hardware specifications describe what the asset can do; the commercial arrangements help determine which capabilities are rewarded.\nSelected historical turning points and their commercial implications\nPeriod Historical event Meaning today — author analysis 1954–1985 PV and lithium-battery system advances [H01][H03] Compatible materials and reliable equipment underpin usable products 1991–2002 Commercial batteries, wind verification and LFP research [H03][H04][H05][H06] Manufacturing and operating evidence connect inventions to customers 2005–2009 China’s law and amendment; EU 2020 targets [H07][H08] Project revenue depends on policy and grid arrangements 2011 U.S. SunShot total-system-cost objective [H09] Installation and permitting belong in the cost equation 2018 U.S. Order 841 storage participation reform [H10] Services need market access and workable settlement Part 3 Market Landscape The value chain from resources to operating assets A useful way to understand this industry is to follow what changes hands. Upstream suppliers sell usable materials. Manufacturers turn them into equipment. Downstream businesses combine equipment with a site, an electricity connection and a customer. The following DEX analysis separates these roles because each creates value and carries risk differently. The representative companies illustrate business functions; placing two names in the same diagram does not establish a supply contract between them.\nUpstream resource access and chemical conversion Albemarle extracts lithium from brine and hard rock and converts it into compounds including carbonate and hydroxide for battery applications. [C11] Economically, extraction and conversion are separate questions: having a resource does not by itself mean having a saleable chemical product. For the next manufacturer, a useful input must arrive in the required form and quality. Upstream value therefore combines resource access, processing capability and dependable supply, rather than the resource alone.\nMany Albemarle customer contracts reference variable lithium-market indices. [C11] This distinction matters when interpreting long-term business. A contract can make a customer\u0026rsquo;s need more visible while leaving the supplier\u0026rsquo;s selling price exposed to the market. The questions are not only how much material can be sold, but also how the price is set and which costs remain when that price falls. More volume does not automatically mean more profit.\nThe other upstream branches supply manufacturing inputs and components. Crystalline-silicon solar manufacturing runs through polysilicon, wafers, cells and module assembly; IRENA’s tool separates material and conversion costs. This taxonomy describes crystalline silicon, not First Solar’s CdTe process. [G10] Wind equipment combines blades, towers and drivetrains, as the historical engineering record illustrates. [H02] Battery electrode materials form another branch before cells and packs. [H03][G09] In this functional classification, the customer is the next component or equipment maker; value comes from compatible inputs, usable quality and dependable delivery.\nMidstream equipment, batteries and systems This report groups module, turbine, battery and storage-system makers in the middle of the chain. The grouping is functional: their immediate commercial output is equipment that another business or household can use. It does not mean their products or factories are interchangeable. The purchasing decision changes with the equipment. Buyers must distinguish the product\u0026rsquo;s specifications from the completed project\u0026rsquo;s performance, and the quoted purchase price from the wider cost of putting it to work.\nFirst Solar makes cadmium telluride (CdTe) thin-film modules and generally sells them on a per-watt basis. [C07] Its buyer is purchasing generating equipment. The manufacturer\u0026rsquo;s revenue question concerns the volume, agreed price and recognition of those sales; the project owner\u0026rsquo;s question concerns the value of electricity produced over time. These questions are connected, but they are not the same. A competitive equipment offer must eventually fit a viable project, while a successful project does not reveal the manufacturer\u0026rsquo;s own margin.\nVestas offers turbine supply, installation or engineering, procurement and construction (EPC) arrangements, together with service contracts. [C08] Here the boundary of the purchase matters: supplying equipment, delivering a more complete project and maintaining it are different obligations. A wider contract can create more value for a customer by assigning responsibility more clearly, but it also extends what the supplier must deliver. Service deserves its own place in the chain because the relationship continues after the original equipment sale.\nCATL and LG Energy Solution represent battery manufacturing; Tesla\u0026rsquo;s Powerwall and Megapack illustrate storage products for household and larger commercial or utility uses. [C05][C06][C10] A battery supplier differs from an operating electricity asset. Selling the hardware transfers a product; owning the project leaves an ongoing task of deciding when and how to use it. The useful customer categories are therefore vehicle or system manufacturers, project businesses and end users. They should not be presented as verified contracts with the named companies.\nThe commercial boundary also changes within batteries: CATL\u0026rsquo;s products include cells, modules and packs, as well as storage cabinets and containers. [C05] In analytical terms, purchasing a component leaves more work with the buyer; purchasing a more complete package moves more responsibility into the supplied product. The customer\u0026rsquo;s decision therefore concerns the scope of the purchase as well as its price. A vehicle producer and a storage-project owner can both buy batteries while evaluating very different final products.\nDownstream development, electricity sales and operation Vistra combines power generation with retail electricity. [C03] This places it downstream of equipment manufacturing: its commercial role includes operating assets and serving electricity customers. The object being sold changes. A manufacturer is selling equipment or services; an electricity business must turn assets into a continuing customer offering. Construction, operation and customer delivery belong to different stages, even when one company participates in several of them.\nINDUSTRY MAP\nRenewable energy and battery value chain Global; examples retain their original geographic and business scope. Classification as of 3 October 2026.\nFit to view Expand all Collapse Full screen Read the full text outline Renewable energy and batteriesUpstreamMineral extraction and conversion [C11]Albemarle lithium [C11] Generation manufacturing inputs [G10,H02]Polysilicon; wafers [G10] Blades; towers; drivetrains [H02] Battery materials and components [H03,G09]Electrode materials [H03] Hard-carbon anodes [G09] MidstreamGeneration equipment [C07,C08,C12]First Solar; Vestas [C07,C08] JinkoSolar modules [C12] Battery cells and packs [C05,C06]CATL [C05] LG Energy Solution [C06] Storage systems [C10]Tesla Megapack [C10] Tesla Powerwall [C10] DownstreamProject development [C12]Masdar project [C12] Generation and storage operation [C03,C04]Vistra assets [C03,C04] Vehicle and household use [G03,C10]EV batteries [G03] Household storage [C10] Service and material recovery [C08,G03]Vestas service [C08] Recovered battery materials [G03] Connecting lines classify functions and do not establish supplier contracts. Companies can operate at several stages. The original figure and editable SVG are linked below.\nSources: [C11] Albemarle Corporation; United States Securities and Exchange Commission filing · [G10] International Renewable Energy Agency (IRENA) · [H02] U.S. Department of Energy · [H03] Royal Swedish Academy of Sciences; Olof Ramström, Nobel Committee for Chemistry · [G09] International Renewable Energy Agency (IRENA); collaboration with China Electric Power Research Institute (CEPRI) · [C07] First Solar, Inc.; United States Securities and Exchange Commission filing · [C08] Vestas Wind Systems A/S · [C12] JinkoSolar, official corporate news · [C05] Contemporary Amperex Technology Co., Limited (CATL) · [C06] LG Energy Solution, official Battery Inside publication · [C10] Tesla, Inc.; United States Securities and Exchange Commission filing · [C03] Vistra Corp.; filed with the United States Securities and Exchange Commission · [C04] Vistra Corp.; United States Securities and Exchange Commission filing · [G03] International Energy Agency (IEA). Reviewed 2026-10-06.\nFigure 1. Global renewable energy and battery value chain, classified at 3 October 2026. Company and product examples retain their original scope. Connecting lines classify functions; they do not establish supply contracts. Sources: [C03][C04][C05][C06][C07][C08][C10][C11][C12][G03][G09][G10][H02][H03].\nView the original mind map (PNG) · Download editable figure (SVG)\nRead the original figure text Renewable energy and batteries\nUpstream\nMineral extraction and conversion\nAlbemarle lithium [C11]\nGeneration manufacturing inputs\nPolysilicon; wafers [G10]\nBlades; towers; drivetrains [H02]\nBattery materials and components\nElectrode materials [H03]\nHard-carbon anodes [G09]\nMidstream\nGeneration equipment\nFirst Solar; Vestas [C07,C08]\nJinkoSolar modules [C12]\nBattery cells and packs\nCATL [C05]\nLG Energy Solution [C06]\nStorage systems\nTesla Megapack [C10]\nTesla Powerwall [C10]\nDownstream\nProject development\nMasdar project [C12]\nGeneration and storage operation\nVistra assets [C03,C04]\nVehicle and household use\nEV batteries [G03]\nHousehold storage [C10]\nService and material recovery\nVestas service [C08]\nRecovered battery materials [G03]\nA concrete connection is JinkoSolar\u0026rsquo;s announced agreement to supply Masdar with 2 GW of Tiger Neo modules for an Abu Dhabi project. [C12] Unlike a general list of customer types, this names both parties and the product quantity. It still describes a signed agreement, not completed delivery. The example shows how a developer\u0026rsquo;s plan can create an equipment purchase without making either the purchased quantity or the planned project already operating capacity.\nDownstream demand includes vehicles and stationary electricity users, while service and material recovery continue after equipment sale. The examples here are EV batteries, Tesla’s Powerwall for household storage, Vestas’ maintenance business and battery-recycling feedstock. [G03][C10][C08] DEX classifies vehicle and household use, project operation, maintenance and recovery separately because they have different customers and sources of value: product performance, electricity service, equipment availability and reusable material. The diagram shows these functions rather than a verified chain of purchases among all named companies.\nA growing market with several measures A solar factory, a battery supplier and a power project sell different things. Their markets therefore need separate measures: generating capacity installed, electricity produced, battery energy deployed and equipment sold. A combined revenue headline would require a consistent boundary to avoid counting intermediate transactions twice. The figures below show where activity is expanding and which part of that activity each figure actually measures.\nIRENA’s same-edition tables put global renewable generating capacity at 5,149.280 GW at the end of 2025, against 4,457.340 GW a year earlier. [G01] The difference is 691.940 GW, or 15.5% growth. [G01] Solar capacity stood at 2,391.584 GW and wind at 1,291.368 GW. [G01]\nThe expansion was uneven: China, the United States and the European Union together represented 79.5% of 2025 renewable additions; Africa represented 1.6%. [G01] These are installation locations, rather than the nationalities of equipment suppliers. Author’s interpretation: regional opportunity depends on where projects can be completed, alongside where equipment can be manufactured.\nElectricity provides a different perspective. Global demand reached 28,600 TWh in 2025, up 3%; renewables supplied 33% of electricity generation. [G07] Capacity measures the size of generating installations, while generation measures their energy output. The two percentages and growth rates answer different questions; neither directly measures equipment revenue.\nBattery demand and factory capacity have different boundaries Global EV battery deployment reached 1.2 TWh in 2025, growing almost 30%. [G03] Global lithium-ion cell nameplate manufacturing capacity exceeded 4 TWh at year-end. [G03] Deployment increased in China and the EU but stagnated in the United States. [G03]\nSelected regions, 2025; percent. Deployment follows vehicle sales; year-end capacity follows factory locations. Regions are not exhaustive.[G03]\nRegion Share of EV battery deployment Share of cell nameplate capacity China 60% [G03] Over 80% [G03] European Union Nearly 15% [G03] 6–7% [G03] United States 10% [G03] 6–7% [G03] MARKET SHARE · January–August 2026\nVehicle battery usage by supplier EV, PHEV and HEV registered during the period; stationary storage is excluded. Total installed battery energy: 844.2 GWh. This is a vehicle-battery segment, not the entire renewable-energy industry.\n80 countries covered by SNE Research · Share of battery energy in registered EV, PHEV and HEV vehicles (%)\nFull screen View the data table Vehicle battery usage by supplier · January–August 2026 · Share of battery energy in registered EV, PHEV and HEV vehicles (%) SupplierShare CATL39.4%BYD15.1%LG Energy Solution8.1%CALB5.3%Gotion4.9%Other suppliers27.2% Download data (CSV) Other suppliers = 100 − 39.4 − 15.1 − 8.1 − 5.3 − 4.9 = 27.2%; the five named suppliers total 72.8%. SNE publishes CATL plus BYD at 54.6%; its one-decimal company values sum to 54.5%. The reason is unverified. Company values are not renormalised. Other suppliers is the complement of the five shown, not the source table’s separate Others category.\nSource: SNE Research — Global EV and Battery Monthly Tracker — January to August 2026 (2026-10-02). Reviewed 2026-10-06.\nFigure 2. Supplier shares of battery energy in registered EV, PHEV and HEV vehicles, January–August 2026, 80 countries. Total: 844.2 GWh. Shares in percent. Other suppliers is a calculated residual; stationary storage is excluded. Source: SNE Research [M01].\nView the original ring chart (PNG) · Download editable figure (SVG) · Download complete chart data (CSV)\nRead the original figure text CATL: 39.4%\nBYD: 15.1%\nLG Energy Solution: 8.1%\nCALB: 5.3%\nGotion: 4.9%\nOther suppliers: 27.2%\nOther suppliers = 100 − 39.4 − 15.1 − 8.1 − 5.3 − 4.9 = 27.2%.\nPublished CATL + BYD total: 54.6%; rounded components: 54.5%; cause unverified.\nThese figures cannot establish a global utilisation rate: EV deployment is one application, whereas cell capacity serves a wider market. [G03] Author’s interpretation: a new factory announcement should be assessed against customer qualification, achievable output and intended applications. A large building and an expanding end market do not by themselves show that a particular production line will earn a return.\nVehicle battery competition within a defined market SNE Research’s release of 2 October 2026 estimates 844.2 GWh of battery energy in registered EV, PHEV and HEV vehicles across 80 countries from January to August 2026, up 19.7% year on year. Its supplier table places CATL at 39.4%, BYD at 15.1%, LG Energy Solution at 8.1%, CALB at 5.3% and Gotion at 4.9%. These are shares of that vehicle-battery usage measure, not of all renewable energy, all batteries or company revenue. [M01]\nThe chart uses the five published rounded supplier values. ‘Other suppliers’ is the calculated residual, 27.2%, covering every supplier outside those five. SNE separately publishes 54.6% for CATL plus BYD, while its one-decimal components sum to 54.5%; the exact reason is unverified. This review preserves the table’s precision and explains the discrepancy. It does not adjust the company values to force a different combined total. [M01]\nLower prices change economics at different levels BloombergNEF’s 2025 survey estimated an average lithium-ion battery-pack price of USD 108/kWh, down 8%. [G06] Stationary-storage packs averaged USD 70/kWh, down 45%, while battery-electric vehicle packs averaged USD 99/kWh. [G06] These are survey estimates for different application groups; a storage pack price is one component of a complete project’s cost.\nIRENA’s 2025 cost study places four-hour utility-scale battery storage installed cost at USD 140/kWh, and four-hour turnkey equipment cost at USD 111/kWh. [G02] The latter excludes engineering, procurement and construction, grid connection and development. [G02] Different samples and cost boundaries prevent these averages from becoming an itemised cost breakdown for an individual project.\nFor projects commissioned in 2025, IRENA reports global weighted-average generation costs of USD 44/MWh for standalone utility-scale solar PV, USD 33/MWh for onshore wind and USD 78/MWh for offshore wind. [G02] These lifetime cost estimates describe generation, rather than the price a customer agrees to pay. Author’s interpretation: falling equipment costs can improve a proposal while leaving its revenue case unresolved.\nThe financial comparison therefore starts with the product being purchased. A battery pack, an installed storage system and delivered electricity have different units and obligations. Author’s interpretation: buyers must compare the required service, financing terms and operating exposure before concluding that a cheaper component makes the whole project cheaper or more profitable.\nRegional competition extends beyond the final factory Battery-pack prices differed geographically in BNEF’s 2025 survey: China averaged USD 84/kWh; North America and Europe were respectively 44% and 56% higher. [G06] Regional averages include different product mixes. They provide a market benchmark, rather than a controlled comparison of two otherwise identical factories.\nIEA’s 2024 supply-chain baseline attributes about 85% of solar PV and 80% of lithium-ion battery manufacturing capacity to China on a value-weighted basis across stages, excluding mining. [G04] Its model identifies manufacturing efficiency as explaining over 40% of the Europe–China battery production-cost gap. [G04] These are specified capacity and model measures, rather than current-year equipment sales shares.\nThe upstream investment cycle also matters. Critical-mineral investment fell 9% in 2025, with battery-metal capital expenditure declining over 20%. [G05] Across the minerals examined, excluding rare earths, the largest refining country’s average share reached 72%, compared with 70% in 2023. [G05] An average across minerals does not describe one country’s share of every material.\nAuthor’s interpretation: localisation should be judged as a functioning production system. A final assembly plant may offer jobs and shorter delivery routes, yet its economic case still depends on reliable inputs, equipment, yields and customer acceptance. Likewise, a decline in mineral investment is a warning about future capacity development, rather than proof that present production has already fallen.\nProfitability depends on the business being measured The figures below cover fiscal years ended 31 December 2025 and retain each company\u0026rsquo;s currency and accounting scope. They illustrate where revenue and profit appear along the chain; they do not form a market-share calculation or a profitability ranking. One column cannot make net profit, operating income, adjusted EBITDA and gross margin equivalent.\nSelected company indicators, FY2025; original currencies and distinct measures\nCompany and scope FY2025 revenue Reported profitability measure Evidence Albemarle — Energy Storage lithium segment $2.710 billion net sales $697 million adjusted EBITDA [C09] CATL — consolidated group RMB423.7 billion RMB72.2 billion profit attributable to listed-company shareholders [C05] LG Energy Solution — consolidated business KRW23.7 trillion KRW1.3 trillion operating profit, including North American production incentives [C06] First Solar — consolidated group $5.219 billion net sales $1.597 billion operating income [C07] Vestas — consolidated group €18.822 billion 5.7% EBIT margin before special items [C08] Tesla — energy generation and storage segment $12.771 billion 29.8% gross margin; manufacturing credits reduced cost of revenue by $1.12 billion [C10] Physical scale also has several meanings. CATL reported 661 GWh of lithium-ion battery sales in 2025 across power and storage applications; Tesla reported 46.7 GWh of storage deployments. [C05][C10] These are company operating measures with different product boundaries. Neither is a compatible numerator for the January–August 2026 vehicle-battery share market used above.\nThe practical comparison is a set of questions, not a league table. What is inside the reported business? Which costs sit above or below the chosen profit measure? Are incentives included? These boundaries must come before any conclusion about economic strength. A large revenue figure may describe a broader business, while an attractive reported margin may reflect a different accounting level. Neither alone proves that one equipment category is inherently more profitable than another.\nAn order book is a schedule of obligations First Solar\u0026rsquo;s year-end future sales contracts covered 50.1 GW and $15.0 billion, with revenue expected through 2030. [C07] This provides both a physical quantity and a financial value, spread over future fulfillment. It offers commercial visibility, but the next question is which deliveries and recognition milestones fall in each year. The full contract value cannot be placed beside one year\u0026rsquo;s revenue as if both were annual sales.\nIn its review of 2025, LG Energy Solution disclosed 140 GWh of ESS backlog and more than 300 GWh for its 46-series business, the latter explicitly at year-end. [C06] These are separate quantities, not a verified combined delivery total. Vestas reported year-end 2025 turbine backlog of €33.2 billion and service backlog of €38.7 billion. [C08] The service figure represents a continuing relationship rather than a single equipment shipment. These quantities cannot be translated into annual deliveries without fulfillment schedules. Keeping the categories and horizons visible prevents different backlogs from being treated as the same kind of near-term demand.\nThese models show where the industry\u0026rsquo;s commercial chain can pause: a material contract may have a moving price, an equipment order still awaits fulfillment, and an operating asset remains an ongoing responsibility. The next useful question is what must happen after the reported number: payment, production, delivery, acceptance or continued operation. The chain is economically meaningful when those obligations can be met, not merely when an impressive quantity is announced.\nPart 4 Industry Challenges A cheaper cell still has to perform inside a system IEA’s 2026 EV report described almost- and all-solid-state designs as prototypes. [G03] DEX distinguishes that status from proven mass-production economics, service life or safety. A useful evaluation must ask which performance has actually been demonstrated, in what configuration and at what scale. A promising chemistry still needs repeatable production and evidence from the application where it will be used.\nSodium-ion offers another route to material diversification, but resource abundance does not settle energy density, hard-carbon processing or manufacturing cost. IRENA’s technology review identifies stationary storage as a potential application where weight matters less. Its comparisons use different underlying dates, so they do not establish a universal 2026 cost advantage over lithium-ion. [G09] DEX’s interpretation is that alternatives should be evaluated for a particular duty cycle and delivery capability, rather than a chemistry label alone.\nGeneration cost is different from dependable supply A low generating cost measures one part of the service. Moving electricity to a different hour requires storage or another source of flexibility; making supply fit demand can require extra generation, networks and operating coordination. The relevant cost boundary therefore changes when the customer asks for dependable output instead of energy whenever it is available. Comparing a standalone generation benchmark with a complete supply service can misstate the economic advantage.\nIRENA illustrates the difference with a hypothetical 100 MW Las Vegas solar case using 2025 costs. Its model reaches a 95% annual energy-coverage target at USD 113.2/MWh, versus USD 43.5/MWh for standalone solar, using 591.8 MWh of storage and 61.9 MW of additional solar. The coverage target is a share of flat annual demand, not a guarantee of uninterrupted supply in every hour. These are local model results, not an operating project or a global price. [G08]\nTiming also affects realised revenue. IEA reports negative wholesale prices in roughly 20% of hours in South Australia and California in the first half of 2026. [G07] Storage may help shift energy away from an oversupplied hour, but that statistic alone does not prove profitable arbitrage. DEX’s analysis requires the actual charging and discharge prices, energy losses, cycling-related wear, availability, network charges and contract restrictions. Connection capacity and market rules can limit a technically capable system’s earning opportunities.\nDiversification needs more than a new factory IEA’s supply-chain analysis finds upstream bottlenecks outside China even where downstream manufacturing is being diversified. Its removal-of-the-largest-supplier exercise is a conditional stress test with specified utilisation assumptions, not a forecast that disruption will occur. [G04] DEX’s interpretation is that building final assembly locally may leave processing equipment, intermediate materials, skills or upstream production dependent on another location. Diversification must follow those dependencies through the chain.\nLower commodity prices can create the opposite incentives for buyers and upstream investors. IEA reports that critical-mineral investment fell 9% in 2025, while battery-metal capital expenditure fell more than 20%. It also identifies cost, equipment, technology and skills constraints for new refining projects. [G05] Cheaper inputs may help current equipment buyers, while weaker investment can reduce the options available for future supply. That is a risk mechanism, not proof of an inevitable shortage.\nRecycling can provide another material source, but its feedstock arrives on a different timetable from new factories. IEA’s 2026 EV report says battery recycling still relied mainly on production scrap. [G03] DEX therefore distinguishes a plant’s announced recycling capacity from the batteries actually available to process. Collection, chemistry, recovery quality and customer acceptance must also fit. Recycling can complement new supply; a capacity announcement alone cannot close the material balance.\nPolicy and traceability alter the commercial calculation Policy support has different expiry dates and accounting effects. The enacted US Public Law 119-21 terminated the Section 30D clean-vehicle credit for vehicles acquired after 30 September 2025. [R04] This is a specific acquisition-date rule, not evidence that all US energy support disappeared. In DEX’s analysis, changes to eligibility can affect the customer’s economics and the timing of purchases, while production incentives shown in company accounts affect a different part of the chain. Their effects should not be combined into one unsupported demand forecast.\nThe EU moved the application date of battery due-diligence obligations from 18 August 2025 to 18 August 2027 in its adopted July 2025 amendment. The official rationale included preparation time for verification bodies and supply-chain arrangements. [R01][R02] The delay does not postpone every battery requirement or establish that a supplier complies. Traceability is a practical ethical issue: documentation must support what materials were bought and how the associated risks were examined, rather than merely repeat a manufacturer’s assurance.\nDEX’s analysis is that a longer preparation period is useful only if it improves the evidence behind purchasing decisions. Supplier identification, audit scope, disputed records and corrective action need clear owners. Certification can support a defined assessment but cannot answer every social or environmental question for every shipment. Poor verification can leave an operator with obligations it did not price into the project, while incomplete records make meaningful accountability harder.\nSafety has to survive construction and operation DOE’s April 2024 storage safety strategy explains why chemistry and control systems cannot settle the whole problem. LFP retains failure risks; high charging rates can contribute to lithium plating. A battery management system has limited ability to intervene once thermal runaway has begun. [R03] DEX’s implication is that protection must be layered across design, installation, monitoring, emergency response and end-of-life handling. Controls that prevent one failure mode do not guarantee containment of every other one.\nThe scope of a test also matters. UL Solutions’ 16 April 2025 announcement describes UL 9540A as a thermal-runaway fire-propagation test method, while UL 9540 addresses complete storage-system safety criteria. [R05] This report has read the public announcement, not the paid standards. Test results should therefore be tied to the tested configuration and the assessment they support; they are not a blanket promise that any installation is risk-free.\nMoss Landing makes the financial consequence tangible without resolving the cause of the fire. The asset write-off and the later cleanup estimate belong to different accounting categories. [C03][C04] For a project decision, DEX would connect the purchase price to the obligations that remain: contracted performance, connection and dispatch limits, degradation, emergency arrangements, insurance terms and eventual removal. Faster manufacturing and larger order books are useful inputs. The decisive question is whether the particular asset can deliver its promised service under those conditions.\nAppendix Supporting Evidence Bracketed identifiers connect the article and figures to the records below. Each record states the claim supported, source location, scope, dates, methods and limits. Source documents are linked to their publishers; original research files are not redistributed.\n[C01] Moss Landing local-emergency ratification report, File 25-048 Supports: Monterey County reported a fire shortly after 3 p.m. on 16 January 2025 at Vistra\u0026rsquo;s 300 MW Phase I battery facility. Approximately 1,200 people were evacuated, and both directions of Highway 1 were closed at Salinas Road.\nInstitution or author: County of Monterey, Office of County Counsel / Board of Supervisors\nPublication date: File created 17 January 2025; agenda dated 21 January 2025; exact online posting day is not separately displayed\nEvent date: Fire and evacuation on 16 January 2025; local-emergency proclamation on 17 January 2025\nStatistical period: Incident status as of 17 January 2025; annual statistical period not applicable\nPage or section: Summary/Discussion, two incident paragraphs in the Full Text view; File 25-048\nGeography: Moss Landing, Monterey County, California, United States\nUnits: MW of facility electrical power; approximate count of evacuated people\nMarket definition: One Phase I battery energy storage facility and its immediate local consequences\nEvidence classification: Primary local-government incident and emergency report\nRead date: 2026-10-06\nMethod and calculation: Directly read the original Full Text report; no calculation\nLimits and uncertainty: Early incident account: it does not establish the fire\u0026rsquo;s cause, lasting health effects, subsequent restart status or the frequency of battery fires worldwide. The evacuation count is approximate and time-specific.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: Moss Landing local-emergency ratification report, File 25-048\n[C02] Moss Landing Battery Fire Community Involvement Plan Supports: EPA\u0026rsquo;s September 2025 plan states that the January 16 fire destroyed the Moss Landing 300 building. At California\u0026rsquo;s request, EPA was overseeing Vistra\u0026rsquo;s battery removal, including making batteries safer, packaging them and arranging recycling or disposal.\nInstitution or author: United States Environmental Protection Agency, Region 9\nPublication date: September 2025; the PDF does not establish an exact day\nEvent date: Fire on 16 January 2025; EPA–Vistra agreement in July 2025; removal-plan approval in August 2025\nStatistical period: Cleanup-plan status in September 2025; annual financial period not applicable\nPage or section: Introduction, printed pp. 4–5 (PDF page 3); Site History, printed pp. 6–7 (PDF page 4); removal oversight and process, printed pp. 8–11 (PDF pages 5–6)\nGeography: Moss Landing, California, United States\nUnits: Incident and cleanup responsibilities; quantitative market unit not applicable\nMarket definition: The damaged Vistra facility and supervised battery-removal process, rather than all storage systems\nEvidence classification: Primary federal-government incident-cleanup plan\nRead date: 2026-10-06\nMethod and calculation: Read original PDF text by the identified sections; no calculation\nLimits and uncertainty: Not a fire-cause or epidemiological study. A peripheral community paragraph says January 15, inconsistent with this document\u0026rsquo;s main sections and C01, which identify January 16; that peripheral date is excluded. The plan is not proof of completed cleanup.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: Moss Landing Battery Fire Community Involvement Plan\n[C03] Vistra Corp. 2025 Annual Report on Form 10-K Supports: Vistra combines retail electricity with power generation. Its 2025 annual report records an approximately $400 million Moss Landing 300 net-book-value write-off in first-quarter depreciation expense. That accounting charge is not the final net economic cost of the incident.\nInstitution or author: Vistra Corp.; filed with the United States Securities and Exchange Commission\nPublication date: 27 February 2026, SEC filing date\nEvent date: Fire on 16 January 2025; write-off recorded in the first quarter of 2025\nStatistical period: Fiscal year from 1 January to 31 December 2025\nPage or section: Item 1, Business—General, printed p. 1 (PDF page 9); Note 8, Loss Events and Insurance Recoveries—Moss Landing 300 Incident, printed pp. 107–108 (PDF pages 115–116)\nGeography: United States company operations; incident in California\nUnits: USD million; accounting asset write-off\nMarket definition: Vistra consolidated business and one storage asset\u0026rsquo;s net book value, not storage-market revenue\nEvidence classification: Primary regulatory annual report with audited financial statements and management disclosures\nRead date: 2026-10-06\nMethod and calculation: Read the issuer-hosted filing PDF and SEC filing index; $400 million equals $0.4 billion\nLimits and uncertainty: Asset write-offs, cleanup estimates, lost revenue and insurance recoveries differ. Do not sum them as net loss. The year-end cleanup estimate is superseded by C04; the cited accounting disclosures do not establish a battery supplier for this facility.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: Vistra Corp. 2025 Annual Report on Form 10-K\n[C04] Vistra Corp. June 2026 Quarterly Report on Form 10-Q Supports: As of 30 June 2026, Vistra estimated Moss Landing removal and cleanup costs at $175 million, including a $65 million second-quarter increase. It had incurred $90 million, leaving $85 million accrued for future costs. These are dated estimates, not a completed project\u0026rsquo;s final bill. The same note discloses fully collected insurance claims by February 2026 under policies with combined limits of USD 500 million, net of deductibles; these are separate from the cleanup estimate.\nInstitution or author: Vistra Corp.; United States Securities and Exchange Commission filing\nPublication date: 10 August 2026, SEC filing date\nEvent date: Moss Landing incident on 16 January 2025; cleanup-estimate revision in the second quarter of 2026\nStatistical period: Quarter and six months ended 30 June 2026; estimates measured at that date\nPage or section: Note 9, Loss Events and Insurance Recoveries—Moss Landing 300 Incident; original HTML Note 9 insurance paragraph, printed p.17\nGeography: Moss Landing, California, United States\nUnits: USD million; incurred cost, estimated total cost and future-cost accrual\nMarket definition: Battery removal, demolition and monitoring covered by the cleanup obligation; not all incident losses\nEvidence classification: Primary regulatory interim report with unaudited financial statements\nRead date: 2026-10-06\nMethod and calculation: Read Note 9 and the filing index; retain reported estimates without projecting beyond June 2026\nLimits and uncertainty: The increase reflects revised remediation timelines and costs, not a conflict with the earlier $110 million estimate. Insurance, litigation and additional operating effects remain separate. The filing did not establish completed cleanup or a definite restart date.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: Vistra Corp. June 2026 Quarterly Report on Form 10-Q\n[H01] On Display: Smithsonian Shares the History of Solar Supports: DOE dates Bell Laboratories\u0026rsquo; photovoltaic cell to 1954 and describes early uses in remote telephone equipment and communications spacecraft.\nInstitution or author: U.S. Department of Energy; author Charlie Gay, then Solar Energy Technologies Office director\nPublication date: 2017-03-01\nEvent date: 1954 Bell Laboratories PV cell; early remote-communications applications; Telstar 1 in 1962\nStatistical period: Historical milestones, not a statistical series\nPage or section: HTML lines 9 and 17–19; photovoltaic-history paragraphs read\nGeography: United States; these examples do not establish a worldwide commercialization date\nUnits: Calendar years; no market-share or capacity unit used\nMarket definition: Photovoltaic electricity conversion; distinct from concentrating solar thermal power\nEvidence classification: Official historical synthesis, not a contemporaneous 1954 experiment\nRead date: 2026-10-06\nMethod and calculation: Direct reading of dated DOE article. No numerical calculation.\nLimits and uncertainty: Supports dated examples only. The article\u0026rsquo;s 1958 first-U.S.-satellite wording is excluded; no earliest-invention claim or present-day efficiency claim is adopted.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: On Display: Smithsonian Shares the History of Solar\n[H02] From 1970s Pioneers to Today’s Wind Industry, Aerospace Researchers Championed Wind Energy Supports: The DOE–NASA Mod-Series ran from the late 1970s through the 1990s. It tested turbine architectures and components, but did not itself launch the commercial wind industry.\nInstitution or author: U.S. Department of Energy\nPublication date: 2024-04-23\nEvent date: Late 1970s–1990s Mod-Series engineering program\nStatistical period: Program history, not a market-wide time series\nPage or section: HTML lines 9, 16–24, 34–38 and 50–60; program aims, limitations and engineering contributions read\nGeography: United States, drawing on U.S. and European turbine research\nUnits: Calendar years; no production or market-share calculation\nMarket definition: Utility-scale wind turbine R\u0026amp;D, including blades, towers, drivetrains and grid operation\nEvidence classification: Official historical synthesis by the sponsoring department\nRead date: 2026-10-06\nMethod and calculation: Direct reading of dated DOE article. No calculation.\nLimits and uncertainty: U.S. program contribution only; no universal first-turbine attribution, sole-cause claim or current component-market share.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: From 1970s Pioneers to Today’s Wind Industry, Aerospace Researchers Championed Wind Energy\n[H03] Scientific Background on the Nobel Prize in Chemistry 2019: Lithium-Ion Batteries Supports: The Academy traces a rechargeable intercalation battery demonstration to 1976, the cobalt-oxide cathode breakthrough to 1979/1980, Yoshino\u0026rsquo;s carbon-anode development to 1985, and commercial release to 1991.\nInstitution or author: Royal Swedish Academy of Sciences; Olof Ramström, Nobel Committee for Chemistry\nPublication date: 2019-10-09\nEvent date: 1976; 1979/1980; 1985; 1991\nStatistical period: Historical scientific milestones\nPage or section: PDF cover and printed pp. 5–10 (PDF pages 6–11); text on intercalation, carbon anodes and commercial release read; references pp. 12–13 checked\nGeography: Research and commercialization examples in the United States, United Kingdom and Japan\nUnits: Calendar years; laboratory energy-density figures are not used\nMarket definition: Lithium-ion development; the early lithium-metal prototype differs from the later carbon-anode ion-transfer cell\nEvidence classification: Official scientific historical synthesis; not the original experiments or sales records\nRead date: 2026-10-06\nMethod and calculation: Direct reading of official Academy PDF text. Dates preserved as stated; no conversion of 1979/1980 into one exact date.\nLimits and uncertainty: Avoid exclusive inventor or universal first claims. The source does not establish current battery economics, chemistry shares or today\u0026rsquo;s safety performance.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: Scientific Background on the Nobel Prize in Chemistry 2019: Lithium-Ion Batteries\n[H04] Public–Private Collaboration Paves the Way for Commercial Wind Power Growth Supports: DOE and EPRI formed the Wind Turbine Verification Program in 1993, using utility-hosted projects to evaluate precommercial turbines, operating performance and maintenance, with consistent data collection.\nInstitution or author: U.S. Department of Energy\nPublication date: 2023-09-26\nEvent date: 1993 program formation; utility demonstration projects during 1995–2001\nStatistical period: Demonstration-program history\nPage or section: HTML lines 7, 10–11, 22–35, 38–46 and 57–61; formation, operating trials and data-sharing passages read\nGeography: United States utility demonstration sites\nUnits: Calendar years; project capacities and funding percentages are not adopted\nMarket definition: Precommercial wind turbines tested in utility operating environments; not all U.S. wind deployment\nEvidence classification: Official retrospective by the program sponsor\nRead date: 2026-10-06\nMethod and calculation: Direct reading of dated DOE article. No calculation or counterfactual estimate.\nLimits and uncertainty: Supports program mechanisms, not proof that this program alone caused wind growth or that every project succeeded.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: Public–Private Collaboration Paves the Way for Commercial Wind Power Growth\n[H05] Phospho-olivines as positive-electrode materials for rechargeable lithium batteries Supports: Padhi, Nanjundaswamy and Goodenough\u0026rsquo;s 1997 paper reported reversible lithium extraction and insertion in LiFePO4/FePO4 and identified rate-related capacity limits.\nInstitution or author: Journal of The Electrochemical Society; author abstract reproduced in U.S. EPA HERO\nPublication date: 1997; exact original publication day not established; EPA page updated 2026-01-23\nEvent date: 1997 paper publication\nStatistical period: Laboratory research, not a market period\nPage or section: EPA HERO record 7748087, bibliography and complete reproduced abstract; journal 144(4), pp. 1188–1194 identifies the article, not full-text reading\nGeography: Laboratory materials research; no geographic market coverage\nUnits: Qualitative reversible electrochemistry; laboratory voltage, current and capacity values not used\nMarket definition: LFP cathode candidate; not a finished commercial cell, pack or storage project\nEvidence classification: Original research abstract reproduced by an official government database\nRead date: 2026-10-06\nMethod and calculation: Read the full publicly reproduced abstract and metadata; original DOI 10.1149/1.1837571 could not be opened. No calculation.\nLimits and uncertainty: Full paper not read. Does not establish commercial mass production, modern performance or a universal first-LFP claim.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: Phospho-olivines as positive-electrode materials for rechargeable lithium batteries\n[H06] Electronically conductive phospho-olivines as lithium storage electrodes Supports: A 2002 Nature Materials paper explicitly addressed the low electronic conductivity limiting LiFePO4 electrode performance and reported an engineered-material approach.\nInstitution or author: Nature Materials; authors Sung-Yoon Chung, Jason T. Bloking and Yet-Ming Chiang\nPublication date: 2002-09-22 online; issue date 2002-10-01\nEvent date: 2002 research publication\nStatistical period: Laboratory materials research\nPage or section: Publisher page, publication metadata and public abstract; Nature Materials 1, pp. 123–128 is the article locator, not pages read\nGeography: Laboratory research; no geographic market sample\nUnits: Qualitative conductivity constraint; numerical improvement claims not used\nMarket definition: Electrode-material conductivity research; distinct from commercial adoption or storage-system economics\nEvidence classification: Original journal research abstract on the publisher website\nRead date: 2026-10-06\nMethod and calculation: Direct reading of the public abstract and dates; DOI 10.1038/nmat732. Subscription full text not read.\nLimits and uncertainty: Does not prove the reported mechanism applies to every modern LFP process, resolve later scientific disputes, or date mass commercialization. Authors disclosed a financial interest.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: Electronically conductive phospho-olivines as lithium storage electrodes\n[H07] Renewable Energy Law of the People’s Republic of China, as amended in 2009 Supports: The law was adopted on 28 February 2005 and states commencement on 1 January 2006. The read 2009-amended text links targets and guaranteed purchase to grid development, storage and cost compensation.\nInstitution or author: National People’s Congress Standing Committee; text published by China’s National Energy Administration\nPublication date: 2012-01-04 official republication\nEvent date: 2005-02-28 adoption; 2006-01-01 stated commencement; 2009-12-26 amendment\nStatistical period: Historical statutory framework; 2009-amended version read\nPage or section: Dated NEA HTML, preamble and Articles 2–4, 7, 11–14, 19–24 and 33; full displayed text read\nGeography: PRC territory and other sea areas under its jurisdiction, as Article 3 states\nUnits: Legal dates and obligations; no tariff amount or market-share unit\nMarket definition: Statutory renewables include wind, solar, hydro, biomass, geothermal and ocean energy, subject to Article 2 qualifications; not battery manufacturing generally\nEvidence classification: Official legal text republication, expressly amended\nRead date: 2026-10-06\nMethod and calculation: Direct reading of full official HTML. Adoption and commencement distinguished from the republication date and amended provisions.\nLimits and uncertainty: Do not backdate all read provisions to 2005. Historical framework only: no claim of universal delivery, subsidy receipt or current-law completeness.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: Renewable Energy Law of the People’s Republic of China, as amended in 2009\n[H08] Directive 2009/28/EC on the promotion of the use of energy from renewable sources Supports: The 2009 directive set binding national targets consistent with at least 20% renewable energy in Community gross final energy consumption by 2020; Article 16 also addressed grids and storage facilities.\nInstitution or author: European Parliament and Council; Official Journal via EUR-Lex\nPublication date: 2009-06-05, Official Journal L 140\nEvent date: 2009-04-23 directive date; 2020 target year\nStatistical period: Historical 2020 policy target, not measured 2020 outcome\nPage or section: Official PDF p. 1 (OJ L140/16), p. 13 (L140/28, Article 3), pp. 20–21 (L140/35–36, Article 16); relevant text read\nGeography: European Community Member States within the directive’s historical framework; text has EEA relevance\nUnits: Percent of gross final energy consumption; not electricity-only share or installed capacity\nMarket definition: All covered renewable final energy uses; national targets and grid provisions have distinct scopes\nEvidence classification: Original official legislative text\nRead date: 2026-10-06\nMethod and calculation: Direct reading of original Official Journal PDF text; target preserved as a target. No realized-share calculation.\nLimits and uncertainty: Historical version, not a statement of 2026 legal requirements. Does not prove targets achieved or that grid access eliminated curtailment; Article 16 includes security conditions.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: Directive 2009/28/EC on the promotion of the use of energy from renewable sources\n[H09] DOE Pursues SunShot Initiative to Achieve Cost Competitive Solar Energy by 2020 Supports: DOE’s 4 February 2011 announcement targeted roughly a 75% reduction in total PV system costs and included manufacturing, installation and permitting work, alongside cell technology.\nInstitution or author: U.S. Department of Energy\nPublication date: 2011-02-04\nEvent date: 2011-02-04 initiative announcement; end-of-decade cost goal\nStatistical period: Program goal announced in 2011, not an observed cost result\nPage or section: HTML date and paragraphs at lines 10–15; total-cost objective and permitting passage read\nGeography: United States utility-scale PV cost-competitiveness objective\nUnits: Approximately 75% target reduction; baseline and price-year calculation not independently reconstructed\nMarket definition: Total photovoltaic system costs, extending beyond the solar cell or module price\nEvidence classification: Contemporaneous official program announcement\nRead date: 2026-10-06\nMethod and calculation: Direct reading of dated announcement. Target not treated as achievement; dollar and electricity-cost equivalents not adopted.\nLimits and uncertainty: Cannot attribute later global cost reductions solely to SunShot or prove every region became competitive. It is an announced U.S. objective.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: DOE Pursues SunShot Initiative to Achieve Cost Competitive Solar Energy by 2020\n[H10] Order No. 841: Electric Storage Participation in Markets Operated by Regional Transmission Organizations and Independent System Operators Supports: FERC issued Order 841 on 15 February 2018, requiring RTO/ISO tariff participation models that recognize storage characteristics and permit technically capable resources to offer capacity, energy and ancillary services.\nInstitution or author: U.S. Federal Energy Regulatory Commission\nPublication date: 2018-02-15 issuance and official public announcement; official PDF compilation contains 2018-02-28 errata\nEvent date: 2018-02-15 Order 841 issuance; not its effective date\nStatistical period: Historical wholesale-market rule\nPage or section: Official PDF physical pp. 28–31, printed pp. 1–4, introduction paragraphs 1–4 and footnotes 1–2; issuance date read; contemporaneous FERC release corroborates date\nGeography: U.S. RTO/ISO wholesale markets under FERC jurisdiction; not every U.S. retail market or a global rule\nUnits: Legal dates and service categories; no revenue, capacity or market-share estimate\nMarket definition: Storage can receive grid electricity, store it and later inject it; covers storage technologies beyond lithium-ion batteries\nEvidence classification: Original official regulatory order and contemporaneous official announcement\nRead date: 2026-10-06\nMethod and calculation: Direct reading of the order’s relevant introduction and dated announcement; not a full 258-page review. No calculation.\nLimits and uncertainty: Issuance is not implementation or commercial success. No inference of guaranteed profit, uniform tariffs or current-law completeness.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: Order No. 841: Electric Storage Participation in Markets Operated by Regional Transmission Organizations and Independent System Operators\n[C11] Albemarle 2025 Annual Report on Form 10-K—business evidence Supports: Albemarle extracts lithium from brine and hard-rock resources and converts it into compounds including carbonate and hydroxide used in batteries. Its Energy Storage business is exposed to lithium pricing; many customer contracts reference variable market indices. This supports an upstream resource-and-conversion role.\nInstitution or author: Albemarle Corporation; United States Securities and Exchange Commission filing\nPublication date: 11 February 2026, SEC filing date\nEvent date: FY2025 annual filing; individual contract-signing dates not provided\nStatistical period: FY2025 report for the year ended 31 December 2025; business and contract-model descriptions as reported\nPage or section: Item 1, Energy Storage segment; Item 2, Properties and Mining Operations; Item 7, Business Outlook\nGeography: Global Albemarle upstream lithium operations and conversion business\nUnits: Business and contract-model descriptions; quantitative financial unit not applicable\nMarket definition: Lithium resource extraction and chemical conversion for multiple end uses, including batteries\nEvidence classification: Primary regulatory annual-report business disclosure\nRead date: 2026-10-06\nMethod and calculation: Read business, property and outlook sections; no financial calculation; financial figures are separately supported by C09\nLimits and uncertainty: General business descriptions do not prove contracts with CATL, LG Energy Solution or Tesla. Variable-price references do not establish the price or duration of any named agreement. This record uses business evidence only, not financial statements from a later amendment.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: Albemarle 2025 Annual Report on Form 10-K—business evidence\n[G10] Solar PV Supply Chain Cost Tool: Methodology, results and analysis Supports: The crystalline-silicon chain covers polysilicon, wafers, cells and module assembly. Manufacturing costs include materials, equipment, facilities, electricity, labour and overheads. The tool separates country conditions and assumes profitability; it does not report realised corporate margins.\nInstitution or author: International Renewable Energy Agency (IRENA)\nPublication date: The European Commission\u0026rsquo;s BUILD UP catalogue records a February 2026 release for the resource; the associated same-edition report was publicly linked by its 8 May 2026 catalogue entry. The report\u0026rsquo;s own exact first release day remains unverified.\nEvent date: Not applicable: cost tool and analytical model.\nStatistical period: 2025 model base; 2030 projections are scenarios. Country input data have source-specific dates.\nPage or section: Overall approach, p7; production-stage and cost-component chapters; operating-profit assumption, p12; projection limitations, p23. PDF pages equal printed page numbers.\nGeography: Six model markets: Australia, China, Germany, India, United States and Viet Nam.\nUnits: Modelled manufacturing USD/Wp; qualitative value-chain and cost categories.\nMarket definition: Crystalline-silicon manufacturing cost model from polysilicon through module assembly; not all solar technologies or installed project cost.\nEvidence classification: Original IRENA methodological model with sourced inputs and explicit assumptions.\nRead date: 2026-10-06\nMethod and calculation: Use stage/cost taxonomy; do not treat assumed profit or 2030 model values as observed results. Cutoff verification: European Commission, BUILD UP catalogue, https://build-up.ec.europa.eu/en/resources-and-tools/tools/solar-pv-supply-chain-cost-tool-analysing-photovoltaic-manufacturing.\nLimits and uncertainty: Six-country modelling is not a global market census. Future calculations hold material prices fixed and omit unexpected volatility, supply constraints and policy shifts. Crystalline-silicon stages cannot automatically describe thin-film manufacturers such as First Solar.\nAttachments and redistribution: Link to the publisher; no original research file is included in the review package.\nOriginal source: Solar PV Supply Chain Cost Tool: Methodology, results and analysis\n[G09] Sodium-ion batteries: A technology brief Supports: Sodium-ion can diversify raw-material supply, but scaling depends on energy density, hard-carbon processing and competitive cost. Lower weight sensitivity makes stationary storage a potential application. Resource abundance alone does not establish mature production or a universally cheaper replacement for lithium-ion.\nInstitution or author: International Renewable Energy Agency (IRENA); collaboration with China Electric Power Research Institute (CEPRI)\nPublication date: 2025, verified by the Spanish Ministry of Industry and Tourism\u0026rsquo;s official library bibliography. Exact month and day not independently verified.\nEvent date: Not applicable: technology review.\nStatistical period: Technology landscape through 2025; some underlying cost and capacity studies are earlier and explicitly identified.\nPage or section: Executive summary, p5; 3.1 Construction and materials, pp12–14; 3.2 Supply chain, pp14–15; 3.3 Pros and cons, pp16–17; 3.4 Applications, pp18–19; Status and outlook, pp20–21. PDF pages equal printed page numbers.\nGeography: Global technology review; country examples are not a current global production census.\nUnits: Qualitative technical claims; no current price or market-share calculation used.\nMarket definition: Sodium-ion batteries transfer sodium ions; cathode choices and hard-carbon anodes have distinct material and manufacturing requirements.\nEvidence classification: Original intergovernmental technical synthesis incorporating third-party studies and company announcements.\nRead date: 2026-10-06\nMethod and calculation: Use qualitative mechanisms; exclude mismatched-year cost comparisons and announced-capacity forecasts from actual-market totals. Cutoff verification: Spanish Ministry of Industry and Tourism, Library catalogue, https://www.mintur.gob.es/es-es/servicios/Documentacion/Biblioteca/Boletines/Boletin_novedades_ene_mar_2026.pdf.\nLimits and uncertainty: Some cost comparisons use sodium-ion 2022 and lithium-ion April 2024 data, so they cannot prove a 2026 price advantage. Announced production is not achieved output. Some sodium cathodes contain nickel or cobalt; universal mineral-free claims are unsupported.\nAttachments and redistribution: Link to the publisher; no original research file is included in the review package.\nOriginal source: Sodium-ion batteries: A technology brief\n[C07] First Solar 2025 Annual Report on Form 10-K Supports: First Solar makes CdTe thin-film modules and generally prices modules per watt. In 2025, net sales were $5.219 billion and operating income $1.597 billion. Year-end future sales contracts covered 50.1 GW, valued at $15.0 billion, with revenue expected through 2030.\nInstitution or author: First Solar, Inc.; United States Securities and Exchange Commission filing\nPublication date: 24 February 2026, SEC filing date\nEvent date: FY2025 filing; future-contract balance measured at 31 December 2025\nStatistical period: Fiscal year from 1 January to 31 December 2025; future revenue horizon through 2030\nPage or section: Item 1, Advanced Module Technology; Item 7, Net Sales; Consolidated Statements of Operations; Note 14, Revenue Contracts with Customers\nGeography: Global consolidated First Solar; contract backlog across its markets\nUnits: USD thousand in financial statements, USD billion after conversion; GW of contracted module power\nMarket definition: Company module sales and operating income; future contracted module sales, excluding specified unsecured India orders\nEvidence classification: Primary regulatory annual report with audited US GAAP financial statements and contractual disclosures\nRead date: 2026-10-06\nMethod and calculation: Net sales $5,219,376 thousand and operating income $1,596,864 thousand divided by 1,000,000 and rounded to three decimals in USD billion; contracts retained as reported\nLimits and uncertainty: Contracted power and contract value are not delivered modules or current-year revenue. Revenue depends on control transfer. Company profitability is not an industry margin, and operating income differs from the net-profit and adjusted-EBITDA metrics used for other companies.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: First Solar 2025 Annual Report on Form 10-K\n[C08] Vestas Annual Report 2025 Supports: Vestas sells wind turbines and service contracts. In 2025 it reported €18.822 billion revenue and a 5.7% EBIT margin before special items. Year-end turbine backlog was €33.2 billion and service backlog €38.7 billion, totaling €71.9 billion across different fulfillment horizons.\nInstitution or author: Vestas Wind Systems A/S\nPublication date: 5 February 2026\nEvent date: Annual report released on 5 February 2026; backlog measured at 31 December 2025\nStatistical period: Fiscal year from 1 January to 31 December 2025; service backlog represents future contractual revenue\nPage or section: Group financial performance, p. 25; Power Solutions order backlog, p. 27; Service, p. 34; Note 1.2, Revenue and contract types, pp. 142–143\nGeography: Global consolidated Vestas, Power Solutions and Service businesses\nUnits: EUR million in financial table, EUR billion after conversion; percent EBIT margin\nMarket definition: Company revenue and EBIT before special items; separate future turbine and service contractual backlogs\nEvidence classification: Primary company annual report with audited financial statements and operating disclosures\nRead date: 2026-10-06\nMethod and calculation: €18,822 million divided by 1,000 gives €18.822 billion; €33.2 billion plus €38.7 billion equals the reported €71.9 billion; use the reported 5.7% margin\nLimits and uncertainty: Backlog does not equal annual deliveries or cash. Service and turbine horizons differ. EBIT before special items excludes identified items and cannot be ranked directly against other firms\u0026rsquo; net income, gross margin or adjusted EBITDA.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: Vestas Annual Report 2025\n[C05] CATL Annual Report 2025 Supports: CATL sells power and energy-storage batteries and solutions. In 2025 it reported RMB423.7 billion revenue, RMB72.2 billion profit attributable to listed-company shareholders and 661 GWh lithium-ion battery sales. Battery sales combine applications and differ from registered EV-battery usage.\nInstitution or author: Contemporary Amperex Technology Co., Limited (CATL)\nPublication date: 10 March 2026\nEvent date: FY2025 results released on 10 March 2026\nStatistical period: Fiscal year from 1 January to 31 December 2025\nPage or section: Section II, Financial Highlights, printed p. 7 (PDF page 8); Section IV, Principal Business, Products and Business Model, printed pp. 12–14; business review, printed pp. 25–27; Note 24, Revenue, printed p. 175\nGeography: Global consolidated CATL group; not the entire battery industry\nUnits: RMB thousand in financial tables; RMB billion when rounded; GWh of company battery sales\nMarket definition: Consolidated revenue, attributable profit and company lithium-ion battery sales across power and storage applications\nEvidence classification: Primary company annual report with audited financial statements and operating disclosures\nRead date: 2026-10-06\nMethod and calculation: Revenue RMB423,701,834 thousand and attributable profit RMB72,201,282 thousand divided by 1,000,000 and rounded to one decimal in RMB billion; 661 GWh reported directly\nLimits and uncertainty: Attributable profit is not total group net profit. Revenue cannot be divided by an unmatched battery-market denominator. Product descriptions do not prove named customer contracts, and annual sales are not installed capacity or outstanding orders.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: CATL Annual Report 2025\n[C06] LG Energy Solution Releases 2025 Financial Results Supports: LG Energy Solution reported 2025 revenue of KRW23.7 trillion and operating profit of KRW1.3 trillion, including North American production incentives. It disclosed an ESS order backlog of 140 GWh and a 46-series backlog exceeding 300 GWh; neither is delivered volume.\nInstitution or author: LG Energy Solution, official Battery Inside publication\nPublication date: 29 January 2026\nEvent date: FY2025 financial results announcement on 29 January 2026\nStatistical period: Fiscal year from 1 January to 31 December 2025; 46-series backlog explicitly at year-end; ESS backlog in the 2025 business review\nPage or section: Opening full-year financial results and 2025 business review, including customer-base and portfolio-diversification paragraphs\nGeography: Global consolidated LG Energy Solution; North American incentive qualification affects profit\nUnits: KRW trillion; GWh of disclosed order backlog\nMarket definition: Company annual revenue and operating profit; separate ESS and 46-series backlog disclosures\nEvidence classification: Primary company earnings announcement with rounded results and forward-looking statements\nRead date: 2026-10-06\nMethod and calculation: Use reported rounded figures; do not add the two backlog categories or substitute an independently calculated margin\nLimits and uncertainty: Profit includes incentives and is not subsidy-free profit. Backlog is not revenue or delivery; category overlap and fulfillment schedules are not established here. The 2026 order and capacity targets are forecasts, excluded from historical results.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: LG Energy Solution Releases 2025 Financial Results\n[C10] Tesla 2025 Annual Report on Form 10-K Supports: Tesla sells Powerwall and Megapack storage products. In 2025, storage deployments were 46.7 GWh; the broader energy generation and storage segment reported $12.771 billion revenue and a 29.8% gross margin. Manufacturing credits reduced that segment\u0026rsquo;s cost of revenue by $1.12 billion.\nInstitution or author: Tesla, Inc.; United States Securities and Exchange Commission filing\nPublication date: 29 January 2026, SEC filing date\nEvent date: FY2025 annual filing; separate single event date not applicable\nStatistical period: Fiscal year from 1 January to 31 December 2025\nPage or section: Item 1, Energy Storage Products; Item 7, Energy Generation and Storage Segment and gross-margin discussion; Note 2, Energy Generation and Storage Revenue\nGeography: Global consolidated Tesla energy segment\nUnits: GWh of storage deployments; USD million/billion of revenue and cost credits; percent gross margin\nMarket definition: Storage deployment includes storage products; segment financials include both generation and storage, not Megapack alone\nEvidence classification: Primary regulatory annual report with audited US GAAP financial statements and operational disclosures\nRead date: 2026-10-06\nMethod and calculation: $12,771 million divided by 1,000 gives $12.771 billion; other deployment, margin and credit values are reported directly\nLimits and uncertainty: Gross margin is not operating or net margin. Incentives affect profitability. Deployments are not orders or factory capacity. This source does not establish Tesla as a supplier to the Vistra facility involved in C01–C04.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: Tesla 2025 Annual Report on Form 10-K\n[C12] JinkoSolar–Masdar 2 GW Tiger Neo module purchase agreement Supports: JinkoSolar announced a signed agreement to supply Masdar with 2 GW of Tiger Neo solar modules for an Abu Dhabi round-the-clock project. This identifies an actual manufacturer-to-developer commercial agreement, while the announcement does not establish completed module delivery or an operating project.\nInstitution or author: JinkoSolar, official corporate news\nPublication date: 12 May 2026, verified on the issuer\u0026rsquo;s dated news index\nEvent date: Agreement signed by the announcement date; exact signing day not disclosed\nStatistical period: One agreement announced in May 2026; annual financial period not applicable\nPage or section: News article\u0026rsquo;s opening agreement paragraph; official news index dated 12 May 2026\nGeography: Abu Dhabi, United Arab Emirates; JinkoSolar–Masdar supply agreement\nUnits: GW of contracted solar-module electrical power, not GWh of battery energy\nMarket definition: Announced 2 GW module purchase agreement for one project; not the full project\u0026rsquo;s capacity or installed global capacity\nEvidence classification: Primary corporate announcement of a named commercial supply agreement\nRead date: 2026-10-06\nMethod and calculation: Read original article and dated company index; retain announced contracted power without converting it to energy or revenue\nLimits and uncertainty: No disclosed price, payment schedule or verified deliveries. Promotional superlatives are not independently established. The project description is a future plan, not evidence of current round-the-clock output; this agreement does not prove another company\u0026rsquo;s storage-supply contract.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: JinkoSolar–Masdar 2 GW Tiger Neo module purchase agreement\n[G03] Global EV Outlook 2026 — Electric vehicle batteries Supports: 2025 EV deployment: 1.2 TWh, almost 30% growth; China 60%, EU nearly 15%, US 10% and stagnant. Cell nameplate: over 4 TWh; China over 80%, EU/US each 6–7%. Almost/all-solid-state prototypes; recycling mainly used production scrap.\nInstitution or author: International Energy Agency (IEA)\nPublication date: 2026-05-20\nEvent date: Not applicable: annual report.\nStatistical period: 2025 actual estimates and year-end manufacturing capacity; technology status at report publication. Future scenarios are separate.\nPage or section: Battery demand; Battery chemistry; Battery manufacturing; Battery technology developments; Battery recycling; reference notes 1, 3, 8 and 9. Publication date: report overview.\nGeography: Global; China, European Union and United States. Deployment geography differs from factory geography.\nUnits: TWh of deployed battery energy and annual nameplate capacity; percent.\nMarket definition: Deployment = volume-weighted average battery size × vehicle sales by mode and region; manufacturing capacity is nameplate.\nEvidence classification: Original IEA analysis using registration, industry and third-party datasets; not a manufacturer census independently audited here.\nRead date: 2026-10-06\nMethod and calculation: Deployment follows vehicle sales and battery size; capacity is nameplate, not output.\nLimits and uncertainty: Different applications prevent a utilisation ratio. Prototype status is not proven commercial performance; announcements are not delivery.\nAttachments and redistribution: Link to the publisher; no original research file is included in the review package.\nOriginal source: Global EV Outlook 2026 — Electric vehicle batteries\n[G01] Renewable capacity statistics 2026 Supports: Year-end 2025 renewable capacity: 5,149.280 GW; 2024: 4,457.340 GW. Net addition: 691.940 GW, or 15.5%. Solar stock: 2,391.584 GW; wind: 1,291.368 GW. China, the United States and the EU represented 79.5% of additions; Africa 1.6%.\nInstitution or author: International Renewable Energy Agency (IRENA)\nPublication date: 2026; exact release day not verified. Official publication-directory path: March 2026.\nEvent date: Not applicable: statistical release. Capacity stock dates: 2024-12-31 and 2025-12-31.\nStatistical period: Year-end 2024 and 2025; calendar-year 2025 net additions. Same 2026-edition tables.\nPage or section: Foreword, PDF p3; notes, printed pIII / PDF p7; total renewable World row, printed p2 / PDF p14; wind World row, printed p14 / PDF p26; solar World row, printed p21 / PDF p33; PV World row, printed p25 / PDF p37.\nGeography: Global; selected regional comparisons explicitly identified.\nUnits: MW in original tables; GW after division by 1,000; percent.\nMarket definition: Maximum net renewable generating capacity, generally installed and connected at year-end; pure pumped storage is excluded.\nEvidence classification: Original intergovernmental statistical compilation.\nRead date: 2026-10-06\nMethod and calculation: (5,149,280 − 4,457,340)/1,000 = 691.940 GW; growth = difference/4,457,340.\nLimits and uncertainty: Capacity is not electricity output or revenue. Solar additions differ between the foreword (510 GW), total-solar table (511.188 GW) and PV table (510.349 GW). The exact discrepancy is unresolved; use stocks and the global total instead.\nAttachments and redistribution: Link to the publisher; no original research file is included in the review package.\nOriginal source: Renewable capacity statistics 2026\n[G07] Electricity Mid-Year Update 2026 — Executive summary Supports: Global electricity demand reached 28,600 TWh in 2025, up 3%; renewables supplied 33% of generation. In the first half of 2026, South Australia and California recorded negative wholesale prices in roughly 20% of hours. Flexibility faces technical, regulatory and contractual barriers.\nInstitution or author: International Energy Agency (IEA)\nPublication date: 2026-07-23\nEvent date: Not applicable: mid-year update.\nStatistical period: 2025 annual estimates; observed first-half 2026 wholesale pricing. 2026–2027 full-year figures are forecasts.\nPage or section: Executive summary: global demand, generation mix, wholesale prices and flexibility; report overview publication date.\nGeography: Global for demand/generation; South Australia and California for negative-price frequency.\nUnits: TWh/year; year-on-year percent; share of wholesale-market hours.\nMarket definition: Electricity generation is energy output, distinct from installed capacity. Negative-price frequency counts wholesale-market hours in specified regions.\nEvidence classification: Original IEA update compiling electricity-system and market data.\nRead date: 2026-10-06\nMethod and calculation: Use annual observations and half-year pricing separately; do not extrapolate negative-price hours into an annual global rate.\nLimits and uncertainty: Wholesale prices are not retail tariffs or a specific generator’s realised price. Negative prices alone cannot prove storage profitability. Forecast renewable overtaking of coal in 2026 is not a completed annual outcome. Hourly capture-price datasets were not provided here.\nAttachments and redistribution: Link to the publisher; no original research file is included in the review package.\nOriginal source: Electricity Mid-Year Update 2026 — Executive summary\n[M01] January to August 2026 global EV battery usage Supports: 844.2 GWh and 19.7% growth; top five shares 39.4%, 15.1%, 8.1%, 5.3% and 4.9%.\nInstitution or author: SNE Research\nPublication date: 2026-10-02\nEvent date: Not applicable\nStatistical period: 2026-01 to 2026-08\nPage or section: Opening, supplier paragraphs, Top 10 table image and footnotes 1 and 2\nGeography: 80 countries in SNE coverage\nUnits: GWh and percent\nMarket definition: Energy in batteries installed in registered EV, PHEV and HEV vehicles; excludes stationary storage.\nEvidence classification: Third-party estimate published by its originator\nRead date: 2026-10-06\nMethod and calculation: Other suppliers = 100 − 39.4 − 15.1 − 8.1 − 5.3 − 4.9 = 27.2%.\nLimits and uncertainty: Rounded estimates. CATL plus BYD is separately stated as 54.6%, versus 54.5% from components; cause unverified. No normalization.\nAttachments and redistribution: Publisher link only; no original redistributed.\nOriginal source: January to August 2026 global EV battery usage\n[G06] Lithium-Ion Battery Pack Prices Fall to $108 Per Kilowatt-Hour Despite Rising Metal Prices: BloombergNEF Supports: BNEF’s 2025 survey estimated average lithium-ion pack prices at USD 108/kWh, down 8%; stationary-storage packs at 70, down 45%; BEV packs at 99. China averaged 84; North America and Europe were 44% and 56% higher respectively.\nInstitution or author: BloombergNEF (BNEF)\nPublication date: 2025-12-09\nEvent date: 2025-12-09 public survey announcement.\nStatistical period: 2025 survey; year-on-year comparison with the survey’s 2024 baseline.\nPage or section: Public release body: global average; stationary-storage and BEV packs; regional prices; survey explanation.\nGeography: Global, China, North America and Europe; geography as defined in the survey.\nUnits: USD/kWh at battery-pack level; percent.\nMarket definition: Surveyed average pack prices across applications, with separate application and regional averages; not total installed-system cost.\nEvidence classification: Third-party survey estimate; original public announcement by the survey publisher.\nRead date: 2026-10-06\nMethod and calculation: Report the published 8% decline; do not recompute it from an older publication vintage.\nLimits and uncertainty: The paid full survey and raw weighting sample were not read. Regional averages reflect application and chemistry mix; they do not isolate causation or show a company’s margin. Pack prices exclude system integration and installation costs.\nAttachments and redistribution: Link to the publisher; no original research file is included in the review package.\nOriginal source: Lithium-Ion Battery Pack Prices Fall to $108 Per Kilowatt-Hour Despite Rising Metal Prices: BloombergNEF\n[G02] Renewable power generation costs in 2025 Supports: 2025 global weighted-average LCOE: solar PV USD 44/MWh, onshore wind 33, offshore wind 78. Since 2010, these fell 89%, 71% and 63%. Four-hour utility-scale BESS installed cost averaged USD 140/kWh. Four-hour turnkey cost averaged 111/kWh, excluding EPC, grid connection and development.\nInstitution or author: International Renewable Energy Agency (IRENA)\nPublication date: July 2026; verified to month precision through PNNL\u0026rsquo;s institutional catalogue. Exact original release day not independently verified.\nEvent date: Not applicable: annual cost study.\nStatistical period: Projects commissioned in 2025; historical comparison with 2010; 2025 storage cost estimates.\nPage or section: Executive summary, printed/PDF pp11–16; Table 1.1, pp14–15; regional comparisons, p16; exclusion-of-China comparison, p23; battery storage, pp139–144; four-hour turnkey definition and cost, p143.; solar scope: p.70, section 3.6 and footnote 17\nGeography: Global cost averages; explicitly separate selected country samples.\nUnits: 2025 USD/MWh for LCOE; 2025 USD/kWh for BESS equipment and installed costs; percent.\nMarket definition: LCOE is a lifetime generation cost metric. Installed BESS, turnkey equipment and battery packs have different cost boundaries. The solar PV benchmark covers standalone utility-scale projects; rooftop PV is outside it.\nEvidence classification: Original IRENA cost study; storage estimates incorporate third-party BloombergNEF data.\nRead date: 2026-10-06\nMethod and calculation: Use reported weighted averages; no pack-to-system subtraction or direct inference about project selling prices. Cutoff verification: Pacific Northwest National Laboratory, Tethys Engineering catalogue, https://tethys-engineering.pnnl.gov/publications/renewable-power-generation-costs-2025.\nLimits and uncertainty: LCOE does not include the same services as firm electricity. Cost averages do not establish company margins. USD 240/kWh in the selected-market component chart is not the global USD 140/kWh benchmark. Raw survey observations were not separately audited.\nAttachments and redistribution: Link to the publisher; no original research file is included in the review package.\nOriginal source: Renewable power generation costs in 2025\n[G04] Energy Technology Perspectives 2026 — Supply chain risks and industrial competitiveness Supports: In 2024, China held about 85% of solar PV and 80% of lithium-ion battery supply-chain capacity on a value-weighted basis excluding mining. Each examined technology had an upstream bottleneck outside China. Manufacturing efficiency explained over 40% of the modelled Europe–China battery cost gap.\nInstitution or author: International Energy Agency (IEA)\nPublication date: 2026-03-26\nEvent date: Not applicable: analytical report.\nStatistical period: 2024 supply-chain baseline and cost model; conditional future scenarios are separate.\nPage or section: Supply chain risks and industrial competitiveness: concentration chart and notes; N−1 supply-chain analysis and notes; Manufacturing efficiency and battery production costs. Publication date: report overview.\nGeography: Global manufacturing stages; China, Europe and the rest of the world.\nUnits: Percent of value-weighted capacity; percent of modelled cost differences.\nMarket definition: Concentration is value-weighted across manufacturing stages, excluding extraction; N−1 removes the largest supplying country under stated assumptions.\nEvidence classification: Original IEA synthesis and scenario/cost modelling using third-party datasets.\nRead date: 2026-10-06\nMethod and calculation: N−1 assumes 85% nameplate utilisation for manufacturing; mineral stages use actual production. Battery comparison models NMC811/graphite.\nLimits and uncertainty: These are 2024 capacity/model results, not 2026 shipments, headquarters shares, contracts or actual disruption losses. Non-Chinese downstream capacity alone does not demonstrate an independent supply chain. A modelled cost gap cannot establish company profitability.\nAttachments and redistribution: Link to the publisher; no original research file is included in the review package.\nOriginal source: Energy Technology Perspectives 2026 — Supply chain risks and industrial competitiveness\n[G05] Global Critical Minerals Outlook 2026 — Executive summary Supports: Critical-mineral investment fell 9% in 2025; battery-metal capital expenditure fell over 20%. The largest refining country’s average share, excluding rare earths, rose to 72% from 70% in 2023. New refining projects outside incumbents face higher costs and equipment, technology and skills constraints.\nInstitution or author: International Energy Agency (IEA)\nPublication date: 2026-07-16\nEvent date: Not applicable: analytical report.\nStatistical period: 2025 investment and refining statistics; comparisons with 2023; policy/pricing developments through report publication.\nPage or section: Executive summary: Investment; Refining concentration; Diversification challenges; Supply-chain ecosystems. Publication date: report overview.\nGeography: Global critical-mineral markets; leading refining country varies by mineral.\nUnits: Year-on-year percent investment changes; average leading-country refining share.\nMarket definition: Refining concentration is the leading single country’s average share across the specified minerals, excluding rare earths.\nEvidence classification: Original IEA market assessment using industry datasets and project analysis.\nRead date: 2026-10-06\nMethod and calculation: Use the specified single-country refining metric; do not compare it with older top-three-country measures.\nLimits and uncertainty: Investment decline is not output decline. Announced public-finance commitments are not disbursements. Higher project costs are estimates with site-specific variation. Refining averages do not establish one country’s ownership of all minerals or guarantee future shortages.\nAttachments and redistribution: Link to the publisher; no original research file is included in the review package.\nOriginal source: Global Critical Minerals Outlook 2026 — Executive summary\n[C09] Albemarle Fourth Quarter and Full Year 2025 Results Supports: Albemarle\u0026rsquo;s lithium-focused Energy Storage segment reported 2025 net sales of $2.710 billion and adjusted EBITDA of $697 million, down 8%. The company attributed the EBITDA decline to lower lithium pricing partly offset by volume and cost improvements. This segment is not a battery-system manufacturer.\nInstitution or author: Albemarle Corporation\nPublication date: 11 February 2026\nEvent date: FY2025 earnings announcement on 11 February 2026\nStatistical period: Fiscal year from 1 January to 31 December 2025; the separate fourth-quarter figures are excluded\nPage or section: Energy Storage Results, full-year EBITDA paragraph; Consolidated Summary of Segment Results, full-year 2025 net-sales column; Non-GAAP Reconciliations\nGeography: Global Albemarle Energy Storage segment\nUnits: USD thousand for segment net sales; USD million for reported adjusted EBITDA\nMarket definition: Albemarle\u0026rsquo;s named lithium business segment, rather than battery storage equipment or consolidated group financials\nEvidence classification: Primary company earnings announcement; unaudited tables and non-GAAP adjusted EBITDA\nRead date: 2026-10-06\nMethod and calculation: Net sales $2,710,035 thousand divided by 1,000,000 and rounded to $2.710 billion; $697 million and 8% decline retained as reported\nLimits and uncertainty: Adjusted EBITDA is not net profit. Management\u0026rsquo;s price explanation is a corporate attribution, not an independent estimate of market causality. The release\u0026rsquo;s 2026 scenarios and long-term-contract assumptions are forward-looking, excluded from 2025 realized results.\nAttachments and redistribution: Publisher link only; no original file redistributed.\nOriginal source: Albemarle Fourth Quarter and Full Year 2025 Results\n[G08] 24/7 renewables: The economics of firm solar and wind Supports: A hypothetical 100 MW Las Vegas solar case uses 2025 costs. At a 95% energy-coverage target, solar plus storage and overbuild costs USD 113.2/MWh versus 43.5 standalone, with 591.8 MWh of storage and 61.9 MW of additional solar.\nInstitution or author: International Renewable Energy Agency (IRENA)\nPublication date: Publicly catalogued by 20 May 2026; KETEP registration date verified. The original publisher\u0026rsquo;s first release date remains unverified.\nEvent date: Not applicable: model illustration, not a constructed project or contract.\nStatistical period: Hypothetical reference case using 2025 cost assumptions; selected-project analysis uses 2024 commissioned projects.\nPage or section: Firm LCOE definition, printed/PDF pp8–9 and p24; Las Vegas illustration, pp25–26; Table 1, p26.\nGeography: Hypothetical Las Vegas, United States reference case; selected global project samples elsewhere.\nUnits: USD/MWh; MW solar capacity; MWh battery capacity; percent of annual demand energy.\nMarket definition: Firm LCOE adds a firming premium. Reliability here means energy coverage of flat annual demand, not system adequacy.\nEvidence classification: Original IRENA model analysis and illustrative scenario.\nRead date: 2026-10-06\nMethod and calculation: Model uses flat hourly demand; 43.5 + 69.7 = 113.2 USD/MWh. Cutoff verification: Korea Institute of Energy Technology Evaluation and Planning, Global Energy catalogue, https://energy.ketep.re.kr/globalenergy/site/main/board/policy_report/31728.\nLimits and uncertainty: The 95% target does not mean uninterrupted supply at every hour or guaranteed peak capacity. Results are location-specific model outputs, not project invoices, average market prices, actual orders or storage returns. Future cost trajectories are conditional projections.\nAttachments and redistribution: Link to the publisher; no original research file is included in the review package.\nOriginal source: 24/7 renewables: The economics of firm solar and wind\n[R04] Public Law 119 21 clean vehicle credit provisions Supports: Section 30D termination changes to vehicles acquired after 30 September 2025.\nInstitution or author: US Congress and Government Publishing Office\nPublication date: 2025-07-04\nEvent date: 2025-07-04 enactment\nStatistical period: Acquisition cutoff 2025-09-30\nPage or section: Sections 70501 to 70503; 139 Stat. 250 to 251; PDF pages 180 and 181\nGeography: United States federal law\nUnits: Dates\nMarket definition: Termination provisions for federal clean vehicle tax credits, especially section 30D.\nEvidence classification: Primary enacted federal statute\nRead date: 2026-10-06\nMethod and calculation: No demand or revenue effect estimated.\nLimits and uncertainty: Does not measure the causal effect on EV sales or imply that all US support disappeared.\nAttachments and redistribution: Publisher link only; no original redistributed.\nOriginal source: Public Law 119 21 clean vehicle credit provisions\n[R01] Adopted amendment on battery due diligence Supports: Moves the application date from 18 August 2025 to 18 August 2027; identifies verification-body and supply-chain preparation constraints.\nInstitution or author: European Parliament and Council\nPublication date: 2025-07-18\nEvent date: 2025-07-18 adoption\nStatistical period: Application date 2027-08-18\nPage or section: Article 1(a), printed page 5 / PDF page 6; recitals 1 to 4, printed pages 2 and 3\nGeography: European Union\nUnits: Dates and legal obligations\nMarket definition: Article 48 battery due diligence obligations; not all battery regulation provisions.\nEvidence classification: Primary adopted legislative text\nRead date: 2026-10-06\nMethod and calculation: Read adopted text; no calculation.\nLimits and uncertainty: Council original rather than unreadable EUR-Lex final journal. Does not establish company compliance or postpone every battery requirement.\nAttachments and redistribution: Publisher link only; no original redistributed.\nOriginal source: Adopted amendment on battery due diligence\n[R02] Council announcement of battery due diligence postponement Supports: Confirms adoption and a two-year postponement; explains time needed for third-party verification arrangements.\nInstitution or author: Council of the European Union\nPublication date: 2025-07-18\nEvent date: 2025-07-18 adoption\nStatistical period: Future application 2027-08-18\nPage or section: Opening and paragraphs on verification bodies; Next steps\nGeography: European Union\nUnits: Calendar dates\nMarket definition: Official announcement of the adopted battery due diligence amendment.\nEvidence classification: Primary official adoption announcement\nRead date: 2026-10-06\nMethod and calculation: Cross-check with R01.\nLimits and uncertainty: Announcement said official publication would follow; it does not establish the eventual journal publication day.\nAttachments and redistribution: Publisher link only; no original redistributed.\nOriginal source: Council announcement of battery due diligence postponement\n[R03] Energy Storage Safety Strategic Plan Supports: LFP retains failure risks; high charging rates can cause lithium plating; a BMS has limited impact after thermal runaway begins.\nInstitution or author: US Department of Energy Office of Electricity\nPublication date: 2024-04; day not stated\nEvent date: Not applicable\nStatistical period: Technical review as of publication\nPage or section: Sections 4.4 and 5.2, PDF pages 25 to 29; cover date on PDF page 1\nGeography: US deployment context; general battery mechanisms\nUnits: Qualitative technical findings\nMarket definition: Lithium-ion stationary storage safety and system engineering.\nEvidence classification: Primary government technical report\nRead date: 2026-10-06\nMethod and calculation: No incident-rate calculation.\nLimits and uncertainty: Not a current local code or a guarantee of safety. Physical damage and interruption differ; no universal loss probability is established.\nAttachments and redistribution: Publisher link only; no original redistributed.\nOriginal source: Energy Storage Safety Strategic Plan\n[R05] UL Solutions announcement on 2025 storage safety testing Supports: UL 9540A provides thermal-runaway propagation test data; UL 9540 supplies complete-system safety criteria and a certification basis.\nInstitution or author: UL Solutions\nPublication date: 2025-04-16\nEvent date: 2025-04-16 announcement\nStatistical period: Fifth-edition testing described in 2025\nPage or section: Paragraphs distinguishing UL 9540A testing and UL 9540 complete-system certification\nGeography: US and Canadian standards context\nUnits: Qualitative testing scope\nMarket definition: Storage testing methods and product safety certification; not incident statistics.\nEvidence classification: Primary testing-provider announcement\nRead date: 2026-10-06\nMethod and calculation: Read public announcement, not paid standards.\nLimits and uncertainty: A testing-provider description of the 2025 edition; not a claim about the latest edition, universal code adoption or zero risk.\nAttachments and redistribution: Publisher link only; no original redistributed.\nOriginal source: UL Solutions announcement on 2025 storage safety testing\nDownload the English evidence register (JSON)\n","date":"2026-10-06T00:00:00Z","image":"/post/renewable-energy-and-batteries/cover-web.webp","permalink":"/post/renewable-energy-and-batteries/","title":"Global Renewable Energy and Battery Industry Report"},{"content":"How engineering information becomes dependable production: the industry's history, value chain, regional markets, business models, and operating challenges. Sources current to 5 October 2026\nGlobal scope, with emphasis on China, Europe and the United States. Market measures retain their exact regions, periods, currencies and definitions. Supporting Evidence appears in a separate appendix after Parts 1–4.\nPart 1: Industry Story On 5 October 2026, Schneider Electric and PTC announced that they had signed a definitive acquisition agreement. The proposed cash price was US$205 per share, valuing PTC\u0026#x27;s equity at approximately US$22.6 billion. A supplier with automation and operations software was seeking to acquire a company whose products manage engineering and product information. The announcement offers a concrete way into a larger industry question: what changes when the information used to design a product is connected to the systems used to make it? The agreement was an announced transaction at this report\u0026#x27;s cutoff; completion is not assumed. [E025, E003, E008, E009]\nThe products make the connection concrete. PTC describes Windchill as a system for product data, lifecycle collaboration and controlled changes, including information from multiple CAD tools. AVEVA, whose remaining shares Schneider acquired in 2023, offers operations software for visualizing equipment, collecting process records and supporting operational decisions. One helps establish and manage what a product is supposed to be. The other helps people understand how an operating process is behaving. Linking those responsibilities is a manufacturing problem with a software and automation business built around it. [E009, E003, E008]\nConsider an engineering change as an analytical example. A revised product definition may require a different production instruction, inspection requirement or machine configuration. The useful outcome is a controlled path through those changes: the relevant people see the approved information, the right system executes it, and the factory can trace what happened. A single corporate owner could make some coordination easier, but ownership alone supplies none of those operating results. Models, interfaces, permissions, testing and responsibility still have to align. This example explains the workflow; it does not describe an undisclosed Schneider or PTC customer project. [E007, E009, E013, E016]\nThe industry\u0026#x27;s central question follows: who can carry engineering information into reliable production, and maintain that connection as products, equipment and security requirements change? The answer involves a long history of translating information into machine actions, a value chain spanning chips to factory operations, and commercial models that reward different kinds of delivery and support. This report follows that sequence. Its global comparison concentrates on China, Europe and the United States, while keeping narrower statistical regions and product categories explicit.\nPart 2: Industry History Putting motion into instructions A useful starting point is the moment when a machine\u0026#x27;s movements could be described as information. In a 1959 MIT programme, researcher J. Francis Reintjes traced the demonstration of numerical-control feasibility to 1952. Encoded numbers on punched tape directed electronic equipment and servomechanisms that moved a machine tool. Douglas Ross then explained APT, which translated descriptions of part geometry and tool movements into machine instructions. The important connection was already visible: engineering information could become a sequence of physical operations. [H001]\nWe can see the beginnings of a recurring division of labour here. The machine, its motion equipment, the programming system and the person translating manufacturing requirements into instructions each contributed something different. Better hardware alone could not remove the work of describing the part and preparing its operations. This helps explain why industrial software developed alongside machinery: the useful product was a working route from design requirements to repeatable execution, with expertise needed at the boundaries. [H001]\nComputer drawing developed a complementary route. Ivan Edward Sutherland\u0026#x27;s January 1963 dissertation described Sketchpad, in which a light pen let a user draw and modify objects on a display. The system stored relationships, reused symbols and applied geometric conditions such as parallel lines. This research example matters because engineering information could carry relationships that changed with a drawing. It was an early technical demonstration, rather than evidence that a commercial CAD market was already established. [H010]\nMaking factory control reconfigurable Programmable control extended this principle to the logic governing factory equipment. Schneider Electric\u0026#x27;s 2019 retrospective dates Modicon\u0026#x27;s conception to 1968 and describes General Motors\u0026#x27; demand for a more flexible alternative to relay wiring. Dick Morley\u0026#x27;s team developed the Modicon 084, allowing functions to be changed through programming without rebuilding the relay connections. The historical significance lies in moving part of the change process into software. It created a continuing role for controller programming, commissioning and maintenance around the physical equipment. [H002]\nProcess plants developed another important control architecture. Yokogawa\u0026#x27;s anniversary account records the announcement of CENTUM in June 1975 and describes its early use of microprocessors and a CRT interface. We can read this DCS milestone as an expansion of the commercial task: suppliers needed to support monitoring and control across a plant, bringing equipment, operator interfaces and engineering together. These capabilities still occupy distinct positions in the industry. Making control programmable also increased the importance of preserving the knowledge embedded in a configured system when a plant later changed or expanded. [H002, H003]\nA Ferranti Argus 500 computer system in Edinburgh, photographed in 1980.\nPhoto: Leocapaldi · Wikimedia Commons · Public domain (author\u0026#x27;s release). Original linked above; web copies resized and converted to WebP, without cropping.\rBuilding product and production models The engineering side was changing too. PTC\u0026#x27;s product history records the launch of Pro/ENGINEER in 1988. Parameters, constraints and associative updates allowed design intent to be carried within a three-dimensional model, so a change could propagate through related elements. This was one company\u0026#x27;s milestone in a much longer CAD history. Its relevance today is the growing value of the model itself: engineering software became a place to maintain relationships and decisions, which could become as consequential to a manufacturer as the drawing or final physical component. [H004]\nPTC introduced Windchill in 1998, using a shared web architecture for lifecycle collaboration. The scope widened from creating a design to coordinating product information across teams. Manufacturing operations needed another kind of shared model. ISA\u0026#x27;s committee history dates the first publication of ISA-95 to 2000: Parts 1 and 2 defined enterprise–control exchanges and their framework, while later parts formalised manufacturing operations management at level 3. These were logical integration boundaries, giving different systems a clearer way to describe their responsibilities and information. [H004, H006]\nThis history helps explain today\u0026#x27;s separate CAD, PLM and manufacturing-operations markets. They grew around different decisions, users and information structures. Connecting them requires agreement on what a product, resource or production operation means, as well as a technical interface. We therefore see migration and integration as work on accumulated models and practices. Moving files or buying a new application addresses only part of that work; the business value depends on preserving useful relationships across the workflow. [H004, H006]\nConnecting systems and renewing software Interoperability became an explicit collective project. The OPC Foundation traces its standards activity to 1996. Its joint group with PLCopen began in 2008, and a controller client specification followed in 2014, supporting exchanges between controllers and with systems such as MES and ERP. The significance was a common route across suppliers and system levels. That route still required implementation and integration. Standardised communication and information models still needed to be matched to the meaning, quality and use of data in a particular plant. [H005]\nGermany\u0026#x27;s Industrie 4.0 initiative, launched in early 2011, gave this integration ambition a broader policy framework. The April 2013 working-group report called for horizontal integration through value networks, engineering integration across the value chain, and vertical integration of networked manufacturing. Those were recommendations requiring further research and implementation. We can see why platform strategies later became attractive: the proposed value depended on connecting previously separate activities. It also made standards, system complexity and the installed factory important to the commercial proposition. [H007]\nLicensing changed alongside this wider scope. PTC announced in January 2018 a broader subscription-only policy for new core-software and ThingWorx licences from January 2019, following a January 2018 transition in the Americas and Western Europe. China and several other markets remained exceptions; existing perpetual licences stayed usable, and Kepware retained both forms. Subscription created a continuing renewal relationship, changing how access, upgrades and spending could be organised. It did not by itself determine where software ran. This distinction remains central when comparing recurring revenue with cloud deployment. [H008]\nBringing AI into industrial engineering On 31 October 2023, Siemens and Microsoft introduced Siemens Industrial Copilot, a jointly developed generative-AI assistant for manufacturing. The announcement described generating, optimising and debugging automation code by combining industrial information with Azure OpenAI Service. The engineering workflow became a meeting point for control expertise and cloud AI capabilities. We can interpret this as another extension of the industry\u0026#x27;s long-running effort to translate intentions into usable instructions, with a new method for interacting with the information and software already involved. [H009]\nThe resulting industry carries several generations of capability at once. Motion and control equipment remain necessary; product and operations models organise decisions; interoperability connects them; licences and services shape the ongoing commercial relationship. AI adds another engineering capability within this accumulated structure. Our reading of the history is that competitive value depends on making these layers work together through actual manufacturing changes. That explains the continuing importance of component suppliers, software specialists, platform companies and integrators, each serving a different part of the same practical task. [H001, H002, H003, H004, H005, H006, H007, H008, H009]\nPart 3: Market Landscape The value chain begins with a physical measurement Industrial automation turns measurements into actions: a sensor detects a condition, a controller evaluates it, and a drive or actuator changes the physical process. Industrial software also defines the product, tests its design, schedules and records production, and manages changes across its life. The useful industry boundary therefore follows the manufacturing workflow. It includes components, control systems, engineering and operations software, robotic equipment, machine integration and factory use. This is the working scope of this report; individual market statistics below retain their publishers\u0026#x27; definitions. [E004, E006, E007, E020]\nUpstream suppliers make the ingredients of an automation system. Texas Instruments lists analog and embedded devices for sensing, communication, motor control and power conversion. SICK\u0026#x27;s 2022 business description covers sensors, cameras and encoders; Inovance\u0026#x27;s product directory includes AC drives, servo drives and motors. Their immediate customers can be equipment and system manufacturers rather than the final factory alone. These examples establish product roles, without establishing a procurement relationship between any two named companies. [E004, E005, E021, E023]\nThe value of this layer is application fitness. A sensor must capture the relevant physical signal, a processor must handle the required computation, and a drive must work with the selected motor and machine. In our analysis, a lower component price creates value only if qualification, replacement availability and engineering work remain acceptable over the equipment\u0026#x27;s service life. NIST\u0026#x27;s discussion of long-lived operational technology explains why support and compatibility belong in the purchasing decision. The available product descriptions do not measure global component shortages or any country\u0026#x27;s import dependence. [E004, E005, E013, E021]\nControl, operations and engineering solve different tasks In the middle of the chain, PLCs execute machine logic and motion tasks; distributed control systems coordinate process operations. Siemens\u0026#x27; 2024 SIMATIC S7-1200 G2 announcement describes TIA Portal engineering and motion functions. SUPCON describes the ECS-700 DCS within an architecture of instruments, I/O, operator and engineering stations, process records and related systems. Inovance\u0026#x27;s H5U PLC provides another disclosed machine-control example. A comparison between these products must start with the process and required control functions, rather than treating every controller as a substitute for every other controller. [E006, E019, E024]\nA WAGO controller in a pharmaceutical monitoring setup, photographed in 2016.\nPhoto: Pierre75000 · Wikimedia Commons · CC BY-SA 4.0 International. Original linked above; web copies resized and converted to WebP, without cropping.\rAbove the immediate control task, HMI and SCADA present operating conditions, while historians preserve process data. AVEVA Operations Control packages visualization, historian, reporting and communication tools with subscription entitlements. MES addresses production execution: recipes, work in progress, quality, material tracking and product genealogy. Rockwell places this software between enterprise systems and shop-floor controls and identifies FactoryTalk ProductionCentre and Plex as examples. For the factory, the commercial question is whether the software can make production information usable at the point of a decision, and whether its interfaces and support fit existing operations. [E007, E008]\nEngineering software acts earlier in the workflow. CAD models product geometry; simulation tests engineering behaviour; PLM governs product information, revisions and change processes. Dassault Systèmes assigns these roles to CATIA, SIMULIA and ENOVIA, and describes DELMIA across manufacturing planning and execution. PTC\u0026#x27;s Windchill manages lifecycle data and collaboration, including multi-CAD information and change traceability. The economic asset here is the accumulated product definition and the work built around it. We infer that migration must preserve usable relationships and revision history, in addition to transferring files; this is a consequence of the disclosed functions, not a measured switching-cost estimate. [E009, E020]\nRobotic equipment occupies another middle-chain role: it carries out physical manufacturing tasks under a configured control system. Inovance\u0026#x27;s portfolio includes industrial robots as well as drives and controllers, showing how a supplier can span several branches. A machine builder or factory engineering team must turn the equipment into an application. Our commercial analysis separates the robot delivery from tooling, programming, cell integration and ongoing support; the agreed contract determines which party provides each. The map below separates functions so that a broad company portfolio is not mistaken for a single product market. It also includes named examples of a machine builder and a factory user, rather than assuming that a robot sale completes an entire production system. [E021, E023, E012]\nStäubli industrial robots on a manufacturing line, photographed in January 2023.\nPhoto: Clemenspool · Wikimedia Commons · CC0 1.0. Original linked above; web copies resized and converted to WebP, without cropping.\rINDUSTRY MAP\nIndustrial automation and industrial software: the value chain Global representative products and users. Expand a branch to explore the four levels: industry, stage, function, and company or product.\nFit to view Expand all Collapse Full screen Read the full text outline Industrial Automation and Industrial SoftwareUpstreamSemiconductorsTexas Instruments analog and embedded chips [E004] Sensing and machine visionSICK sensors, cameras and encoders [E005] Drives and motorsInovance AC drives, servo drives and motors [E021] MidstreamPLC and distributed control systemsSiemens SIMATIC; SUPCON ECS-700 DCS [E019, E006] Inovance H5U PLC [E024] HMI, SCADA and process dataAVEVA Operations Control [E008] Manufacturing execution systemsRockwell FactoryTalk ProductionCentre and Plex [E007] CAD and engineering simulationDassault Systèmes CATIA and SIMULIA [E020] Product lifecycle managementPTC Windchill; Dassault Systèmes ENOVIA [E009, E020] Manufacturing planning and operationsDassault Systèmes DELMIA [E020] Industrial robotic equipmentInovance industrial robot portfolio [E021] DownstreamMachine building and integrationthyssenkrupp Automation Engineering [E012] Factory operationsSiemens Electronics Factory Erlangen [E012] Connections classify product roles; they do not establish procurement contracts. Companies may appear in several roles. Evidence cutoff: 5 October 2026.\nSources: E004 · E005 · E021 · E019 · E006 · E024 · E008 · E007 · E020 · E009 · E012. Reviewed 2026-10-06.\nFigure 1. Industry value chain: global representative products and users. The hierarchy is industry → upstream, midstream and downstream → functional segment → company or product. Source references appear at the leaves; branches classify roles. A company can supply equipment and software while also operating its own factory. [E004, E005, E006, E007, E008, E009, E012, E019, E020, E021, E024]\nDownload printable map (SVG) · Download map nodes (CSV)\nThe downstream customer buys a working production process Machine builders and integrators connect mechanical equipment, control programs and information flows for an application. Factory operators then run, maintain and modify the resulting assets. Siemens\u0026#x27; November 2024 release identifies thyssenkrupp Automation Engineering as a special-machine and plant builder using its engineering copilot in a battery quality-inspection machine. The same release describes an operations-copilot application on soldering machines at Siemens\u0026#x27; Electronics Factory Erlangen. These are distinct positions in the chain, even though a common supplier supports both. The disclosure verifies the applications; it supplies no independently audited payback period. [E012]\nBusiness models consequently follow different deliverables. A component supplier provides devices; a controls supplier combines equipment with engineering tools and support; an engineering-software vendor licenses software supporting a workflow; and a service organization provides agreed implementation, maintenance or continuing support. AVEVA explicitly offers subscription software across on-premises, cloud and hybrid deployments. PTC\u0026#x27;s licensing transition also left existing perpetual rights intact. Subscription describes a payment and entitlement relationship; deployment describes where the software runs. Keeping these dimensions separate is essential when comparing commercial offers. [E008, H008, E104]\nFor the buyer, our suggested comparison is the cost of the complete operating workflow: equipment and licences, integration and validation, training, support, future changes and a feasible exit. For the supplier, the corresponding questions are what is delivered once, what must be maintained over time, and which party bears the cost of an implementation problem. This framework explains why recurring revenue can coexist with substantial engineering work. It also explains why a vendor\u0026#x27;s software label cannot, on its own, establish either its profit margin or the customer\u0026#x27;s total cost. The segment disclosures later in this section provide concrete accounting evidence. [E008, E013, E104]\nMarket size depends on what is being counted The available evidence describes several substantial markets. Three figures are useful because each points to a different source of demand: engineering software used to develop products, industrial-software product revenue reported in China, and control systems sold into Chinese process industries. Their boundaries need to remain visible when discussing market size. [E207, E208, E201]\nScope and sourceStatistical year; amountPublished growthGlobal broad PLM economy — CIMdata estimate2025; USD 88.3bn+9.9%China industrial-software products — MIIT2025; RMB 333.0bn+9.7%China DCS — MIR estimate2024; approximately RMB 11.76bn−3.6%\rThese are separately defined revenue/value measures, in their original currencies. They are not components of a combined total. Sources: [E207, E208, E201].\nCIMdata’s PLM definition stretches well beyond a product-data-management application. It includes engineering tools such as mechanical CAD, CAM, simulation and analysis, electronic-design automation and architecture/engineering/construction software, alongside innovation platforms and digital manufacturing. Its release identifies notable growth in EDA and AEC. This helps explain why a broad engineering-software economy can grow for reasons extending beyond manufacturers’ spending on shop-floor systems. China’s MIIT product-revenue category covers a different geography and statistical boundary; DCS includes hardware, software, services and engineering. Adding the three amounts would therefore mix different years, overlap software categories and combine unlike businesses. [E207, E208, E201]\nMIIT reports RMB 333.0bn of industrial-software product revenue for 2025 and growth of 9.7%. Its earlier 2024 release reported RMB 294.0bn. Dividing those two published amounts produces approximately 13.3%: (333.0/294.0 − 1) × 100. The basis for reconciling those amounts with the official growth rate has not been established here, so 9.7% remains the cited growth rate and the two releases are not turned into a continuous growth chart. [E208, E209]\nMIR’s public DCS estimate offers a different competitive signal: a market can contract while an established domestic supplier retains a substantial position. The excerpt places SUPCON at 40.4% of China’s 2024 DCS market, whose delivered offering includes services and engineering. This attributed estimate identifies a specific process-control position; it does not establish supplier profitability or individual system prices. [E201, E202]\nDifferent regions, different installation cycles Industrial-robot installations provide a cleaner physical comparison across regions. IFR reports that global additions exceeded 600,000 in 2025, up 11%. China expanded faster than that global pace, the United States also grew, and EU27 installations fell. These figures describe new industrial robots installed during the same calendar year, making the direction of demand comparable even though the regions differ greatly in population and manufacturing scale. [E204, E205, E206]\nMarket2025 new installations (units)Year-on-year changeChina354,000+20%United States38,400+12%EU2760,500−11%\rIFR public releases. US value follows the country-specific release; the global summary says almost 38,500. EU27 is not all Europe. Counts retain published precision. Sources: [E204, E205, E206].\nChina represented 59% of global deployments according to IFR. Chinese suppliers installed 195,000 units in their home market, up 15%, but their category share fell from 57% to 55% as total Chinese installations grew faster. The point is economically useful: increasing domestic suppliers’ volume can coexist with a lower share of a faster-growing market. The evidence does not identify a single cause, such as price, quality or policy, for that change. [E204]\nINDUSTRIAL ROBOT SUBMARKET · Calendar 2025\nChina industrial-robot installations by supplier origin Annual new industrial-robot installations, grouped by supplier origin. IFR reports approximately 354,000 installations in China in 2025.\nChina · Share of annual new installations (%)\nFull screen View the data table China industrial-robot installations by supplier origin · Calendar 2025 · Share of annual new installations (%) Supplier originShare Chinese suppliers55%All other suppliers45% Download data (CSV) IFR directly reports Chinese suppliers at 55%; all other suppliers = 100% − 55% = 45%. Percentages retain the published integer precision. These are shares of installations in a named submarket, not global automation or software revenue, or individual vendor shares. [E204]\nSource: International Federation of Robotics (IFR) — Five Million Robots now Operate in Factories Globally (2026-09-24). Reviewed 2026-10-06.\nFigure 2. China industrial-robot installations by supplier origin, 2025: Chinese suppliers 55%; all other supplier origins 45%. The second category is calculated as 100% − 55%. Denominator: annual new installations in China; source: IFR [E204].\nThe chart is a supplier-origin split, not a ranking of individual robot manufacturers. It also counts installations rather than sales revenue: a unit installed in one application need not have the same selling price or integration cost as a unit in another. Its value is to show how much of China’s annual deployment is supplied by the domestic category, without pretending to measure the whole value captured by automation suppliers. A common vendor-share denominator covering this report\u0026#x27;s combined global scope was not established in the reviewed public evidence. ARC/Control\u0026#x27;s automation definition, for example, excludes robotics and several other categories. The chart consequently uses the explicitly defined robot submarket and installation measure. [E204, E212]\nThe United States illustrates the distinction between robot manufacturing and deployment capability. IFR describes many domestic system integrators, while saying that most robots are still imported from Japan and Europe. Automotive remained its largest customer sector, with 13,500 installations, down 1%, even as total installations rose 12%; food and beverage grew 30% to 2,900. Those figures support a reading of demand spreading beyond the largest established customer sector, rather than an automotive-led explanation for all growth. [E205]\nEU27 presents a large installed base with weaker new investment. Its operating stock reached 712,000 units at year-end 2025, while new installations fell 11%. Automotive additions declined 25% to 14,900 and metal-industry additions 13% to 13,000; food and beverage increased 4% to 5,300. This sector pattern helps explain the aggregate decline without claiming that it exhausts the causes. Germany accounted for 41% of EU27 installations, making the largest national market particularly important to regional equipment demand. A declining annual flow does not erase the existing base that may need maintenance, integration and upgrades; that last implication is an analytical inference, not a measured service-market estimate. [E206]\nReading company results through their business models Recent company disclosures show where growth and profit appeared within large suppliers. Siemens and Rockwell report fiscal third quarters ending 30 June 2026; Schneider’s comparison below uses the six months ending on that date. Each operates globally, so these are company operating results rather than indicators of demand in its home country. [E101, E102, E105]\nBusiness; reporting periodRevenue and growthProfit and recurring-business measuresSiemens Digital Industries; Q3 FY2026EUR 4.9bn; +10% comparable. Software subset EUR 1.8bn; +15%.Segment profit EUR 923m; margin 18.7%. ARR EUR 5.7bn; +11% organic.Rockwell Software \u0026amp; Control; Q3 FY2026USD 751m; +19% reported/+18% organic.Segment operating earnings USD 261m; margin 34.8%. Group organic ARR +6%.Schneider Industrial Automation; H1 calendar 2026EUR 3.585bn; +7.7% organic.Adjusted EBITA EUR 501m; margin 14.0%. AVEVA ARR +11% at 30 June.\rCurrencies and reporting periods remain separate. Schneider’s Q2 Industrial Automation growth was 11% organic [E103]; the table uses H1 amounts from E105. Profit measures are company-defined; group ARR and AVEVA ARR are labelled separately. Sources: [E101, E102, E103, E105].\nSiemens provides a direct link between the software contribution and segment profitability: Digital Industries’ profit rose 44%, and the company identified software as the largest contributor to the improvement. The segment margin rose from 14.5% to 18.7%. Orders were EUR 4.9bn, up 9% comparably, alongside revenue of the same rounded amount, up 10%. Orders indicate incoming business; revenue records what was recognized in the quarter. ARR supplies a separate view of recurring business. Their different growth rates describe different commercial stages, rather than three versions of the same sale. [E101]\nRockwell permits a useful comparison within one company and quarter. Segment operating margins were 34.8% in Software \u0026amp; Control, 20.0% in Intelligent Devices and 15.1% in Lifecycle Services. Its filing describes predominantly point-in-time product revenue in Intelligent Devices, a product/software combination in Software \u0026amp; Control, and predominantly over-time solutions/services revenue in Lifecycle Services. Large systems and services are mainly sold directly; products also rely on distributors. Our interpretation is that product reuse, engineering delivery and the timing of contract performance create different operating demands, so an automation supplier needs more than one earnings model. [E102, E104]\nHigher volume was Rockwell’s stated main driver of Software \u0026amp; Control margin improvement. Strong project execution and the Sensia dissolution benefited Lifecycle Services’ margin despite lower sales. Schneider similarly reported support-function cost leverage while Industrial Automation’s gross margin slipped slightly because productivity and pricing did not fully offset raw-material inflation and tariffs. Its adjusted EBITA margin nevertheless improved from 13.7% to 14.0%. These disclosures show why revenue growth and margin movement need separate explanations; they do not isolate software subscriptions as the sole cause of profitability. [E102, E105]\nContract indicators add another dimension. Rockwell disclosed about USD 1.375bn of remaining performance obligations, of which about USD 820m was expected to become revenue within 12 months. The measure excludes specified short contracts, invoice-based services and unexercised renewals. Its ARR instead represents the annual contract value of active recurring contracts and is expressly separate from revenue, contract liabilities and backlog. A growing ARR can help track a recurring relationship; RPO can help assess committed work still to be recognized. Neither alone supplies a complete forecast of future sales or cash. [E102, E104]\nSchneider’s Industrial Automation business also grew unevenly by geography: H1 organic growth was 5% in North America, 7% in Europe and 12% in China and East Asia. The company associated Chinese growth with discrete automation, packaging/material handling and improving OEM activity, while AVEVA contributed across regions. These are segment-specific explanations for 2026 sales, alongside the different 2025 robot-installation cycle. Read together, the evidence points to a market whose direction depends on application, geography and how the supplier earns and recognizes revenue. [E105, E204, E205, E206]\nCompetition moves across the workflow The competitive landscape has three overlapping forms. Controls companies can add engineering and operations software to their installed equipment base. Software specialists can deepen the product model or manufacturing workflow. Integrators can combine products from different suppliers for a particular process. These positions are visible in the portfolios and applications above. In our analysis, the practical dividing line is where a supplier takes responsibility: an interface, a software task, a configured machine or a continuing production outcome. Portfolio breadth and implementation responsibility should therefore be assessed separately. [E006, E007, E008, E009, E012, E020]\nCompleted acquisitions show established suppliers extending that scope. Schneider Electric completed the purchase of the AVEVA shares it did not already own on 18 January 2023. Emerson completed its purchase of the remaining AspenTech shares on 12 March 2025, making it a wholly owned subsidiary within Control Systems \u0026amp; Software. Siemens completed its Altair acquisition on 26 March 2025 for an enterprise value of approximately US$10 billion, adding simulation, high-performance computing, data science and AI capabilities. These transactions have different histories; in particular, buying remaining shares is a change from an existing ownership position. [E003, E002, E001]\nThe strategic attraction is a broader route from engineering decisions to operating information. Our interpretation is that a connected portfolio can reduce handoffs if models, permissions and change processes actually align. It can also concentrate dependency in one supplier and create a larger migration task when a customer changes platforms. Acquisition announcements establish ownership and stated intentions; the buyer still needs evidence of compatible versions, usable interfaces, support terms and application-level results. The disclosed transactions do not supply a common measure of global industry concentration. [E001, E002, E003, E009, E016]\nFor a supplier, this makes the renewal and expansion of a working customer process a useful test of competitiveness. For a customer, it makes continuity through the next product change, equipment upgrade or security update a useful test of value. These are our proposed evaluation criteria, rather than a ranking of companies. They connect the regional demand and segment economics above to the central industry question: which supplier can carry engineering information into dependable production while keeping the system supportable over time? [E009, E012, E013, E104]\nPart 4: Industry Challenges Existing factories make verification part of the product A factory upgrade must fit a process that already works. NIST\u0026#x27;s OT security guidance describes requirements for predictable response, safety and continuity, alongside legacy systems that may run unsupported operating systems. Routine rebooting can be unsuitable, and software changes require testing and staged implementation. The resulting challenge extends beyond installing a newer application: a change in communications, controller software or equipment behaviour can affect production and safety together. In our assessment, the scarce resource in a brownfield project is often a controlled opportunity to verify the whole workflow, including its response when a component fails. [E013]\nA practical approach is to document dependencies, test changes away from the live process where feasible, agree an acceptance window, and prepare recovery steps before deployment. Buyers also need a support plan covering controller firmware, operating systems, application versions and available engineering expertise. These are analytical recommendations derived from the operating constraints, rather than a universal installation recipe. A test environment may omit physical conditions present in production; a successful trial therefore reduces uncertainty without removing the need for site acceptance. Support commitments also have to survive future version changes and changes of supplier. [E013]\nInteroperability needs meaning, permission and responsibility The OPC Foundation describes OPC UA as a framework combining communications, information models and access mechanisms, with authentication, encryption and auditing. IDTA\u0026#x27;s 10 June 2025 AAS announcement adds fine-grained access control for properties, submodels and registry or repository services. These mechanisms offer a useful route through a basic integration problem: two systems may exchange a value yet interpret its identity, units, revision or permitted use differently. Their availability in a standard establishes a design path; implementation in a particular product and agreement between particular systems still require verification. [E016, E017]\nOur proposed starting point is a bounded workflow, such as carrying an approved engineering change into production records. PTC describes Windchill\u0026#x27;s multi-CAD management, controlled product changes and traceability, giving that workflow an identifiable information-management role. Implementation then needs agreed identifiers, version rules, authorized users and responsibility for exceptions. Technical permissions can restrict access to selected fields, while contracts must establish who may retain, reuse or transfer the data. Neither layer resolves every semantic disagreement. Buyers should therefore assess export and migration arrangements alongside the interface demonstration, with validation costs included in the commercial decision. [E009, E016, E017]\nAI assistance shifts work towards review and verification Siemens and Microsoft introduced Industrial Copilot on 31 October 2023 with functions for generating, optimizing and debugging automation code. Siemens\u0026#x27; November 2024 release described a thyssenkrupp Automation Engineering battery-inspection machine using the engineering assistant, and an operations assistant used on soldering machines at Siemens\u0026#x27; Erlangen factory. These vendor-disclosed applications show where assistance enters engineering and maintenance. The difficult transition is from a plausible answer to an approved machine change: code must fit the controller, process sequence and failure conditions, while maintenance advice must match the installed equipment and current documentation. [H009, E012, E013]\nWe would evaluate an assistant through a reviewed workflow: identify its information sources and versions, restrict access, test proposed code, and retain human approval for deployment. Evaluation should count review time, integration, computing and ongoing support alongside time saved in drafting. Rockwell\u0026#x27;s filing distinguishes product, software and services delivery, a reminder that the commercial package extends beyond an assistant\u0026#x27;s licence. The cited AI announcements provide no independently measured return on investment. A useful pilot should consequently measure accepted work and its total cost at the customer site, including corrections, rather than treating the volume of generated code as the outcome. [H009, E012, E013, E104]\nSecurity duties continue after installation Europe\u0026#x27;s Cyber Resilience Act gives lifecycle security a concrete timetable. Reporting obligations began on 11 September 2026: the Commission describes an early warning within 24 hours of awareness and full notification within 72 hours for relevant actively exploited vulnerabilities and severe security incidents. These are reporting deadlines. The main product obligations apply from 11 December 2027. Application depends on the product and statutory scope, so a manufacturer needs to establish its actual role before assigning responsibilities. The operational consequence is already clear for covered reporting: someone must recognize an incident, gather information and escalate it in time. [E014, E022]\nNIST\u0026#x27;s recommended segmentation, authorized communications and OT-specific training provide practical engineering paths. Our analysis is that these measures work best when linked to maintenance ownership and recovery procedures: a restricted network still needs necessary production traffic, and trained staff still need access to current equipment information. China\u0026#x27;s AI + Manufacturing policy, published on 7 January 2026, similarly emphasizes industrial reliability, safety, data governance and evaluation, with objectives for 2027. Those objectives indicate policy direction. Project results must be assessed separately through operating evidence, with responsibilities and validation suited to the process in each jurisdiction. [E013, E015]\nSupply choices and delivery economics meet at acceptance Industrial software depends on physical equipment. Texas Instruments describes sensing, communications, motor-control and power-conversion components; Inovance lists drives, motors, controllers and interfaces. These product roles help locate dependencies, while establishing neither a particular supplier relationship nor a current shortage. Our supply-chain analysis therefore starts with qualification: replacing a component or drive may require checking electrical behaviour, interfaces, software compatibility and support arrangements. Alternative suppliers can create options, but each qualified configuration takes engineering work. Procurement should weigh that work against price and availability instead of assuming that products in the same category are interchangeable. [E004, E021, E013]\nDelivery economics differ across the chain. Rockwell\u0026#x27;s quarterly filing for the period ended 30 June 2026 describes product revenue recognized predominantly at a point in time for Intelligent Devices, a product-and-software mix in Software \u0026amp; Control, and mainly over-time solutions and services in Lifecycle Services. In our assessment, reusable software and standard configurations can spread development work across customers, while site integration, acceptance and continuing support consume resources with each deployment. Revenue-recognition categories alone establish neither cash timing nor software-only profitability. A sound decision therefore asks what will be accepted, who will maintain it, how a change will be verified, and what the full supported workflow will cost. [E104, E013]\nAppendix: Supporting Evidence Each record identifies the claim, source, dates, statistical scope and limits. Company disclosures establish what the company reports; third-party estimates retain their original definitions. The evidence cutoff is 5 October 2026; verification includes reading on 6 October. Original research materials are linked to their publishers.\nDownload the complete English report (TXT) · Download source records (JSON)\n[E025] Schneider Electric to acquire PTC, creating the next level of Energy and Industrial IntelligenceSupported claim: On 5 October 2026, Schneider Electric and PTC announced that they had signed a definitive acquisition agreement. The proposed cash price is US$205 per share, valuing PTC equity at approximately US$22.6 billion.\nSchneider Electric and PTC; financial release hosted by Schneider Electric · Primary joint acquisition-agreement announcement, original PDF first page visually read\nRead the original source\nDates, definitions and reading limitsPublication date5 October 2026, release dateline and official index.Event dateAnnouncement of a signed agreement on 5 October 2026; actual execution date and completion not established from the read page.Statistical periodAnnouncement event at the report\u0026#x27;s evidence cutoff.Source locationOriginal 10-page PDF, file/printed page 1: dateline, first paragraph and first Key Highlights bullet. Official index corroborates date and announcement identity.GeographyFrance and United States, as in the joint-release dateline; global industrial-software transaction.Metric and unitsUS$205 per share; approximately US$22.6 billion equity value for 100% of share capital. Equity value is distinct from enterprise value, revenue and market size.DefinitionAn agreement to acquire PTC, distinct from a completed acquisition and from market revenue.Support and limitsOnly original page 1 was visually read. Completion conditions, closing timetable, financing and the remaining nine pages are not used. Stated strategic benefits and synergies are management expectations, not achieved integration results.CalculationNone.Reading statusFirst page actually viewed at readable original resolution by both reference researcher and root on 6 October 2026. Official index actually read. Full PDF not read; inaccessible SEC/search excerpts are not the verification basis.Reuse of original materialIssuer link only; no original material redistributed.Read date2026-10-06\r[E003] Schneider Electric announces completion of transaction to acquire entire share capital of AVEVASupported claim: Schneider Electric announced on 18 January 2023 that the scheme to acquire the AVEVA shares it did not already own had become effective and the transaction was complete.\nSchneider Electric · Primary company transaction-completion announcement\nRead the original source\nDates, definitions and reading limitsPublication date18 January 2023Event dateTransaction completed on 18 January 2023.Statistical periodNot applicable: transaction event.Source locationSingle-page PDF, file page 1, title and first two body paragraphs; lines 16–27. No printed page number.GeographyFrench acquirer, UK software business; global operations.Metric and unitsOwnership and completion date. No market metric used.DefinitionCompany ownership, not a market denominator.Support and limitsSupports the ownership note for AVEVA. It is an earlier portfolio example, not a 2025 event or evidence that later product integration has been completed.CalculationNone.Reading statusRead the cited sections of the publisher\u0026#x27;s original on 5 October 2026. This is not a claim that the entire source was read.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-05\r[E008] AVEVA Operations Control SoftwareSupported claim: AVEVA Operations Control offers subscription entitlements including HMI/SCADA, historian, reporting, communication drivers and collaboration, with on-premises, cloud and hybrid deployment options and disclosed MQTT, OPC UA and REST interfaces.\nAVEVA · Primary vendor product, deployment and subscription description\nRead the original source\nDates, definitions and reading limitsPublication datePublication and update dates not disclosed.Event dateNot applicable: product and subscription description.Statistical periodNot applicable: no market statistics.Source locationProduct overview, lines 17–21; Deployment options, lines 108–115; interface support, lines 119–121; Subscription Entitlements, lines 124–126.GeographyGlobal vendor product description; regional contract terms not disclosed.Metric and unitsSubscription portfolio and deployment options, not market revenue.DefinitionOperations-control software bundle spanning visualization, data collection and related tools.Support and limitsVerifies the subscription and deployment choices. It does not independently verify unlimited performance, customer ROI or effortless interoperability. Features cannot be backdated to unspecified earlier versions.CalculationNone.Reading statusRead the cited sections of the publisher\u0026#x27;s original on 5 October 2026. This is not a claim that the entire source was read.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-05\r[E009] Windchill PLM Software | Enterprise PLM SystemSupported claim: PTC identifies Windchill as enterprise PLM connecting product data, processes and people across the product lifecycle, with multi-CAD management and connectors. Its regulated-industry section also describes traceability for product data and changes, document control and audit trails.\nPTC · Primary vendor product description\nRead the original source\nDates, definitions and reading limitsPublication datePublication and update dates not disclosed.Event dateNot applicable: product functions and roles.Statistical periodNot applicable: no market statistics.Source locationWhat is Windchill?, web lines 1–3; multi-CAD section, lines 401–408; regulated-industry section, lines 409–418.GeographyGlobal vendor product description; regional deployment and sales breakdowns are not disclosed.Metric and unitsPLM functions and product identity; no customer improvement or market figures used.DefinitionEnterprise product lifecycle and data/process management, distinct from CAD geometric design and PLC real-time control.Support and limitsSupports the disclosed PLM, change-traceability and multi-CAD roles. It does not prove cost-free migration, universal lossless format conversion, achieved regulatory compliance or industry-wide customer benefits.CalculationNone.Reading statusRead the cited sections of the publisher\u0026#x27;s original on 5 October 2026. This is not a claim that the entire source was read.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-05\r[E007] What is Manufactuing Execution System?Supported claim: Rockwell Automation places MES between shop-floor controls and enterprise systems, with recipe, quality, work-in-progress and genealogy, performance and material-tracking functions, and lists FactoryTalk ProductionCentre and cloud-native Plex as examples.\nRockwell Automation / FactoryTalk · Primary vendor MES product and application explanation\nRead the original source\nDates, definitions and reading limitsPublication datePublication and update dates not disclosed.Event dateNot applicable: product and architecture description.Statistical periodNot applicable: no market statistics.Source locationWhat is MES?, lines 44–61; Do I Need an MES?, lines 83–94; MES and ERP, lines 113–117; examples, lines 145–146.GeographyUS vendor site describing general manufacturing applications; no regional sales data.Metric and unitsFunctions and product categories.DefinitionManufacturing execution software between enterprise planning and shop-floor control.Support and limitsSupports MES functionality and product identity. It does not establish that MES replaces PLC/DCS, that every customer achieves value in three months or that cloud deployment itself proves a particular pricing model.CalculationNone.Reading statusRead the cited sections of the publisher\u0026#x27;s original on 5 October 2026. This is not a claim that the entire source was read.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-05\r[E013] Guide to Operational Technology (OT) SecuritySupported claim: NIST describes OT requirements for deterministic response, safety and continuity, warns that routine rebooting can be unsuitable, and calls for testing and staged software changes. Its guidance also covers network segmentation, authorized flows and training tailored to OT roles. Its typical OT component-life illustration is 10–15 years, and legacy systems may use unsupported operating systems.\nNational Institute of Standards and Technology (NIST); Keith Stouffer et al. · Primary official technical guidance; not a universal mandatory law\nRead the original source\nDates, definitions and reading limitsPublication dateSeptember 2023; exact day not disclosed.Event dateNot applicable: published guidance. The CSRC page notes a Revision 4 initial public draft dated 21 September 2026; that draft is not adopted final guidance.Statistical periodNot applicable: technical guidance, not a market or lifetime survey.Source locationSection 2.3 and Table 1, printed pp.28–31 / file pp.45–48; network architecture printed pp.71–72 / file pp.88–89; segmentation printed p.102 / file p.119; training printed p.108 / file p.125.GeographyUS NIST guidance; technical discussion can inform global OT analysis, without imposing US law abroad.Metric and unitsA typical OT component-life comparison of 10–15 years is a guide-level illustration, not every installation\u0026#x27;s lifetime.DefinitionOperational technology controlling or interacting with physical systems, including industrial control; not limited to manufacturing.Support and limitsSupports the E013 engineering mechanisms, segmentation and training paths. It does not quantify a skills shortage, retrofit budget, failure probability or guaranteed security result. Deployment must account for the specific process, safety and communications. The Revision 4 initial public draft is not adopted final guidance.CalculationNo calculation. File pages are one-based; cited zero-based PDF indices are converted by adding one.Reading statusSection 2.3 and Table 1 reread on 6 October 2026. Architecture, segmentation and training sections read on 5 October 2026; entire 316-page guide not read.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-06\r[E016] Unified Architecture – LandingpageSupported claim: The OPC Foundation describes OPC UA as a platform-independent framework combining communications, information models and access mechanisms, with encryption, signing, authentication and audit functions for information exchange from devices to enterprise and cloud systems.\nOPC Foundation · Primary technical explanation by the standard-maintaining organization\nRead the original source\nDates, definitions and reading limitsPublication datePage publication and update dates not disclosed; 2008 is the original UA release date, not this page\u0026#x27;s version date.Event dateNot applicable: architecture and mechanism description.Statistical periodNot applicable: no adoption statistics.Source locationArchitecture, Platform Independence and Security; Information Modeling and Access, web lines 103–157, particularly lines 128 and 144–157.GeographyGlobal industrial interoperability standard.Metric and unitsTechnical features only; companion-specification counts not used.DefinitionCommunications, information modelling and access infrastructure; not a substitute for MES or PLM application functions.Support and limitsSupports a path to common interfaces and semantics. It does not show that connecting via OPC UA automatically removes model, real-time, version or engineering differences. Security depends on configuration and operations.CalculationNone.Reading statusCited original sections reread on 6 October 2026; earlier verification on 5 October 2026.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-06\r[H001] MIT Science Reporter—Automatically Programmed Tools (1959)Supported claim: In this 1959 programme, MIT researchers date the demonstration of numerical-control feasibility to 1952. They explain how APT translated descriptions of part geometry and tool movements into machine-control instructions.\nMassachusetts Institute of Technology; J. Francis Reintjes and Douglas Ross interviewed by Robert Woodbury · Primary historical programme, preserved by MIT with a transcript\nRead the original source\nDates, definitions and reading limitsPublication date1959 programme; archive posting date not disclosed.Event date1952 feasibility demonstration; APT demonstrated and explained in the 1959 programme.Statistical periodHistorical events, not market statistics.Source locationTranscript: Reintjes on numerical control, web lines 30–36; Ross on the APT workflow, lines 57–66.GeographyUnited States; MIT and aircraft manufacturing.Metric and unitsDates and technical workflow; no sales or productivity measurement.DefinitionNumerical control using encoded instructions and servomechanisms; APT part programming. This is not a claim that these machines had modern CNC architecture.Support and limitsSupports a specific demonstration and programming mechanism. It does not establish sole invention, a universal first, or quantified commercial impact. Archive upload timing is unknown; the preserved programme itself dates to 1959.CalculationNone. Connections to today\u0026#x27;s software-controlled manufacturing are author interpretation.Reading statusOriginal archive transcript sections cited above read on 6 October 2026; video not watched.Reuse of original materialLink to MIT; no original video or transcript redistributed.Read date2026-10-06\r[H010] Sketchpad A man-machine graphical communication systemSupported claim: Ivan Edward Sutherland\u0026#x27;s January 1963 dissertation describes Sketchpad: drawing directly on a display with a light pen, reusing symbols and imposing geometric conditions while preserving drawing relationships.\nIvan Edward Sutherland; University of Cambridge Computer Laboratory archival edition; original dissertation at MIT · Primary research dissertation in an archival reprint\nRead the original source\nDates, definitions and reading limitsPublication dateOriginal dissertation January 1963; Cambridge technical report September 2003.Event dateResearch system described in the January 1963 dissertation.Statistical periodHistorical research, not commercial market statistics.Source locationArchival PDF file pp.1–2 date and origin; file pp.9–10 Abstract; file pp.17–18 Introduction. MIT catalogue separately dates the dissertation to 1963.GeographyUnited States research; UK archival publication.Metric and unitsDates and system functions; no sales or productivity figures.DefinitionInteractive computer drawing and geometric constraints, a research milestone relevant to CAD.Support and limitsNot a claim that CAD began here or that a commercial industry was established in 1963. Original printed pagination differs from archival file pagination.CalculationNone.Reading statusSpecified archival original-text sections and MIT catalogue actually read on 6 October 2026; entire 149-page reprint not read.Reuse of original materialPublisher link only; no dissertation or image redistributed.Read date2026-10-06\r[H002] Schneider Electric celebrates 50 years of Modicon, the programmable controller that maximizes operational profitability (translated title)Supported claim: Schneider\u0026#x27;s 2019 retrospective dates Modicon\u0026#x27;s conception to 1968. It describes a General Motors request, Dick Morley\u0026#x27;s team and the Modicon 084: electronic programmable control allowed changes without reconfiguring relay wiring.\nSchneider Electric · Primary company historical retrospective, in French\nRead the original source\nDates, definitions and reading limitsPublication date14 January 2019Event date1968 conception; Modicon 084 development in the late 1960s.Statistical periodHistorical product development, not market statistics.Source locationDate and history paragraphs, web lines 19–28.GeographyUnited States development; French company retrospective.Metric and unitsHistorical dates and reconfiguration mechanism; no cost estimate used.DefinitionProgrammable logic control replacing hard-wired relay logic in the described application.Support and limitsThe world\u0026#x27;s-first wording belongs to the vendor and is not independently established. The source dates conception, not a verified first installation; avoid treating 1968 and 1969 as interchangeable milestones.CalculationNone. Do not use the release\u0026#x27;s later controller instruction-rate comparison as a market or ROI measure.Reading statusRead the cited French original on 6 October 2026.Reuse of original materialPublisher link only; original redistribution permission not established.Read date2026-10-06\r[H003] Yokogawa Celebrates the 50th Anniversary of the CENTUM Distributed Control Systems: A Pioneering AchievementSupported claim: Yokogawa\u0026#x27;s anniversary announcement dates CENTUM\u0026#x27;s announcement to June 1975 and describes early use of microprocessors and a CRT interface in this distributed control system.\nYokogawa Electric Corporation · Primary company anniversary announcement\nRead the original source\nDates, definitions and reading limitsPublication date19 June 2025Event dateCENTUM announced in June 1975.Statistical periodHistorical product launch, not market statistics.Source locationAnnouncement date and first two body paragraphs, web lines 584–588.GeographyJapan; process-industry applications internationally.Metric and unitsLaunch month and technology description; no market-share or availability percentage used.DefinitionCENTUM DCS for plant and process monitoring and control.Support and limitsThe world\u0026#x27;s-first claim is Yokogawa\u0026#x27;s own retrospective wording. Use a representative early DCS milestone without asserting independent priority or a precisely measured productivity effect.CalculationNone. The impact on today\u0026#x27;s process-control supplier structure is author analysis.Reading statusRead the cited original announcement on 6 October 2026.Reuse of original materialPublisher link only; original redistribution permission not established.Read date2026-10-06\r[H004] A Quick History of Creo at PTC: From Parametric to the Cloud and AISupported claim: PTC\u0026#x27;s dated history records the 1988 launch of Pro/ENGINEER and the 1998 introduction of Windchill. It describes parameter- and constraint-based modelling, associative updates and shared web-based lifecycle collaboration.\nPTC · Primary company product-history retrospective\nRead the original source\nDates, definitions and reading limitsPublication date16 February 2026Event datePro/ENGINEER in 1988; Windchill in 1998.Statistical periodHistorical product launches, not market statistics.Source locationDate; Origins and Scaling the digital thread sections, web lines 5–18.GeographyUnited States-based supplier; global engineering software.Metric and unitsLaunch years and product mechanisms; no market share used.DefinitionParametric 3D CAD and web-based product lifecycle management; distinct product functions.Support and limitsVendor claims about first commercial success and first internet-based PLM are not independently verified. These dates are PTC product milestones, not the beginning of all CAD or PLM.CalculationNone. Effects on switching costs and competitive boundaries require explicit author-analysis wording.Reading statusRead the cited dated original on 6 October 2026; published before the cutoff.Reuse of original materialPublisher link only; original redistribution permission not established.Read date2026-10-06\r[H006] The ISA-95 Enterprise-Control System Integration standardsSupported claim: ISA\u0026#x27;s committee-participant retrospective dates the first publication of ISA-95 to 2000. Parts 1 and 2 defined enterprise–control data exchanges and a model framework; later Parts 3 and 4 formalized manufacturing operations management at level 3.\nInternational Society of Automation; Chris Monchinski · Primary standards-body retrospective by a committee participant\nRead the original source\nDates, definitions and reading limitsPublication dateSeptember/October 2020 InTech special edition.Event dateFirst ISA-95 publication in 2000; later multipart development.Statistical periodStandard history, not software market statistics.Source locationWeb lines 38–51; edition identification at line 61.GeographyUS-based standards organization; international manufacturing use.Metric and unitsPublication year, standard parts and logical levels; no market metric.DefinitionEnterprise–control integration and manufacturing operations management. Levels describe logical boundaries, not compulsory physical network topology.Support and limitsDoes not establish when MES was invented or prove universal adoption. Formalising level 3 does not mean manufacturing software began in 2000; no licensed standard text was accessed or reproduced.CalculationNone. Integration-cost implications are objectives or analysis, not measured savings.Reading statusRead the cited public retrospective on 6 October 2026; full standards not read.Reuse of original materialPublisher link only; no standard or publication reproduced.Read date2026-10-06\r[H005] OPC Foundation and PLCopen release version 1.02 of the OPC UA for IEC61131-3 specificationSupported claim: The OPC Foundation identifies its standards work as dating from 1996. Its joint PLCopen group began in 2008; the controller client specification was first released in 2014, enabling data exchanges between controllers and with MES/ERP systems.\nOPC Foundation and PLCopen · Primary standards-body release with historical context\nRead the original source\nDates, definitions and reading limitsPublication dateWeb date 24 November 2020; release dateline 25 November 2020.Event date1996 standards activity; 2008 working group; 2014 client specification; v1.02 in November 2020.Statistical periodSpecification milestones, not installed-base statistics.Source locationDate/dateline, lines 86–90; group and client history, lines 97–102; foundation background, line 110.GeographyInternational standards ecosystem; US announcement and PLCopen collaboration.Metric and unitsYears and specification versions; no adoption share used.DefinitionOPC UA exposes PLC information and supports horizontal and vertical information exchange.Support and limitsSpecification functionality does not guarantee plug-and-play integration or deployment coverage. This source does not establish a single OPC UA launch date; specification development proceeded through multiple stages.CalculationNone. Keep both publication dates instead of silently resolving the one-day discrepancy.Reading statusRead the cited original on 6 October 2026; no paid specification accessed.Reuse of original materialPublisher link only; original redistribution permission not established.Read date2026-10-06\r[H007] Recommendations for implementing the strategic initiative INDUSTRIE 4.0: Final report of the Industrie 4.0 Working GroupSupported claim: The April 2013 working-group report organized Industrie 4.0 around horizontal integration, engineering integration across the value chain, and vertical integration of networked manufacturing. It presented an implementation and research agenda rather than completed deployment results.\nForschungsunion / acatech; Henning Kagermann, Wolfgang Wahlster and Johannes Helbig · Primary policy and engineering working-group report\nRead the original source\nDates, definitions and reading limitsPublication dateApril 2013 in PDF; publisher catalogue dates publication to 8 April 2013.Event dateWorking-group final recommendations in April 2013; initiative launched in early 2011.Statistical periodPolicy and research roadmap, not market statistics.Source locationPDF cover and imprint, file pages 1–2; Executive summary printed p.7/file p.8, lines 154–183; initiative history printed p.76/file p.77, lines 3411–3417. Publisher catalogue lines 7–12.GeographyGermany; international manufacturing-policy context.Metric and unitsPublication and initiative dates; three integration directions, no adoption metric.DefinitionCyber-physical manufacturing and networked integration strategy.Support and limitsRecommendations and anticipated benefits do not establish realised factory autonomy or productivity. The URL upload path is 2018; the document and catalogue establish its original 2013 publication.CalculationNone. Its effect on today\u0026#x27;s platform competition is author interpretation.Reading statusRead cited PDF sections and the German publisher catalogue on 6 October 2026.Reuse of original materialPublisher links only; PDF copyright is reserved.Read date2026-10-06\r[H008] PTC Continues to Accelerate Subscription Business Model GloballySupported claim: PTC announced on 17 January 2018 that new core-software and ThingWorx licences would generally become subscription-only globally on 1 January 2019. The Americas and Western Europe had transitioned on 1 January 2018; regional and product exceptions remained.\nPTC · Primary company licensing-policy announcement\nRead the original source\nDates, definitions and reading limitsPublication date17 January 2018Event dateRegional transition 1 January 2018; announced broader transition 1 January 2019.Statistical periodLicensing policy; no historical bookings statistic used.Source locationWeb lines 6–14, especially transition, retained perpetual rights and exceptions.GeographyAmericas and Western Europe; wider rollout excluded China, India, Korea, Russia, Taiwan and Turkey from a complete transition in this announcement.Metric and unitsPolicy effective dates; no price or revenue estimate.DefinitionNew licence sales by subscription; existing perpetual licences remained usable and active support renewable. Kepware retained both licence forms.Support and limitsAn announced policy is not proof of every later implementation. Subscription is not equivalent to SaaS or mandatory cloud deployment. This company case does not establish a universal industry transition.CalculationNone; no inference about customers\u0026#x27; lifetime costs or vendor margins.Reading statusRead the cited original announcement on 6 October 2026.Reuse of original materialPublisher link only; original redistribution permission not established.Read date2026-10-06\r[H009] Siemens and Microsoft partner to drive cross-industry AI adoptionSupported claim: On 31 October 2023, Siemens and Microsoft introduced Siemens Industrial Copilot, a jointly developed generative-AI assistant for manufacturing. The announcement describes generating, optimising and debugging automation code using industrial information and Azure OpenAI Service.\nSiemens AG; Microsoft is the named partner. · Primary partnership and product announcement\nRead the original source\nDates, definitions and reading limitsPublication date31 October 2023Event dateIndustrial Copilot introduction on 31 October 2023.Statistical periodAnnouncement event, not market statistics.Source locationDate and introduction, web lines 48–60; described function and technology, line 66.GeographyGerman–US partnership; manufacturing applications internationally.Metric and unitsIntroduction date and described functions; no productivity number used.DefinitionAn engineering assistant combining industrial domain information and generative AI; not autonomous control of all factories.Support and limitsSupports the announcement and proposed functions. Claimed reductions from weeks to minutes are not independent performance evidence; no industry-first, broad adoption or ROI claim is established.CalculationNone. Convergence between control expertise and cloud AI is author interpretation.Reading statusRead the cited original release on 6 October 2026.Reuse of original materialPublisher link only; original redistribution permission not established.Read date2026-10-06\r[E004] Industrial automationSupported claim: Texas Instruments\u0026#x27; industrial automation application page identifies analog and embedded products for industrial communication, motor control, power conversion and sensing, including field transmitters, HMI, PLC/DCS/PAC and servo or stepper drives.\nTexas Instruments · Primary vendor application and product description\nRead the original source\nDates, definitions and reading limitsPublication datePublication and update dates not disclosed.Event dateNot applicable: product-role description.Statistical periodNot applicable: no market statistics.Source locationOverview and Featured applications, web lines 0–30.GeographyGlobal application description; regional shipment and supply patterns not disclosed.Metric and unitsProduct functions and categories; no performance or sales figures used.DefinitionEmbedded processors and analog components used inside automation equipment.Support and limitsSupports a representative semiconductor role. It does not establish a supply contract with a named controller company, market share, import dependence or vendor performance superiority. The undated page cannot establish when a feature became available in an earlier version.CalculationNone.Reading statusCited original sections reread on 6 October 2026; earlier verification on 5 October 2026.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-06\r[E006] Distributed Control Systems for Critical Process OperationsSupported claim: SUPCON describes a DCS architecture joining field instruments, I/O, operator and engineering stations, historians, asset management and safety systems for continuous and batch operations; ECS-700 is a disclosed product example.\nSUPCON · Primary vendor DCS product description\nRead the original source\nDates, definitions and reading limitsPublication datePublication and update dates not disclosed; copyright year is not a release date.Event dateNot applicable: product-role description.Statistical periodNot applicable: no market statistics.Source locationReliable Control, lines 23–25; ECS-700, lines 31–43; Key Capabilities, lines 88–110; Applications, lines 112–177.GeographyGlobal product site. Availability of particular models can be region-specific.Metric and unitsControl architecture, interface and application categories.DefinitionProcess-automation distributed control systems for continuous and batch plants.Support and limitsSupports the DCS role and disclosed architecture. Protocol support depends on configuration. It does not verify a named supply contract, zero downtime or return on investment. Inconsistent FAQ entries are not used.CalculationNone.Reading statusRead the cited sections of the publisher\u0026#x27;s original on 5 October 2026. This is not a claim that the entire source was read.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-05\r[E020] Dassault Systèmes’ Software Portfolio: Unified by the 3DEXPERIENCE PlatformSupported claim: Dassault Systèmes identifies CATIA as 3D CAD for design and engineering, SIMULIA as simulation for structural, fluid and electromagnetic virtual testing, ENOVIA as collaborative PLM, and DELMIA as manufacturing and supply-chain planning, management, optimization and execution software.\nDassault Systèmes · Primary vendor software-portfolio description\nRead the original source\nDates, definitions and reading limitsPublication datePublication and update dates not disclosed.Event dateNot applicable: product-role description.Statistical periodNot applicable: no market statistics.Source locationPortfolio section: CATIA, lines 15–17; SIMULIA, lines 25–27; ENOVIA, lines 28–30; DELMIA, lines 35–37; industry applications, lines 83–134.GeographyGlobal portfolio description; regional sales and legal headquarters are not established by this page alone.Metric and unitsProduct and function categories; no market or performance ranking used.DefinitionCAD, simulation/CAE, PLM and manufacturing-operations tools are separate segments; the whole portfolio is not a single segment\u0026#x27;s denominator.Support and limitsSupports multiple engineering-software roles from one source. It does not prove world leadership, complete interoperability, replacement of controllers or quantified customer improvements.CalculationNone.Reading statusRead the cited sections of the publisher\u0026#x27;s original on 5 October 2026. This is not a claim that the entire source was read.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-05\r[E005] Updated Environmental Statement 2022Supported claim: SICK\u0026#x27;s 2022 environmental statement identifies sensors, camera systems, encoders and distance-measurement systems for factory production, packaging, assembly, quality assurance and machine safety, and instruments and measurement systems for process automation.\nSICK · Primary company environmental statement, business-description section\nRead the original source\nDates, definitions and reading limitsPublication date2022 document version; exact public-release date not disclosed.Event dateNot applicable: business-role description.Statistical period2022 version; business descriptions only, not environmental statistics.Source locationTHE 3 BUSINESS FIELDS, printed page 05 / file page 5 (zero-based P4), lines 126–149.GeographySICK\u0026#x27;s factory, process and logistics applications; regional sales breakdown not disclosed.Metric and unitsProduct and application categories; no numerical market metric.DefinitionSensors, cameras, encoders, distance measurement and process instrumentation.Support and limitsSupports the disclosed product roles. It does not show that every 2022 product remains unchanged in 2026, establish market share or verify a current named customer contract.CalculationNone.Reading statusRead the cited sections of the publisher\u0026#x27;s original on 5 October 2026. This is not a claim that the entire source was read.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-05\r[E021] A complete industrial automation portfolioSupported claim: Inovance\u0026#x27;s European product directory lists AC drives, servo drives and motors, PLCs and HMI, motion controllers and I/O, CNC and industrial robots in its automation portfolio.\nInovance Technology Europe GmbH · Primary vendor product directory\nRead the original source\nDates, definitions and reading limitsPublication datePublication and update dates not disclosed; copyright 2026 is not a release date.Event dateNot applicable: product classification.Statistical periodNot applicable: no market statistics.Source locationPortfolio overview, lines 45–47; category headings, lines 49–65; publisher copyright, lines 68–70.GeographyEuropean product directory; Chinese group headquarters separately supported by E023.Metric and unitsProduct categories, not regional sales or localization rates.DefinitionAutomation drives, motors, controls, interfaces and motion equipment; the company spans several value-chain roles.Support and limitsSupports the brand\u0026#x27;s representative drive and control roles at category level only. The H5U functions are separately verified in E024. Category links are not supply contracts or evidence of regional sales, localization rates or market share.CalculationNone.Reading statusRead the cited sections of the publisher\u0026#x27;s original on 5 October 2026. This is not a claim that the entire source was read.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-05\r[E023] About InovanceSupported claim: Inovance\u0026#x27;s official company page states that the group is headquartered in Shenzhen, China, supplies automation solutions to OEMs and end users, and uses industrial automation in its own manufacturing facilities.\nInovance Technology Europe GmbH · Primary company description; benefits and superiority statements are vendor claims\nRead the original source\nDates, definitions and reading limitsPublication datePublication and update dates not disclosed; copyright 2026 is not a release date.Event dateNot applicable: company and business roles. Exact factory upgrade dates not disclosed.Statistical periodNot applicable: no dated R\u0026amp;D statistics from the page are used.Source locationAbout Inovance, lines 45–47; World-class manufacturing, lines 68–70; A global organisation, lines 73–75.GeographyGroup headquarters in Shenzhen, China; global OEM and end-user customers. The described factory lines are not individually located.Metric and unitsHeadquarters, customer types and manufacturing roles; no improvement figures used.DefinitionAutomation supplier and user of automation in its own production; not a measure of regional sales or domestic component content.Support and limitsSupports headquarters and disclosed roles. It does not independently verify efficiency leadership, every product certification or specific factory commissioning. Headquarters do not determine where every component is manufactured.CalculationNone.Reading statusRead the cited sections of the publisher\u0026#x27;s original on 5 October 2026. This is not a claim that the entire source was read.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-05\r[E019] Debut at Hannover Messe 2024: Siemens announces a new generation of controller with Simatic S7-1200 G2, part of Siemens XceleratorSupported claim: Siemens announced the SIMATIC S7-1200 G2 controller generation on 16 April 2024, describing TIA Portal engineering, integrated motion functions and control of coordinated axes and simple kinematics.\nSiemens Digital Industries · Primary product-announcement release\nRead the original source\nDates, definitions and reading limitsPublication date16 April 2024Event dateProduct announced on 16 April 2024; availability in winter 2024 was a plan in that release, not proof of delivery.Statistical periodNot applicable: product announcement.Source locationDate and title; body lines 49–69; New features, lines 64–69.GeographyGermany release describing machine-builder applications; regional launch timing not separately verified.Metric and unitsProduct-role and engineering functions; performance comparisons not used.DefinitionPLC and machine motion control, distinct from manufacturing execution or product lifecycle software.Support and limitsSupports SIMATIC\u0026#x27;s PLC role and the functions disclosed then. Version-sensitive NFC or firmware details are not used as current 2026 facts, and the release does not independently prove productivity gains.CalculationNone.Reading statusRead the cited sections of the publisher\u0026#x27;s original on 5 October 2026. This is not a claim that the entire source was read.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-05\r[E024] H5U PLCSupported claim: Inovance\u0026#x27;s H5U page describes a compact EtherCAT-enabled industrial PLC with axis control, simulation for offline debugging and CANlink, CANopen and Modbus RTU communications.\nInovance Technology Europe GmbH / Inovance India · Primary product description\nRead the original source\nDates, definitions and reading limitsPublication datePublication and update dates not disclosed; copyright 2023 is not a release date.Event dateNot applicable: product functions; original launch date not disclosed.Statistical periodNot applicable: no market statistics.Source locationProduct title and function list, web lines 45–54.GeographyInovance product described on its India site; Chinese group headquarters is separately supported by E023.Metric and unitsFunctions only; I/O counts and performance comparisons not used.DefinitionIndustrial machine PLC and axis control, not the whole industrial automation or software market.Support and limitsSupports the specific PLC and its disclosed interfaces and offline simulation. It does not show that simulation removes all site validation, that all firmware remains unchanged or that the product is supplied in every region.CalculationNone.Reading statusRead the cited sections of the publisher\u0026#x27;s original on 5 October 2026. This is not a claim that the entire source was read.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-05\r[E012] Siemens Industrial Copilot expanded, adopted by thyssenkruppSupported claim: Siemens disclosed in November 2024 that thyssenkrupp Automation Engineering had integrated Engineering Copilot into an electric-vehicle battery inspection machine, using TIA Portal, PLC SCL code and WinCC Unified visualization. Siemens\u0026#x27; Erlangen electronics factory used Operations Copilot on soldering machines for error explanations and maintenance information.\nSiemens AG · Primary vendor announcement describing named applications\nRead the original source\nDates, definitions and reading limitsPublication date12 November 2024Event dateExisting integrations were described on the release date; exact implementation dates not disclosed. A global rollout from 2025 was a plan at that time.Statistical periodNot applicable: named application examples.Source locationRelease date, web line 48; thyssenkrupp case, lines 63–66; Erlangen electronics-factory example, line 68; commercial availability, line 80.GeographyGermany-based equipment builder and German factory; planned worldwide rollout is not verified completion.Metric and unitsApplication scope; no productivity percentage, developer forecast or industry ranking used.DefinitionEngineering and operations assistants linked to automation tools and machine documentation, not autonomous process control.Support and limitsSupports vendor-disclosed applications and downstream roles. It does not independently prove productivity gains, safety certification or completion of the planned worldwide rollout.CalculationNone.Reading statusCited original sections reread on 6 October 2026; earlier verification on 5 October 2026.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-06\r[E104] Rockwell Automation Form 10-Q for quarter ended 30 June 2026Supported claim: Intelligent Devices product revenue is predominantly recognized at a point in time; Software \u0026amp; Control combines product and software revenue; Lifecycle Services mostly recognizes solutions/services over time. Products use distributors/direct sales; large systems/services mainly direct sales. Remaining performance obligations were about USD 1.375bn; about USD 820m expected within 12 months.\nRockwell Automation, Inc. · Primary quarterly regulatory filing.\nRead the original source\nDates, definitions and reading limitsPublication date4 August 2026, official IR 10-Q link dated 8/4/2026.Event dateQuarter ended 30 June 2026.Statistical periodQ3 FY2026 and position at 30 June 2026.Source locationPrinted p 12/PDF file page 12 (zero-based P 11), Note 2 Revenue Recognition, lines 466–486; profit exclusions printed p 25/P24, lines 1044–1054; margin table printed p 31/P30, lines 1269–1271.GeographyGlobal Rockwell group/segments.Metric and unitsRevenue-recognition/channel descriptions; USD remaining performance obligations; company-defined segment operating margin.DefinitionRPO includes existing contractual obligations subject to disclosed expedients; ARR is separately defined in E102.Support and limitsRPO excludes short contracts, specified invoice-based services and unexercised renewals. Segment results exclude corporate/other and acquisition-intangible amortization; not pure-software net profit.CalculationNo calculation.Reading statusSpecified official PDF sections actually read 6 October 2026; not a full 50-page reading.Reuse of original materialPublisher link only; original redistribution rights not established.Read date2026-10-06\r[E207] CIMdata Publishes Executive PLM Market ReportSupported claim: CIMdata estimates the broad 2025 global PLM economy at USD 88.3bn, +9.9%; EDA/AEC growth was notable.\nCIMdata, Inc.; researcher-signed release hosted by Industrial Machinery Digest. · Third-party estimate in the researcher’s public release.\nRead the original source\nDates, definitions and reading limitsPublication date4 June 2026.Event dateMarket-report announcement, 4 June 2026.Statistical periodCalendar 2025.Source locationCIMdata byline/date line 18; dateline 21; total/growth 25; scope 39; regional modules 40.GeographyGlobal; regional modules use Americas, EMEA and Asia-Pacific. EMEA is not equivalent to Europe.Metric and unitsPLM-related software and services revenue, US$ billions.DefinitionBroad PLM: Tools, Product Innovation Platform and Digital Manufacturing; Tools include MCAD, CAM, S\u0026amp;A, EDA and AEC.Support and limitsBroad PLM includes tools/platforms/digital manufacturing; not total industrial software. Paid methodology and chart numbers not read. CIMdata domain returned 403; signed public-release text on IMD actually read.CalculationDirect published values; rounded totals are not used to reconstruct reported growth.Reading statusSigned public release re-read 6 October 2026 on IMD; CIMdata-hosted page inaccessible; paid report/segment graphics not read.Reuse of original materialPublisher link only; original redistribution rights not established.Read date2026-10-06\r[E208] Software Industry Performance in 2025 (translated title)Supported claim: MIIT reports China’s 2025 industrial-software product revenue of RMB 333.0 billion and reported growth of 9.7%.\nMinistry of Industry and Information Technology, Operation Monitoring and Coordination Bureau. · Government statistical release.\nRead the original source\nDates, definitions and reading limitsPublication date30 January 2026, 15:16; webpage does not state timezone.Event dateNot applicable: annual statistics.Statistical periodCalendar 2025.Source locationSection II, first paragraph, web line 13; publication timestamp, line 3.GeographyChina.Metric and unitsIndustrial-software product revenue; RMB 100 million in original.DefinitionMIIT industrial-software product-revenue statistical category; page does not detail product/enterprise/service boundaries.Support and limitsUse the official amount and growth as published. The two released annual amounts do not establish a reconciled growth series or vendor shares.Calculation3330×RMB 100m=RMB 333.0bn. (3330/2940−1)×100≈13.2653%, not the official 2025 +9.7%; reason for the discrepancy not established.Reading statusRelevant original public sections actually re-read on 6 October 2026; locator specifies scope.Reuse of original materialPublisher link only; original redistribution rights not established.Read date2026-10-06\r[E201] Fourteen consecutive wins! SUPCON’s 2024 DCS market share rises to 40.4%, setting another industry record (translated title)Supported claim: MIR estimates China’s 2024 DCS market at approximately RMB 11.76bn, down 3.6%, and SUPCON supplier share at 40.4%.\nMIR Industry/MIR DATABANK; author-approved Eefocus republication. · Researcher-authored third-party estimate; public excerpt.\nRead the original source\nDates, definitions and reading limitsPublication date23 July 2025.Event dateNot applicable: annual estimate.Statistical periodCalendar 2024.Source locationDate/author line 81; amount/scope/share line 91; decline line 94; sector discussion lines 103–105; supplier statement line 122; republication notice line 153.GeographyChina; territorial inclusions not further stated.Metric and unitsRMB market value; supplier share percentage; historical share measurement not fully established.DefinitionChina DCS, including hardware, software, services and engineering.Support and limitsUse attributed 40.4% in prose. E202 does not verify its 2024 historical methodology; no full-vendor share chart or share-derived supplier revenue.Calculation117.6×RMB 100m=RMB 11.76bn; 40.4% quoted without further calculation.Reading statusRelevant original public sections actually re-read on 6 October 2026; locator specifies scope.Reuse of original materialPublisher link only; original redistribution rights not established.Read date2026-10-06\r[E209] China’s Software Industry Performed Well in 2024 (translated title)Supported claim: MIIT reported 2024 industrial-software product revenue of RMB 294.0 billion and growth of 7.4%.\nMinistry of Industry and Information Technology, Operation Monitoring and Coordination Bureau. · Government statistical release.\nRead the original source\nDates, definitions and reading limitsPublication date26 January 2025, 15:21; timezone not stated.Event dateNot applicable.Statistical periodCalendar 2024.Source locationSection II, first paragraph, web line 29; publication timestamp, line 4.GeographyChina.Metric and unitsIndustrial-software product revenue; RMB 100 million in original.DefinitionMIIT industrial-software product-revenue statistical category; page does not detail product/enterprise/service boundaries.Support and limitsUse the official amount and growth as published. The two released annual amounts do not establish a reconciled growth series or vendor shares.Calculation2940×RMB 100m=RMB 294.0bn. (3330/2940−1)×100≈13.2653%, not the official 2025 +9.7%; reason for the discrepancy not established.Reading statusRelevant original public sections actually re-read on 6 October 2026; locator specifies scope.Reuse of original materialPublisher link only; original redistribution rights not established.Read date2026-10-06\r[E202] Automation Product Data Updates: Content and Timing (translated title)Supported claim: MIR lists DCS supplier data and annual vendor totals under sales value.\nMIR DATABANK / MIR Industry. · Publisher’s methodology document.\nRead the original source\nDates, definitions and reading limitsPublication dateUndated; historical version not established.Event dateNot applicable.Statistical periodNot year-specific.Source locationPDF file page 2, zero-based P 1, DCS rows at bottom, extracted lines 109–113.GeographyChinese automation-market dataset.Metric and unitsSales value.DefinitionDCS supplier and industry data.Support and limitsUndated method only. Prior actual reading on 5 October establishes availability in the previous evidence record, not a publication date or a verified 2024 version. No new dated statistics used.CalculationNone.Reading statusDCS table actually re-read 6 October 2026; publisher text extraction, not a claim of reading a paid report.Reuse of original materialPublisher link only; original redistribution rights not established.Read date2026-10-06\r[E204] Five Million Robots now Operate in Factories GloballySupported claim: IFR reports 2025 China installations of 354,000, +20%, representing 59% of global deployments. Chinese suppliers installed 195,000, +15%, with domestic share 55% versus 57% in 2024. Global installations exceeded 600,000, +11%.\nInternational Federation of Robotics. · Association statistics; official public release.\nRead the original source\nDates, definitions and reading limitsPublication date24 September 2026.Event dateWorld Robotics 2026 release, 24 September 2026.Statistical periodCalendar 2025.Source locationDateline/global totals line 9; China share/units/growth lines 14–15; US global-summary figure line 23.GeographyGlobal and China; supplier share uses China’s denominator.Metric and unitsAnnual industrial-robot installations, units; supplier-origin share, percent.DefinitionAnnual factory industrial-robot installations; 55% supplier-origin category share within China, not individual-vendor or revenue share.Support and limitsRounded public values; detailed origin classification/full report not read. E205 reports US 38,400 versus global-summary almost 38,500; not reconciled.CalculationOther supplier origins=100%−55%=45%; share change 55%−57%=−2 percentage points. No unrounded values inferred.Reading statusRelevant original public sections actually re-read on 6 October 2026; locator specifies scope.Reuse of original materialPublisher link only; original redistribution rights not established.Read date2026-10-06\r[E205] U.S. now Second-Largest Robotics Market, Following ChinaSupported claim: IFR reports US 2025 installations 38,400, +12%; automotive 13,500, -1%; metal/machinery 3,000, -15%; food/beverage 2,900, +30%. The US has many robot system integrators; most robots are imported from Japan/Europe.\nInternational Federation of Robotics. · Association statistics; official public release.\nRead the original source\nDates, definitions and reading limitsPublication date24 September 2026.Event dateWorld Robotics 2026 release, 24 September 2026.Statistical periodCalendar 2025.Source locationPDF file page 1, zero-based P 0: dateline/totals lines 15–18; industries 24–28; integrators/imports 34–38.GeographyUnited States.Metric and unitsAnnual industrial-robot installations, units.DefinitionAll U.S. industrial-robot installations; automotive is a subset.Support and limitsUS scope. Global-summary almost 38,500 differs from US release 38,400; retain the latter with attribution; underlying difference not established.CalculationDisclosed values only; industry subsets are not added to total.Reading statusRelevant original public sections actually re-read on 6 October 2026; locator specifies scope.Reuse of original materialPublisher link only; original redistribution rights not established.Read date2026-10-06\r[E206] European Union’s Industrial Robot Stock Hits Record 700,000 UnitsSupported claim: IFR reports EU27 installations 60,500 in 2025, -11%, versus year-end operational stock 712,000. Automotive 14,900, -25%; metal 13,000, -13%; food/beverage 5,300, +4%. Germany installed 24,800 units, representing 41% of EU27 installations.\nInternational Federation of Robotics. · Association statistics; official public release.\nRead the original source\nDates, definitions and reading limitsPublication date24 September 2026, PDF text; filename’s SEP-14 is not the publication date.Event dateWorld Robotics 2026 release, 24 September 2026.Statistical periodCalendar 2025; stock at year-end.Source locationPDF file page 1/P0: date line 15; stock lines 20–23; annual flow 25–29; industries 31–35; Germany 37–40.GeographyEU27, not all Europe.Metric and unitsAnnual installations and year-end operational stock, units; distinct measures.DefinitionIndustrial robots in EU27.Support and limitsEU27, not Europe. Stock and annual additions are distinct; country/industry subsets cannot be summed with aggregate totals.CalculationDirect published rounded totals and percentages.Reading statusRelevant original public sections actually re-read on 6 October 2026; locator specifies scope.Reuse of original materialPublisher link only; original redistribution rights not established.Read date2026-10-06\r[E212] Control’s Top 50 Automation CompaniesSupported claim: The ARC/Control definition includes control, instrumentation and related software but excludes robotics, material handling and supply-chain-management software.\nControl Global; ARC Advisory Group methodology. · Joint research-series methodology.\nRead the original source\nDates, definitions and reading limitsPublication dateNot disclosed on the read methodology page.Event dateNot applicable.Statistical periodMethodology, not a numerical year.Source locationTop 50 methodology and Technologies included/not included, web lines 81–110; explicit exclusions, lines 103–110.GeographyGlobal and North American lists.Metric and unitsAutomation-business revenue; not whole-company revenue.DefinitionARC process-control and automation technology scope.Support and limitsNot the report\u0026#x27;s entire automation/software scope; the Top 50 total is not an industry denominator. The undated method page is used for scope explanation, not year-specific quantitative evidence.CalculationNone; no shares constructed.Reading statusCited original sections reread on 6 October 2026; earlier verification on 5 October 2026.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-06\r[E101] Record third quarter – Outlook raisedSupported claim: Q3 FY2026 Digital Industries orders: EUR 4.9bn, +9% comparable; revenue: EUR 4.9bn, +10% comparable. Software revenue: EUR 1.8bn, +15%. ARR: EUR 5.7bn, +11% organic. Segment profit: EUR 923m, +44%; margin: 18.7% versus 14.5%. Siemens attributes the largest contribution to profit improvements to software.\nSiemens AG · Primary company results release.\nRead the original source\nDates, definitions and reading limitsPublication date6 August 2026, release dateline.Event dateQuarter ended 30 June 2026.Statistical periodQ3 FY2026; three months ended 30 June 2026.Source locationRelease dateline, web line 49; Revenue growth at all industrial businesses, lines 86–87; comparable-growth definition, line 78.GeographyGlobal Siemens Digital Industries segment.Metric and unitsEUR orders, revenue, segment profit; comparable growth; organic ARR; company-defined segment margin.DefinitionDigital Industries segment; software revenue is a subset. Comparable growth excludes currency and portfolio effects.Support and limitsARR is not quarterly revenue. Segment profit definitions differ across companies. Software +15% is quoted without relabelling as organic/comparable.CalculationDirect disclosed figures; no share or precise book-to-bill inferred from rounded EUR 4.9bn amounts.Reading statusRelevant original public sections actually re-read on 6 October 2026; locator specifies scope.Reuse of original materialPublisher link only; original redistribution rights not established.Read date2026-10-06\r[E102] Rockwell Automation Reports Third Quarter 2026 ResultsSupported claim: Q3 FY2026 group sales were USD 2.313bn, +8% reported/+10% organic. Software \u0026amp; Control sales were USD 751m, +19% reported/+18% organic; operating earnings USD 261m; margin 34.8% versus 31.6%. Lifecycle Services sales USD 482m, -12% reported/-2% organic; margin 15.1% versus 13.3%. Organic ARR grew 6%; Sensia dissolution completed 1 April 2026.\nRockwell Automation. · Primary company results release.\nRead the original source\nDates, definitions and reading limitsPublication date4 August 2026, explicit Published line at web line 464.Event dateQuarter ended 30 June 2026; Sensia dissolution 1 April 2026.Statistical periodQ3 FY2026; three months ended 30 June 2026.Source locationFinancial results lines 11–14; segment results lines 34–43; ARR lines 46,59–63; explicit publication line 464.GeographyGlobal Rockwell group and Software \u0026amp; Control segment.Metric and unitsUSD sales and operating earnings; reported/organic growth; company-defined segment margin; organic ARR.DefinitionARR measures annual value of active recurring contracts; excluded from interpretation as recognized revenue, contract liabilities or backlog. Product/revenue-recognition mix is detailed in E104.Support and limitsSoftware \u0026amp; Control margin is a company segment metric. Organic growth adjusts disposals; Sensia and execution affect service profitability. No pure-software margin inferred.Calculation(2313/2144−1)×100≈7.88%; (751/629−1)×100≈19.40%. Use disclosed rounded growth and margins.Reading statusRelevant original public sections actually re-read on 6 October 2026; locator specifies scope.Reuse of original materialPublisher link only; original redistribution rights not established.Read date2026-10-06\r[E105] Schneider Electric Half-Year Financial Report 2026 — issuer text distributed unchangedSupported claim: H1 Industrial Automation revenue was EUR 3.585bn, +7.7% organic; adjusted EBITA EUR 501m, 14.0% margin versus 13.7%. AVEVA revenue grew high-single digits organically; ARR grew 11% at 30 June. Segment regional organic growth: North America 5%, Europe 7%, China/East Asia 12%, South Asia/International 6%. Management cites support-cost leverage while gross margin fell slightly amid raw-material/tariff costs.\nSchneider Electric SE; issuer financial-report text carried unchanged by Publicnow/MarketScreener. · Primary issuer financial report in an explicitly unedited republication; original PDF not read.\nRead the original source\nDates, definitions and reading limitsPublication date30 July 2026; issuer publication/unchanged-distribution statement at line 2005; official index corroborates date (E103).Event dateSix-month period ended 30 June 2026.Statistical periodH1 calendar 2026; ARR at 30 June 2026.Source locationNote 3 segment table lines 1084–1122; business/regional review 1690–1704; adjusted EBITA definition 1730–1733; segment margin/cost discussion 1765–1768; publication/distribution statement 2005.GeographyGlobal Industrial Automation segment; company reporting regions, not individual-country markets.Metric and unitsEUR segment revenue/adjusted EBITA; organic growth; company-defined annualized recurring revenue growth.DefinitionIndustrial Automation includes automation/control for discrete, process and hybrid industries. Adjusted EBITA excludes restructuring/other operating items and acquisition-intangible amortization.Support and limitsCompany regions differ from IFR country/EU27 scopes. H1 profit is not quarterly or pure-software net profit. Read source is unchanged company text hosted externally; not the original PDF.CalculationDisclosed figures; 14.0%−13.7%=0.3 percentage points reported improvement, distinct from disclosed approximately 50 bps organic improvement.Reading statusSpecified unchanged issuer-text sections actually read 6 October 2026; original PDF unavailable and not claimed read.Reuse of original materialSource link only; original report redistribution rights not established.Read date2026-10-06\r[E103] Financial resultsSupported claim: Official results summary reports Q2 2026 Industrial Automation revenue growth of 11% organically; H1 group revenue EUR 21.2bn and adjusted-EBITA margin 19.3% relate to the wider group.\nSchneider Electric. · Primary official investor-relations release index and results summary.\nRead the original source\nDates, definitions and reading limitsPublication date30 July 2026, dated half-year-release index row; ongoing page.Event dateQuarter/half-year ended 30 June 2026.Statistical periodQ2/H1 calendar 2026, distinct periods.Source locationResults-summary lines 16–29; dated release index lines 40–44.GeographyGlobal Schneider Industrial Automation segment; group figures separately labelled.Metric and unitsQ2 Industrial Automation organic growth; separately identified H1 group metrics.DefinitionIndustrial Automation is one segment; group also includes Energy Management.Support and limitsThe index is used for Q2 summary and release timing; Industrial Automation H1 amounts and profitability are established separately in E105. October 29 Q3 results are excluded.CalculationNo calculation.Reading statusRelevant original public sections actually re-read on 6 October 2026; locator specifies scope.Reuse of original materialPublisher link only; original redistribution rights not established.Read date2026-10-06\r[E002] Emerson Completes Acquisition of Remaining Outstanding Shares of AspenTechSupported claim: Emerson completed the acquisition of AspenTech shares it did not already own on 12 March 2025. AspenTech became a wholly owned subsidiary and an independent business unit consolidated in Control Systems \u0026amp; Software.\nEmerson · Primary company transaction-completion announcement\nRead the original source\nDates, definitions and reading limitsPublication date12 March 2025Event dateTransaction completed on 12 March 2025; tender offer expired on 11 March.Statistical periodNot applicable: transaction and post-transaction ownership.Source locationOpening paragraphs, web lines 195–202; Successful Completion of Tender Offer and Merger, lines 206–213.GeographyUnited States; global industrial software business.Metric and unitsOwnership and segment classification. US$265.00 per share applies to the remaining shares acquired.DefinitionCompany ownership and reporting segment, not industry market share.Support and limitsThis was the purchase of the remaining shares, not Emerson\u0026#x27;s first acquisition of control. The release does not establish industry concentration, total historical investment or integration success.CalculationNone.Reading statusRead the cited sections of the publisher\u0026#x27;s original on 5 October 2026. This is not a claim that the entire source was read.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-05\r[E001] Siemens acquires Altair to create most complete AI-powered portfolio of industrial softwareSupported claim: On 26 March 2025, Siemens announced that it had completed the acquisition of Altair Engineering for an enterprise value of approximately US$10 billion, adding mechanical and electromagnetic simulation, high-performance computing, data science and AI capabilities.\nSiemens AG · Primary company transaction-completion announcement\nRead the original source\nDates, definitions and reading limitsPublication date26 March 2025Event dateAcquisition completed on 26 March 2025.Statistical periodNot applicable: transaction event.Source locationRelease date and first body paragraph; web lines 48–60.GeographyGermany-based acquirer and US software target; global business.Metric and unitsApproximately US$10 billion of enterprise value; not annual revenue or market size.DefinitionAcquisition of an industrial simulation and analysis software company; no industry-wide denominator.Support and limitsVerifies the disclosed completion, value and capability scope. Promotional claims about the world\u0026#x27;s most complete portfolio are not independent rankings or proof of integration results.CalculationNo calculation; retain the source\u0026#x27;s approximate US$10 billion enterprise value.Reading statusRead the cited sections of the publisher\u0026#x27;s original on 5 October 2026. This is not a claim that the entire source was read.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-05\r[E017] Milestone for industrial digitalisation: Asset Administration Shell standard receives security specificationSupported claim: On 10 June 2025, IDTA announced an updated AAS specification bundle including Part 4: Security, with fine-grained access control for properties, submodels and registry or repository services.\nIndustrial Digital Twin Association (IDTA) · Primary specification-release announcement\nRead the original source\nDates, definitions and reading limitsPublication date10 June 2025Event dateBundle-release announcement on 10 June 2025; the exact specification-version documents were not separately examined.Statistical periodNot applicable: specification update.Source locationDate, title and release overview, web lines 79–85; access-control description, lines 86–90.GeographyGermany-based association; intended for international industrial ecosystems.Metric and unitsSpecification and authorization functions; no market statistics.DefinitionAsset Administration Shell information models for standardized digital twins, not the entire digital-twin software market.Support and limitsSupports the addition of an access-control mechanism. It does not prove that all products implement it or that the 2025 announcement describes the latest 2026 release.CalculationNone.Reading statusCited original sections reread on 6 October 2026; earlier verification on 5 October 2026.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-06\r[E014] Cyber Resilience Act - Reporting obligationsSupported claim: The European Commission states that manufacturers\u0026#x27; CRA reporting obligations started on 11 September 2026 for actively exploited vulnerabilities and severe security incidents in products with digital elements; an early warning is due within 24 hours of awareness and a full notification within 72 hours.\nEuropean Commission, Shaping Europe\u0026#x27;s digital future · Primary regulator implementation explanation\nRead the original source\nDates, definitions and reading limitsPublication dateOriginal publication date not disclosed; read version last updated 11 September 2026.Event dateReporting obligations apply from 11 September 2026 and were already applicable at the research cutoff.Statistical periodNot applicable: regulatory timetable.Source locationOpening paragraphs, lines 4–7; What are the main rules?, lines 8–13; reporting platform, lines 14–16; Last update field.GeographyEuropean Union market and relevant manufacturers; application to a particular product or operator must be checked.Metric and units24-hour warning and 72-hour notification deadlines measured from awareness; these are not patch-release deadlines.DefinitionSecurity reporting for relevant products with digital elements, not every internal OT installation without exceptions.Support and limitsSupports the reporting phase and deadlines. It does not make all CRA product obligations applicable in 2026. Open-source software steward obligations have a different timetable. Full statutory scope and exclusions were not independently resolved.CalculationNo calculation; deadlines are quoted from the regulator explanation.Reading statusCited original sections reread on 6 October 2026; earlier verification on 5 October 2026.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-06\r[E022] Cyber Resilience ActSupported claim: The European Commission states that the CRA entered into force on 10 December 2024; its main product obligations apply from 11 December 2027 and reporting obligations from 11 September 2026. The framework covers lifecycle cybersecurity for relevant hardware and software products with digital elements.\nEuropean Commission, Shaping Europe\u0026#x27;s digital future · Primary regulator policy overview\nRead the original source\nDates, definitions and reading limitsPublication dateOriginal release and complete update dates not disclosed; read content includes the 27 July 2026 guidance and the September reporting phase.Event dateEntered into force 10 December 2024; reporting applies from 11 September 2026; main obligations apply from 11 December 2027.Statistical periodNot applicable: phased regulatory timetable.Source locationBody lines 8–13, especially line 12; reporting link and description, lines 75–78.GeographyEuropean Union market; not a domestic Chinese or US law.Metric and unitsDates and lifecycle responsibilities; no cost or fine figures used.DefinitionRelevant products with digital elements. Detailed exceptions and operator roles require statutory scope checks.Support and limitsTogether with E014, supports the separation of entry into force and application phases. The full statutory exclusions and article-level interpretation were not independently verified.CalculationNone.Reading statusCited original sections reread on 6 October 2026; earlier verification on 5 October 2026.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-06\r[E015] Implementation Opinions on the \u0026#x27;Artificial Intelligence + Manufacturing\u0026#x27; Special Action (translated title)Supported claim: China\u0026#x27;s eight-department AI + Manufacturing policy was made public on 7 January 2026 after being dated 25 December 2025. It calls for models suited to industrial real-time, reliability and safety requirements, cloud-edge-device deployment, lighter models, data governance, evaluation and high-quality datasets, with objectives for 2027.\nMinistry of Industry and Information Technology (MIIT), Cyberspace Administration of China, National Development and Reform Commission, Ministry of Education, Ministry of Commerce, SASAC, State Administration for Market Regulation, and National Data Administration · Primary issued policy document; objectives and proposed measures are not completed outcomes\nRead the original source\nDates, definitions and reading limitsPublication date7 January 2026, 16:43 as shown on the publisher page.Event dateDocument dated 25 December 2025; separate statutory commencement date not disclosed.Statistical periodPolicy objectives to 2027, not actual 2027 results.Source locationFile and printed p.1, objectives; p.2, Section II(ii) industry models and II(iii) data; p.3, Section III(v) workflow transformation. Publisher page lines 8–17 distinguish document and release dates.GeographyChina, national policy.Metric and unitsPolicy targets, not current installations, orders or revenue. The targets of 3–5 general models, 1,000 agents, 100 datasets and 500 scenarios must be explicitly labelled as 2027 objectives if used.DefinitionAI applied across manufacturing design, pilot validation, production and operations; not a statistical definition of the industrial software market.Support and limitsSupports the issued policy and stated paths. It does not prove achievement, a subsidy amount or a vendor award. Only the cited opening sections were read, not every annex.CalculationNone.Reading statusRead the cited sections of the publisher\u0026#x27;s original on 5 October 2026. This is not a claim that the entire source was read.Reuse of original materialPublisher link only; no original material redistributed.Read date2026-10-05\rImage Credits The original cover is retained from the reviewed report. The three inline photographs below are existing photographs; their authors, source pages and reuse terms are listed with each image.\nA Ferranti Argus 500 computer system in Edinburgh, photographed in 1980.\nLeocapaldi · Source photograph · Public domain (author\u0026#x27;s release). Web copies are resized, uncropped WebP derivatives; the original file is linked from the photograph.\nA WAGO controller in a pharmaceutical monitoring setup, photographed in 2016.\nPierre75000 · Source photograph · CC BY-SA 4.0 International. Web copies are resized, uncropped WebP derivatives; the original file is linked from the photograph.\nStäubli industrial robots on a manufacturing line, photographed in January 2023.\nClemenspool · Source photograph · CC0 1.0. Web copies are resized, uncropped WebP derivatives; the original file is linked from the photograph.\n","date":"2026-10-06T00:00:00Z","image":"/post/industrial-automation-software-industry-report/cover.png","permalink":"/post/industrial-automation-software-industry-report/","title":"Industrial Automation and Industrial Software: History, Value Chain, Markets, and Challenges"},{"content":"NVIDIA’s founding history, products and software, revenue, customers, manufacturing partners and disclosed financial results.  Founded5 April 1993 Nasdaq tickerNVDA FY2026 revenue$215.938bn Latest quarter in this reportQ2 FY2027 Cover: NVIDIA Voyager headquarters. Official NVIDIA photograph; the logo is part of the original image.\nInformation cutoff 3 October 2026\nThis report is based on company announcements, filings with the US Securities and Exchange Commission (SEC), product documentation and research papers. Retrospective company accounts and financial disclosures are identified through the references.\nUnless otherwise stated, financial amounts in the English edition are in USD billions; earnings per share are in US dollars. NVIDIA’s fiscal year differs from the calendar year: fiscal 2026 ended on 25 January 2026. The latest disclosed quarter is the second quarter of fiscal 2027, which ended on 26 July 2026.[24][25]\n1 Company founding and development Selected milestones A selection of documented events; the company history in the article provides additional context.\nApril 5, 1993 NVIDIA was founded by Jensen Huang, Chris Malachowsky, and Curtis Priem to develop 3D graphics for gaming and multimedia.\nSource: NVIDIA company history 1995 NVIDIA introduced its first product, NV1, which was used in Diamond Edge 3D products.\nSource: NVIDIA early company history 1999 NVIDIA completed its initial public offering on January 22, and GeForce 256 was released on October 11.\nSource: NVIDIA investor FAQ Source: GeForce 256 release retrospective 2006 NVIDIA unveiled CUDA, providing a programming architecture for GPU parallel computing beyond graphics rendering.\nSource: NVIDIA company history 2012 The AlexNet paper by Alex Krizhevsky, Ilya Sutskever, and Geoffrey Hinton described model training on two NVIDIA GTX 580 GPUs with 3 GB of memory each.\nSource: original AlexNet paper, pages 2 and 7 April 5, 2016 NVIDIA announced DGX-1, an integrated deep-learning system with eight Tesla P100 GPUs and NVLink interconnects.\nSource: DGX-1 announcement April 27, 2020 NVIDIA completed its acquisition of Mellanox, with an announced transaction value of US$7 billion.\nSource: acquisition completion announcement March 18, 2024 NVIDIA announced the Blackwell platform, including the GB200 Grace Blackwell Superchip and the GB200 NVL72 rack-scale system.\nSource: Blackwell platform announcement 1 1993 The three founders establish NVIDIA NVIDIA was founded on 5 April 1993 by Jensen Huang, Chris Malachowsky and Curtis Priem. Its initial development work focused on three dimensional graphics and multimedia processing for personal computers.\nThe founders’ professional backgrounds included semiconductors and computer graphics. Huang had worked at AMD and LSI Logic. Malachowsky had held engineering and technical leadership positions at Hewlett-Packard and Sun Microsystems. Priem had designed graphics hardware at Sun Microsystems, including work on its GX graphics products, and subsequently served as NVIDIA’s chief technology officer.[1][2][3][4]\nIn an official retrospective published in 2023, NVIDIA described meetings at a Denny’s restaurant in Silicon Valley where the founders discussed the company’s concept, including enabling more realistic three dimensional graphics on PCs. Huang returned to the restaurant in September 2023 for the unveiling of a commemorative plaque. The restaurant account comes from the company’s retrospective narrative.[5]\n2 1994 to 1995 Manufacturing cooperation and the first product NV1 In 1994, NVIDIA established a partnership with SGS-Thomson. Diamond Multimedia subsequently used NVIDIA chips in its graphics cards.\nNVIDIA introduced its first product, NV1, in 1995. Products incorporating the chip included Diamond Edge 3D. NV1 supported two dimensional and three dimensional graphics and used quadratic texture mapping. NVIDIA also participated in bringing Sega games to the PC platform during this period.[6]\n3 1996 NV1 sales end and NV2 development is discontinued NVIDIA’s IPO prospectus reported that NV1 sales declined as Microsoft Direct3D and SGI OpenGL gained adoption. The company stopped selling NV1 in the first quarter of 1996.\nNVIDIA subsequently discontinued development of NV2 for a game console project and shifted development work to RIVA 128. This period included product discontinuations and changes in development direction.[4]\n4 1997 to 1998 The RIVA product family enters the market RIVA 128 entered commercial sales in 1997. According to NVIDIA’s corporate timeline, shipments exceeded one million units in the first four months. In 1998, the company introduced RIVA 128ZX and RIVA TNT and established a long term manufacturing partnership with TSMC.[6]\nThe early prospectus reported net revenues of USD 1.182 million, USD 3.912 million and USD 29.071 million for 1995, 1996 and 1997, respectively. These figures use the annual periods reported at that time and are presented separately from the more recent fiscal year data below.[4]\n5 1999 The IPO and GeForce 256 launch NVIDIA listed on Nasdaq on 22 January 1999 under the ticker NVDA. Its initial public offering price was USD 12 per share. This was the original issue price, without adjustment for subsequent stock splits.[7]\nOn 11 October that year, NVIDIA introduced GeForce 256. The product integrated graphics transformation and lighting calculations into hardware. NVIDIA used the term GPU in its product marketing.[8]\n6 2001 Supplying chips for a game console In 2001, NVIDIA supplied the graphics processor and media communications processor for Microsoft’s original Xbox. Its products had therefore entered gaming devices beyond personal computers.[9]\n7 2006 CUDA is announced NVIDIA announced CUDA in 2006. CUDA provides a programming environment and tools that allow developers to use GPUs for computing tasks beyond graphics rendering.\nThis product approach includes both hardware and software: GPUs perform computation, while compilers, runtimes and software libraries support program development and execution.[1][10]\n8 2012 The AlexNet paper documents training on NVIDIA GPUs In 2012, Alex Krizhevsky, Ilya Sutskever and Geoffrey Hinton published the image classification research paper commonly known as the AlexNet paper.\nThe paper states that the researchers trained the model on two NVIDIA GTX 580 GPUs, each with 3GB of memory, over approximately five to six days. This is a GPU use case explicitly documented in a research paper. The model’s authors and the chip supplier were separate parties.[11]\n9 2016 DGX 1 is introduced as an integrated system On 5 April 2016, NVIDIA introduced DGX-1, integrating GPUs, interconnects, software and development tools into a deep learning computing system.\nThe original DGX-1 contained eight Tesla P100 GPUs connected through NVLink. NVIDIA supplied an integrated system through this product, rather than only an individual processor.[12][13]\n10 2018 The Turing architecture and RTX products are introduced In August 2018, NVIDIA introduced the Turing architecture and Quadro RTX professional products, followed by GeForce RTX products for gaming.\nTuring included RT cores for ray tracing and Tensor cores for related matrix computations. The corresponding hardware capabilities were incorporated into professional graphics and consumer gaming products.[14][15]\n11 2020 The Mellanox acquisition is completed On 27 April 2020, NVIDIA announced the completion of its acquisition of Mellanox. The announcement described a transaction value of USD 7 billion.\nMellanox’s business included InfiniBand and Ethernet networking products. Following completion, these networking products became part of NVIDIA’s data center product portfolio.[16]\n12 2022 The Arm acquisition agreement is terminated In February 2022, NVIDIA and SoftBank announced the termination of the Arm acquisition agreement, citing regulatory challenges. The transaction did not close, and Arm did not become an NVIDIA subsidiary through this transaction.\nNVIDIA recorded a USD 1.353 billion acquisition termination charge in its fiscal 2023 financial statements.[17][18]\n13 2024 The Blackwell platform is announced On 18 March 2024, NVIDIA announced the Blackwell platform. Related products included the B200 GPU, the GB200 superchip and the GB200 NVL72 system.\nGB200 combines two Blackwell GPUs with one Grace CPU. GB200 NVL72 further organizes these computing components into a rack system.[19]\n14 2026 Vera Rubin production progress is disclosed On 31 May 2026, NVIDIA announced that the Vera Rubin platform was ramping into full production. This records the production status disclosed by the company; anticipated deployment volumes and revenue in the announcement are not presented as completed operating results.[20]\n2 Components of the product portfolio COMPANY PRODUCT MAP\nNVIDIA’s product and software map Product uses and delivery forms. These branches are not accounting segments or verified market-share categories.\nFit to view Expand all Collapse Full screen Read the full text outline NVIDIAHardwareGeForce NVIDIA RTX Data-center GPUs / Grace CPUs DGX / GB200 systems NetworkingSpectrum ConnectX BlueField SoftwareCUDA TensorRT AI Enterprise Omniverse PlatformsDRIVE Isaac Jetson Examples of current product families described by NVIDIA. A branch indicates product use or delivery form; it does not imply a separate financial reporting segment.\nSources: NVIDIA product catalog · CUDA development platform · GB200 NVL72 product documentation · Ethernet product portfolio · TensorRT documentation · AI Enterprise licensing guide · Omniverse documentation · DRIVE platform · Isaac platform · Jetson modules. Reviewed 2026-10-05.\nNVIDIA’s products include processors, computing systems, networking equipment and software. Their functions and delivery formats differ.\nProduct category Examples Uses Consumer graphics GeForce Game rendering and graphics computation on PCs Professional graphics NVIDIA RTX professional products Workstation graphics, design and visualization Data center computing Data center GPUs, Grace CPUs and DGX AI and other computing workloads Networking Spectrum, ConnectX and BlueField Data transfer, network connectivity and related infrastructure processing Automotive computing DRIVE In vehicle computing and autonomous driving system development Embedded computing Jetson Computing in robots and other devices These categories describe product uses and do not correspond directly to the accounting segments in the financial statements.[21]\nThe distinction between training and inference AI model training involves using data to adjust model parameters. Inference uses a trained model to process inputs and generate results.\nNVIDIA’s TensorRT documentation describes a workflow from training models in frameworks such as PyTorch and TensorFlow to optimizing and deploying them for inference. The same GPU product portfolio can therefore participate in different stages of model development, while training and inference remain different computing tasks.[22]\nThe distinction between chips and servers and rack systems From computing components to a rack system Example: the NVIDIA GB200 NVL72 single-rack reference configuration. The levels below describe composition, rather than a sequence of computing tasks.\n1. Computing components Grace CPU and Blackwell GPU\nThe CPU and GPU are separate computing processors used in this configuration.\n2. GB200 Grace Blackwell Superchip 1 Grace CPU + 2 Blackwell GPUs\nNVLink-C2C connects the CPU and GPUs; the Superchip combines multiple processors.\n3. GB200 compute tray 2 GB200 Superchips = 2 Grace CPUs + 4 Blackwell GPUs\nA liquid-cooled compute node based on the MGX reference design.\n4. GB200 NVL72 rack system 18 compute trays + 9 NVLink switch trays\nTotal: 36 Grace CPUs and 72 Blackwell GPUs. NVLink connects the GPUs within this liquid-cooled rack.\nInfiniBand or Ethernet provides network connectivity beyond the rack; this is separate from the rack's NVLink GPU interconnect.\nCounts apply to this single-rack reference configuration, rather than to every NVIDIA server or rack product.\nSources: NVIDIA GB200 NVL72 technical description; GB200 NVL72 product specifications; Blackwell platform announcement. A chip is a computing component. A server assembles processors, memory and other components into a device. A rack system further integrates computing devices, connections, power delivery and cooling.\nFor example, the official GB200 NVL72 documentation specifies 72 Blackwell GPUs and 36 Grace CPUs, liquid cooling and GPU connections through NVLink. Purchasing or deploying this system involves a different product scope from purchasing an individual GPU.[23]\nComponents of the networking portfolio NVIDIA’s Ethernet portfolio includes Spectrum switches, ConnectX network adapters, BlueField data processing units, LinkX connectivity products and associated software.\nThese products handle connectivity, data transfer and related tasks. Computing interconnects between GPUs, network connections between servers and access to external data perform functions at different levels of a system.[26]\n3 Software and development platforms CUDA Program development and execution CUDA includes compilation, runtime, debugging, performance analysis and software library components. Developers can use these tools to write and run GPU computing programs.\nThe CUDA platform and an individual graphics card are different product concepts: CUDA is a software and programming system, while the graphics card is hardware that executes programs.[10]\nTensorRT Model inference optimization TensorRT is a software development toolkit for optimizing and executing deep learning inference. It addresses deployment and execution of models that have already been trained.[27]\nNVIDIA AI Enterprise Enterprise licensing and support NVIDIA AI Enterprise provides enterprise AI software and support services. Licensing options include subscriptions, usage based billing through cloud marketplaces and perpetual licenses with a specified support term.\nThe company’s software business consequently includes charging and service arrangements that differ from one time hardware sales.[28]\nOmniverse Three dimensional data and simulation tools Omniverse provides software components for three dimensional scenes, simulation and related application development. Its technology is built on OpenUSD, an open framework developed by Pixar for describing three dimensional scenes and exchanging and organizing three dimensional data across tools.[29]\nDRIVE and Isaac and Jetson DRIVE provides automotive computing and development tools covering sensor data, model training, simulation and in vehicle computing.\nIsaac provides robotics development tools. Isaac Sim supports simulation, while Isaac ROS provides robotics software components. Jetson is an embedded computing hardware platform with software tools such as JetPack. These platforms address automotive development, robotics software development and device based computing, respectively.[30][31][32]\n4 Revenue sources and business classifications REVENUE COMPOSITION · FY2026\nNVIDIA revenue composition by market platform Share of NVIDIA’s total FY2026 revenue by its disclosed market-platform classification. This is company revenue composition, not industry market share.\nNVIDIA consolidated worldwide revenue · Platform revenue / total company revenue\nFull screen View the data table NVIDIA revenue composition by market platform · FY2026 · Platform revenue / total company revenue Market platformShare Data Center89.7%Gaming7.4%Professional Visualization1.5%Automotive1.1%OEM and Other0.3% Download data (CSV) Percentages are calculated from the five reported platform revenues and the USD 215.938 billion total, then rounded to one decimal place. The exact amounts appear in the table below. Fiscal 2026 ended on 25 January 2026. FY2027 uses a changed platform presentation and is not treated as a continuous category trend.\nSource: NVIDIA / U.S. SEC — Fiscal 2026 annual report Form 10 K (2026-02-25). Reviewed 2026-10-05.\nNVIDIA discloses its business through both accounting segments and market platforms. Their boundaries differ, and amounts cannot be added across the two classifications.\nMarket platform revenue in fiscal 2026 Market platform Revenue USD bn Data Center 193.737 Gaming 16.042 Professional Visualization 3.191 Automotive 2.349 OEM and Other 0.619 Total 215.938 Data Center revenue in this presentation includes computing and networking businesses.[33]\nMarket platform revenue in the second quarter of fiscal 2027 Market platform Revenue USD bn Data Center Hyperscale 48.710 Data Center AI Clouds, Industrial \u0026amp; Enterprise 40.313 Data Center subtotal 89.023 Edge Computing 7.198 Total 96.221 Hyperscale is the hyperscale customer category. ACIE refers to AI Clouds, Industrial \u0026amp; Enterprise. The Data Center subtotal already includes the first two rows.\nNVIDIA changed its presentation of market platform revenue in fiscal 2027 and recast comparative data. During the second quarter, it also reclassified one customer from ACIE to Hyperscale following a change in the customer’s business model. Categories such as Gaming in the earlier table therefore cannot be treated as having the same statistical scope as Edge Computing in the newer table.\nThe company’s reportable accounting segments remain Compute \u0026amp; Networking and Graphics, presented separately from the market platform classifications above.[34]\n5 Customers and sales channels and revenue recognition NVIDIA distinguishes between direct and indirect customers.\nDirect customers purchase from NVIDIA and include distributors, equipment manufacturers, cloud service providers, AI model developers and system integrators. Indirect customers purchase or use the products through direct customers and may include enterprises, cloud service providers and public sector organizations.\nIn the first half of fiscal 2027, three direct customers accounted for 16%, 15% and 13% of total revenue, respectively. The filing did not name them in this disclosure, so these percentages cannot be assigned to specific companies without additional evidence.[35]\nGeographic revenue is classified by the location of the customer’s headquarters. This describes the customer’s location and does not directly identify the eventual installation or use location of the equipment.\nProduct revenue is generally recognized when control transfers. Support and service revenue is recognized according to the service arrangements. Revenue recognition and cash collection can occur on different dates; sales revenue, accounts receivable and cash flow are therefore different financial items.[33]\n6 Manufacturing arrangements and supply chain NVIDIA uses external manufacturing partners. Product design, wafer fabrication, memory supply, packaging and testing, and system assembly involve different participants.\nPartners identified in its annual report include the following:\nStage Partners identified in the annual report Wafer fabrication TSMC and Samsung Memory supply SK hynix, Micron and Samsung Assembly and testing and related manufacturing services Hon Hai, Wistron, Fabrinet and others The annual report also discusses advanced packaging technologies such as CoWoS. This list describes disclosed relationships and does not mean that every product is manufactured by the same group of suppliers.[33]\n7 Operating data for the last five fiscal years GAAP annual results, USD billions. Fiscal years, not calendar years. FY2026 ended on 25 January 2026. Sources: FY2023, FY2025 and FY2026 results. On small screens, scroll the chart horizontally. Select it to open at full size; the exact figures appear in the table below. Amounts below follow the company’s disclosed US generally accepted accounting principles (GAAP) figures.\nFiscal year Revenue USD bn Gross margin Operating income USD bn Net income USD bn Operating cash flow USD bn 2022 26.914 64.9% 10.041 9.752 9.108 2023 26.974 56.9% 4.224 4.368 5.641 2024 60.922 72.7% 32.972 29.760 28.090 2025 130.497 75.0% 81.453 72.880 64.089 2026 215.938 71.1% 130.387 120.067 102.718 The data come from the fiscal 2023, fiscal 2025 and fiscal 2026 financial announcements.[18][36][24]\nThese items use different calculations:\nRevenue: sales and service revenue recognized during the reporting period. Gross margin: revenue less cost of revenue, expressed as a percentage of revenue. Operating income: gross profit less operating expenses such as research and development, sales and administration. Net income: income after other income and expenses and income tax. Operating cash flow: net cash generated by operating activities, including working capital movements and related adjustments. Fiscal 2023 operating income includes the Arm acquisition termination charge described earlier. Fiscal 2026 net income also includes income outside operating income; the two measures are not interchangeable.\n8 Results for the latest disclosed quarter NVIDIA announced its second quarter fiscal 2027 results on 26 August 2026. The reporting period ended on 26 July 2026.\nMetric Q2 fiscal 2027 Prior year quarter Revenue USD bn 96.221 46.743 GAAP gross margin 75.0% 72.4% Operating expenses USD bn 8.408 5.413 Operating income USD bn 63.734 28.440 Net income USD bn 59.688 26.422 Diluted earnings per share USD 2.46 1.08 The company reported revenue growth of 106% year over year and 18% from the preceding quarter.[25]\nDifferences between GAAP and adjusted figures For the same quarter, NVIDIA reported GAAP diluted earnings per share of USD 2.46 and non-GAAP earnings per share of USD 2.22.\nNon-GAAP figures adjust company designated items, including the exclusion of gains from equity securities. Net gains from equity securities were USD 7.771 billion in the quarter, so adjusted earnings per share can be lower than the GAAP figure.\nBeginning in the first quarter of fiscal 2027, NVIDIA’s non-GAAP measures no longer exclude stock based compensation expense. The historical comparative figures presented were updated accordingly. Non-GAAP figures published in different years therefore need to be distinguished according to the adjustment definitions in the relevant announcements.[34]\n9 Cash flow and receivables and the balance sheet Profit and cash flow in the first half of fiscal 2027 Metric Amount USD bn Net income 118.010 Operating cash flow 74.421 Company defined free cash flow 69.895 NVIDIA defines free cash flow as operating cash flow less purchases related to property and equipment and intangible assets, and principal payments on those assets.[25]\nThe quarterly report states that net gains from equity securities were primarily driven by unrealized gains. Recognizing such gains in income does not mean the company received an equivalent amount of cash from selling the investments.[35]\nSelected balance sheet items as of 26 July 2026 Item Amount USD bn Cash and cash equivalents 22.443 Marketable debt securities 34.143 Marketable equity securities 42.783 Accounts receivable net 63.059 Inventories 31.575 Total assets 320.272 Total liabilities 91.288 Shareholders’ equity 228.984 Cash, debt securities and equity securities are presented separately in the financial statements. Marketable equity securities are investment assets.[25]\nThe CFO commentary reported days sales outstanding of 60 days in the second quarter, compared with 45 days in the preceding quarter. The company attributed the change to extended payment terms on large, multi-quarter agreements with certain investment-grade customers.\nDuring the same quarter, NVIDIA issued USD 25 billion of senior unsecured notes for general corporate purposes.[34]\n10 Research spending and employees and management Research and development expenses Fiscal year R and D expenses USD bn 2023 7.339 2024 8.675 2025 12.914 2026 18.497 These amounts are presented in the respective annual financial statements as operating expenses.[18][36][24]\nAt the end of fiscal 2026, NVIDIA had approximately 42,000 employees across 38 countries, including approximately 31,000 working in research and development.[33]\nManagement and the founder’s ownership Management identified in the 2026 proxy statement included President and Chief Executive Officer Jensen Huang, Chief Financial Officer Colette Kress, Ajay Puri with responsibility for worldwide field operations, and Debora Shoquist with responsibility for operations.\nAs of 23 March 2026, the proxy statement reported Huang’s beneficial ownership at 870,604,104 shares, or 3.58%. This measure includes shares associated with relevant trusts, entities and a foundation; it should not all be treated as shares held directly in his personal name.\nExecutive equity compensation includes performance stock units (PSUs) and restricted stock units (RSUs), which have different vesting conditions and assessment arrangements.[37]\n11 Disclosed contractual commitments and guarantees Selected contractual commitments as of 26 July 2026 Category Disclosed amount USD bn Supply and capacity 279 Cloud service agreements 29 Data center leases not commenced 25 Equity investments 25 Capital expenditures 8 Total of the above items 366 These figures are future contractual commitments disclosed in the financial statements and cover different periods. Certain arrangements are conditional. They are distinct from cash already paid and expenses already recognized during the reporting period.[35]\nSB Energy related guarantees signed in August 2026 NVIDIA subsequently disclosed guarantees signed in August 2026 in connection with SB Energy, providing credit support for the PORTS campus project in Ohio and involving leases for an OpenAI affiliate.\nThe guarantees are capped at USD 105 billion in total and become effective in phases subject to conditions such as the commencement of the relevant leases. Payment obligations are triggered by specified tenant defaults and are limited to defined portions of lease and power payments.\nThe amount is a conditional guarantee cap, not cash already paid or a loss already recognized. The signing date was after the quarterly balance sheet date of 26 July.[35]\n12 Export licensing events that have occurred On 9 April 2025, the US government informed NVIDIA that exports of H20 products to China and related destinations required a license.\nIn the first quarter of fiscal 2026, which ended on 27 April 2025, NVIDIA recorded a USD 4.5 billion charge related to H20 excess inventory and purchase obligations. GAAP gross margin for that quarter was 60.5%.\nThese are events that had occurred and were reflected in the financial statements. Undelivered orders, estimates of future sales and recognized revenue use different reporting bases.[38]\n13 Related computing products offered by other companies Other companies offer announced processors and software systems for AI computing:\nCompany Product or platform Uses disclosed by the company AMD Instinct accelerators and ROCm software AI training, inference and other computation Google Cloud TPU Cloud based machine learning computation AWS Trainium AI training and inference These products differ in hardware format, software interfaces and delivery arrangements. The table records their existence and uses; it does not establish market shares or quantify an effect on NVIDIA’s revenue.[39][40][41]\n14 Stock trading data and measurement basis NVIDIA common stock trades on Nasdaq under the ticker NVDA. The historical price page records a closing price of USD 233.95 per share for the US trading session on 2 October 2026.[42]\nThe company reported GAAP diluted earnings per share of USD 4.90 for fiscal 2026, USD 1.84 for the first half of fiscal 2026 and USD 4.85 for the first half of fiscal 2027. Combining these disclosed rounded figures gives approximate trailing twelve month earnings per share of:\n4.90 − 1.84 + 4.85 = USD 7.91.\nDividing that closing price by these earnings per share gives approximately 29.6 times. This calculation uses historical GAAP earnings, including investment gains in the corresponding periods. It is an approximate historical price to earnings ratio calculated from disclosed data.[24][34]\nMeasurement conventions used in this report Historical events are presented by event date, financial results by company fiscal period, balance sheet amounts by reporting date and stock prices by US trading session. Accounting segments, market platforms, customer headquarters and final product uses each have their own definitions. Future revenue guidance, anticipated deployment volumes and investment return forecasts are not included as realized results.\nSources and image credit Corporate timeline. Return to citation Jensen Huang biography. Return to citation Chris Malachowsky biography. Return to citation IPO prospectus. Return to citation Retrospective account of the Denny’s founding story. Return to citation Early product and manufacturing partnership timeline. Return to citation Investor relations FAQs. Return to citation GeForce 256 launch retrospective. Return to citation Corporate timeline for 2001. Return to citation CUDA development platform. Return to citation Original AlexNet research paper. Return to citation DGX 1 launch announcement. Return to citation DGX 1 and Pascal hardware description. Return to citation Turing architecture announcement. Return to citation GeForce RTX announcement. Return to citation Mellanox acquisition completion announcement. Return to citation Arm transaction termination announcement. Return to citation Fiscal 2023 financial results announcement. Return to citation Blackwell platform announcement. Return to citation Vera Rubin production progress announcement. Return to citation NVIDIA product catalog. Return to citation TensorRT workflow documentation. Return to citation GB200 NVL72 product documentation. Return to citation Fiscal 2026 financial results announcement. Return to citation Second quarter fiscal 2027 financial results announcement. Return to citation Ethernet product portfolio. Return to citation TensorRT documentation. Return to citation AI Enterprise licensing guide. Return to citation Omniverse documentation. Return to citation DRIVE platform. Return to citation Isaac platform. Return to citation Jetson modules. Return to citation Fiscal 2026 annual report Form 10 K. Return to citation Second quarter fiscal 2027 CFO commentary. Return to citation Second quarter fiscal 2027 quarterly report Form 10 Q. Return to citation Fiscal 2025 financial results announcement. Return to citation 2026 proxy statement. Return to citation First quarter fiscal 2026 financial results announcement. Return to citation AMD Instinct products. Return to citation Google Cloud TPU. Return to citation AWS Trainium. Return to citation NVIDIA historical stock prices. Return to citation NVIDIA Voyager headquarters photograph. GB200 NVL72 architecture and reference configuration. Source review for this website edition: 5 October 2026. The report’s information cutoff remains 3 October 2026. The Vera Rubin announcement date has been corrected to 31 May 2026, and the AlexNet paper link now points to the original NeurIPS publication.\n","date":"2026-10-05T22:00:00+08:00","image":"/post/nvidia-company-research/cover.webp","permalink":"/post/nvidia-company-research/","title":"NVIDIA Company Research Report"},{"content":"Early life, electrical engineering studies, family, the founding of NVIDIA, public interviews, education gifts and documented honors.  Born1963 · TaiwanCaltech biography Earned degreesElectrical engineeringB.S. 1984 · M.S. 1992 NVIDIA co-founded5 April 1993Company history Company rolesPresident · CEO · DirectorExecutive biography Cover: Jensen Huang, 5 January 2017. Photo: Maurizio Pesce · Source · CC BY 2.0. View full-size cover · 5184 × 3456.\nDownload the Chinese and English biography (Word)\nInformation cutoff: October 3, 2026.\nThis biography draws on school records, public interviews, company announcements, regulatory filings, and award organizations. Retrospective personal stories retain attribution to the speaker. The downloadable Chinese and English editions use the same reference numbers.\n1 From Taiwan to the United States Childhood places TaiwanBorn in 1963 ThailandPart of his childhood WashingtonStayed with an uncle in Tacoma KentuckyBoarding and elementary school OregonReunited with his parents Sequence of places recorded by Caltech; the exact early dates are approximate. Sources: Caltech · Oneida school newsletter.\nJensen Huang is also identified as Jen-Hsun Huang in earlier university and company records. Born in Taiwan in 1963, he spent part of his childhood in Thailand. Around age nine, he and his older brother moved to the United States before their parents. Caltech\u0026rsquo;s published biography records an initial stay with an uncle in Tacoma, Washington, followed by a move to Kentucky. He later reunited with his parents in Oregon as a teenager.[2]\nHe has often recalled his childhood stay at Oneida Baptist Institute in Kentucky. The school\u0026rsquo;s newsletter supplies a specific distinction: although the institute then operated principally as a high school, it allowed the young brothers to live on campus. They attended classes at nearby Oneida Elementary School. Their residence and the school where they took classes were separate institutions.[3]\nPublic accounts establish the sequence of places, but retrospective descriptions do not consistently agree on exact early dates and ages. This biography therefore uses approximate ages rather than turning an uncertain recollection into a precise date.\n2 Boarding life in Kentucky At the NBA All-Star Technology Summit on February 14, 2025, Huang told interviewer Becky Quick that he had shared a dormitory room with a sixteen-year-old boy while he was much younger. He recalled that other boys worked on a tobacco farm, whereas he cleaned bathrooms because he was the youngest. He described the work as difficult and said he tried to perform his assigned tasks well.[4]\nThese details come from Huang\u0026rsquo;s later recollection. The school independently records the brothers\u0026rsquo; accommodation and elementary-school attendance.[3][4]\nHe remained connected to the institution. Its 2019 newsletter records a teaching and residential building supported by Huang and his wife, bringing a place from his childhood back into the documented account of his adult life.[29]\n3 Oregon high school, table tennis and restaurant work After rejoining his parents, Huang attended Aloha High School in Oregon. HKUST\u0026rsquo;s honorary doctorate citation records that he graduated at sixteen and had been a nationally ranked table-tennis player. The sport was table tennis, rather than tennis.[5]\nHe also worked at Denny\u0026rsquo;s, including washing dishes, clearing tables, and serving customers. NVIDIA\u0026rsquo;s 2023 retrospective records this employment and his return to the San Jose restaurant associated with the company\u0026rsquo;s founding for a commemorative event.[10]\n4 University studies and family life Huang studied electrical engineering at Oregon State University, receiving his bachelor\u0026rsquo;s degree in 1984. He earned a master\u0026rsquo;s degree in electrical engineering from Stanford in 1992. Oregon State\u0026rsquo;s alumni record notes that he completed his graduate studies while working full time; eight years separated the two degree dates.[6]\nHis wife, Lori Huang, formerly Lori Mills, is also an Oregon State graduate. The university\u0026rsquo;s 2022 engineering alumni magazine records that they met as laboratory partners in a first-year engineering course and married a few years later.[7]\nThey have a son, Spencer, and a daughter, Madison. Huang recalled in 2010 that he continued taking Stanford classes while working, married, and raising children. His course load decreased after the children were born. He continued because he enjoyed studying and completed the master\u0026rsquo;s degree over eight years.[36]\n5 An engineer at AMD and LSI Logic BIOGRAPHY MAP\nEducation, work and public milestones Explore the institutions, roles and events documented in this biography.\nFit to view Expand all Collapse Full screen Read the full text outline Jensen HuangEducationSchool yearsOneida Elementary · classes Oneida Baptist Institute · boarding Aloha High School · Oregon Earned university degreesOregon State · electrical engineering B.S., 1984 Stanford · electrical engineering M.S., 1992 Semiconductor employmentAMD · microprocessor designer LSI Logic · director of CoreWare NVIDIA · since 1993Co-founder with Malachowsky and Priem President, CEO and director NVIDIA milestones during his tenurePublic listing and graphics1999 · Nasdaq IPO 1999 · GeForce 256 release Parallel computing and AI systems2006 · CUDA architecture unveiled 2016 · DGX-1 announced; delivery to OpenAI 2024 · Blackwell announced Education giftsUniversitiesStanford · Huang Engineering Center Oregon State · Collaborative Innovation Complex HKUST · Top Engineering Scholars Award School campusOneida · Jen-Hsun Huang Hall Branches organize documented facts. NVIDIA products were developed by company teams; the map does not assign their invention to one person. Gift donors and amounts are detailed in Chapter 14.\nSources: NVIDIA: Jensen Huang executive biography · Oneida Baptist Institute: July–August 2018 newsletter · HKUST: 2024 honorary doctorate citation · Oregon State: Huang Engineering Hall of Fame biography · Stanford Technology Ventures Program: Huang biography · NVIDIA: corporate history timeline · NVIDIA: DGX-1 launch announcement, April 5, 2016 · NVIDIA: DGX-1 delivery to OpenAI, August 2016 (republication archive) · NVIDIA: Blackwell platform announcement, March 18, 2024 · Stanford: fiscal 2008 financial review, Huang gift commitment · Oneida Baptist Institute: Fall 2019 On Campus newsletter · Oregon State: $50 million Huang gift, October 14, 2022 · HKUST Engineering: Top Engineering Scholars Award gift, September 2, 2025. Reviewed 2026-10-03.\nBefore founding NVIDIA, Huang worked at AMD and LSI Logic. Stanford\u0026rsquo;s entrepreneurship program records his roles as a microprocessor designer at AMD and director of CoreWare at LSI Logic, positions in semiconductor design and related technology operations.[8]\nNVIDIA\u0026rsquo;s other founders were Chris Malachowsky and Curtis Priem. Before the startup, they worked at Sun Microsystems while Huang worked at LSI Logic.[8][12]\nBy the time he completed his master\u0026rsquo;s degree in 1992, Huang had already worked in industry. His earned academic qualifications when he began the company were a bachelor\u0026rsquo;s and a master\u0026rsquo;s degree in electrical engineering. His later honorary doctorates were honors conferred by universities.[1][6]\n6 Three founders in 1993 NVIDIA\u0026rsquo;s official history gives April 5, 1993, as its founding date and names Huang, Malachowsky, and Priem as the founders. Its initial focus was three-dimensional graphics for gaming and multimedia. Huang has served as president, chief executive officer, and a director since the company\u0026rsquo;s inception.[1][9]\nThe Denny’s restaurant at 2484 Berryessa Road in San Jose, associated with NVIDIA’s founding discussions. Photographed on 2 July 2023.\nPhoto: Coolcaesar · Source · CC BY-SA 4.0.\nView full-size photo · 4568 × 2924 Some pre-founding discussions took place at a Denny\u0026rsquo;s restaurant in San Jose, California. NVIDIA returned to the restaurant for a commemorative event in 2023. The restaurant was a meeting place; incorporation, financing, hiring, and product development involved additional steps.[10]\nIn an April 2009 Stanford eCorner talk, Huang recalled treating his thirtieth birthday, February 17, 1993, as his first day working on the startup. He had not taken business or marketing classes and bought startup books and prepared financing presentation materials.[11]\nHe also recalled hiring a lawyer who introduced the founders to venture investors. Investors considered their work history, qualifications, and market opportunity. The team began building the company without first completing a detailed business plan. These details come from his account of its first few months.[11]\nHuang took the CEO role, but the enterprise was founded by three partners. Subsequent chips, software, and systems were developed by engineering teams.\n7 NV1, NV2 and the early product reversal NVIDIA\u0026rsquo;s first commercial chip, NV1, entered the market in 1995. The company\u0026rsquo;s IPO prospectus states that it stopped selling NV1 in the first quarter of 1996 and abandoned NV2 development. RIVA 128, introduced in August 1997, used a subsequent graphics architecture.[13]\nIn Sequoia\u0026rsquo;s interview, Huang recalled that the original graphics approach diverged from the interfaces and architecture adopted by the broader market. He described asking Sega to release NVIDIA from an ongoing development arrangement. In his account, Sega agreed to relieve the contractual obligation while paying the original contract amount.[12]\nThis negotiation story is a participant\u0026rsquo;s later recollection. The biography does not treat every contractual detail, date, or payment as separately verified by the prospectus. The prospectus directly establishes the sequence of NV1 discontinuation, NV2 abandonment, and RIVA 128\u0026rsquo;s introduction.[12][13]\nAs CEO, Huang participated in handling the product direction and customer relationship. The company was still private during this period, which included the discontinuation of its first product and cancellation of its next project.\n8 The 1999 IPO and GeForce 256 NVIDIA went public on Nasdaq on January 22, 1999, under the ticker NVDA. Its investor FAQ gives an original IPO price of $12 per share, before subsequent stock-split adjustments.[14]\nThat year, NVIDIA introduced GeForce 256 as a GPU. Its anniversary retrospective records October 11, 1999, as the release milestone and describes hardware transform and lighting that shifted graphics-processing work from the CPU.[15]\nThe IPO and the product release were distinct events: one brought the company into the public capital markets; the other was a processor development. Both occurred during Huang\u0026rsquo;s tenure, with the product\u0026rsquo;s engineering performed by company teams.\n9 CUDA and general-purpose parallel computing NVIDIA\u0026rsquo;s corporate history records the unveiling of CUDA architecture in 2006. CUDA provided a route for developers to program parallel computations on GPUs, extending their use beyond graphics rendering into scientific and other computing tasks.[9][16]\nGPUs can execute large numbers of suitable calculations in parallel, but applications still require programs, tools, and algorithms. NVIDIA\u0026rsquo;s CUDA developer materials provide tools, libraries, and programming resources. The company\u0026rsquo;s work at this stage included both processor hardware and software support.[16]\nHuang continued as CEO. CUDA is a technology platform developed and maintained by NVIDIA teams.[1][16]\n10 Deep learning and computing systems after 2012 Jensen Huang at GTC Taiwan on 21 September 2016. The stage display shows the Tesla P4 and P40 inference accelerators.\nPhoto: NVIDIA Taiwan · Source · CC BY 2.0.\nView full-size photo · 7952 × 5304 In 2012, Alex Krizhevsky, Ilya Sutskever, and Geoffrey Hinton published their image-classification neural-network paper. It describes using two NVIDIA GeForce GTX 580 GPUs and roughly five to six days to train the network. The architecture, training method, and experimental results were the researchers\u0026rsquo; work.[17]\nIn April 2016, NVIDIA announced DGX-1, combining multiple GPUs and associated software in a deep-learning system. An August company account records Huang personally delivering a DGX-1 to OpenAI in San Francisco. These are product-launch and delivery events; the delivery alone does not establish that a later model was produced entirely on that machine.[18][19]\nIn March 2024, NVIDIA announced Blackwell, describing GPUs, CPU combinations, and networking components. Huang\u0026rsquo;s public product presentations thus involved chips and larger computing systems, as well as personal-computer graphics cards.[23]\n11 Other events during his tenure Early product reversals were not the company\u0026rsquo;s only difficulties. NVIDIA\u0026rsquo;s fiscal 2009 annual report disclosed a weak die/packaging material set in certain previous-generation notebook GPUs and related chips. The company recorded a $196 million charge for anticipated warranty, repair, return, replacement, and associated costs. This was a documented product-quality event during Huang\u0026rsquo;s tenure.[20]\nOn April 27, 2020, NVIDIA announced completion of its Mellanox acquisition, valued at approximately $7 billion. Mellanox\u0026rsquo;s businesses included networking technology, which became part of NVIDIA following the transaction.[21]\nIn February 2022, NVIDIA and SoftBank announced termination of NVIDIA\u0026rsquo;s proposed Arm acquisition, citing significant regulatory challenges that prevented completion. The acquisition did not close, and Arm did not become a NVIDIA subsidiary through that transaction.[22]\n12 How he describes his management Speaking at Stanford on January 29, 2003, Huang discussed corporate culture. He identified innovation, acknowledging mistakes, and self-criticism as elements of NVIDIA\u0026rsquo;s culture and argued that a CEO should devote time to it. These were the management views he expressed at that time.[24]\nIn a 2024 Stanford GSB interview, he described presenting employees with facts, data, assumptions, and unknowns, and explaining his reasoning. He said that fewer information-transfer layers and greater access to decision context helped employees make independent decisions. This is his description of his approach, rather than a survey of all employees\u0026rsquo; experiences.[25]\nIn his November 23, 2024, HKUST conversation with Harry Shum, he again emphasized learning and accepting uncertainty. He said a leader need not know everything and could pursue a direction confidently while retaining room to learn. The university\u0026rsquo;s account preserves the discussion.[26]\nIn Stanford\u0026rsquo;s 2010 interview, he recalled feeling let down when a younger manager resigned while he was managing a design center at twenty-four. Years later, he considered his reaction mistaken and said he should respect colleagues\u0026rsquo; freedom to choose other jobs.[36]\n13 Public appearances and family roles Leather jackets recur in Huang\u0026rsquo;s public appearances. HKUST\u0026rsquo;s account of its November 2024 event records Huang and Shum wearing jackets bearing the university\u0026rsquo;s name. A 2025 university donation announcement also records a charitable auction involving a jacket. The clothing appears in both event and fundraising records.[26][31]\nHis family also appears in company filings. NVIDIA\u0026rsquo;s 2026 proxy statement discloses that his son and daughter work at the company, are not executive officers, and do not report directly to him. It describes their compensation arrangements. Those disclosures do not establish a family succession plan.[35]\nAt this biography\u0026rsquo;s cutoff, NVIDIA\u0026rsquo;s official biography continues to identify Huang as a founder, president, CEO, and director. Product presentations are one part of his public work; the formal positions also involve ongoing management and board responsibilities.[1]\n14 Four documented education gifts Education gifts in the public record Amounts retain their original currencies. Commitments, matching grants and announced gifts are identified individually.\nFiscal 2008\nUS$30 million\nStanford UniversityA commitment by Jensen and Lori Huang for the engineering center; the building was dedicated in October 2010.\nSources: Stanford: fiscal 2008 financial review, Huang gift commitment · Stanford Engineering: Huang Center dedication\n2019 school newsletter\nUS$2 million\nOneida Baptist InstituteA matching grant from the couple for Jen-Hsun Huang Hall, with dormitory rooms and classrooms.\nSources: Oneida Baptist Institute: Fall 2019 On Campus newsletter\n14 October 2022\nUS$50 million\nOregon State UniversityThe announced gift from the couple to the university foundation for the Collaborative Innovation Complex.\nSources: Oregon State: $50 million Huang gift, October 14, 2022\nSeptember 2025\nHK$10 million\nHKUSTA matching gift from the Jen-Hsun and Lori Huang Foundation for the Top Engineering Scholars Award; Terry Tsang separately pledged HK$10 million.\nSources: HKUST Engineering: Top Engineering Scholars Award gift, September 2, 2025\nJen-Hsun Huang Engineering Center at Stanford University, photographed on 16 December 2016. The dedication described in the text took place in 2010.\nPhoto: Frank Schulenburg · Source · CC BY-SA 4.0.\nView full-size photo · 5247 × 3498 Stanford\u0026rsquo;s fiscal 2008 financial review records Huang and Lori\u0026rsquo;s $30 million commitment for the new School of Engineering Center. A separate engineering-school account records the dedication on October 5, 2010. The center bears the name Jen-Hsun Huang.[27][28]\nOneida Baptist Institute\u0026rsquo;s Fall 2019 newsletter records a $2 million matching grant from the couple for Jen-Hsun Huang Hall. The building contains thirty-two dormitory rooms accommodating up to 128 girls, plus twelve classrooms. It also records the first resident assistants serving in the building during the 2019–20 school year.[29]\nOn October 14, 2022, Oregon State announced a $50 million gift from the couple to the university foundation for the Jen-Hsun and Lori Huang Collaborative Innovation Complex. The announcement lists research in AI, materials science, and robotics. This account records the announced gift and purpose; the projected opening date in that announcement is not treated as a confirmed completion date.[30]\nA September 2025 HKUST announcement records Terry Tsang\u0026rsquo;s HK$10 million pledge at a charitable event and a further HK$10 million matching gift from the Jen-Hsun and Lori Huang Foundation for the Top Engineering Scholars Award. These records distinguish the donors—individual, couple, or foundation—and the currencies involved.[31]\n15 Honors, public responsibilities and the present Huang entered Oregon State\u0026rsquo;s Engineering Hall of Fame in 2013. On November 23, 2024, HKUST conferred an honorary doctorate in engineering. University biographies also record his election to the US National Academy of Engineering.[5][6]\nThe 2025 Queen Elizabeth Prize for Engineering recognized seven modern-machine-learning laureates: Bill Dally, Fei-Fei Li, Geoffrey Hinton, John Hopfield, Huang, Yann LeCun, and Yoshua Bengio. The prize organization identifies Huang and Dally\u0026rsquo;s contribution with the hardware platforms supporting machine learning. Huang was one of seven recipients.[32]\nHuang received the 2026 IEEE Medal of Honor at the April 24 ceremony in New York City. On March 25 of the same year, the White House announced his appointment to the President\u0026rsquo;s Council of Advisors on Science and Technology, whose role includes providing science and technology advice to the president.[33][34]\nAs of October 3, 2026, his public career continues. This biography records education, employment, founding, product events, philanthropy, and appointments already documented, without assigning a predicted ending to his career, wealth, or company.\nAppendix Selected chronology Selected life and career milestones Personal events and NVIDIA milestones during Huang’s tenure. The chapters above provide the context and attribution.\n1963Born in Taiwan; later spent part of childhood in Thailand\nSources: [2]\nAround age nineMoved to the US with his brother before their parents; later lived and studied in Kentucky\nSources: [2][3]\n1984Electrical engineering bachelor\u0026#x27;s degree, Oregon State\nSources: [6]\n1992Electrical engineering master\u0026#x27;s degree, Stanford\nSources: [6]\nApril 5, 1993Founded NVIDIA with Malachowsky and Priem\nSources: [9]\nAugust 1997NVIDIA introduced RIVA 128\nSources: [13]\nJanuary 22, 1999NVIDIA\u0026#x27;s Nasdaq IPO\nSources: [14]\n2006NVIDIA unveiled CUDA architecture\nSources: [9]\nOctober 5, 2010Stanford Huang Engineering Center dedication\nSources: [28]\n2016DGX-1 announced; Huang delivered a system to OpenAI\nSources: [18][19]\nOctober 2022Huang and Lori\u0026#x27;s $50 million Oregon State foundation gift announced\nSources: [30]\n2025Queen Elizabeth Prize for Engineering shared with six other recipients\nSources: [32]\nMarch–April 2026PCAST appointment and IEEE Medal of Honor presentation\nSources: [33][34]\nPersonal recollections are attributed to the identified interviews. Original or archival links appear in the references. Some historical pages and PDFs may change their online availability; their titles, publishers, and dates support further retrieval. Sources NVIDIA: Jensen Huang executive biography. Return to citation Caltech: 2024 commencement speaker biography. Return to citation Oneida Baptist Institute: July–August 2018 newsletter. Return to citation NBA: Huang interview transcript, Technology Summit, February 14, 2025. Return to citation HKUST: 2024 honorary doctorate citation. Return to citation Oregon State: Huang Engineering Hall of Fame biography. Return to citation Oregon State: 2022 Oregon Stater Awards magazine. Return to citation Stanford Technology Ventures Program: Huang biography. Return to citation NVIDIA: corporate history timeline. Return to citation NVIDIA: Huang returns to Denny’s, September 2023. Return to citation Stanford eCorner: The First Six Months of NVIDIA, April 8, 2009. Return to citation Sequoia Capital: Crucible Moments, NVIDIA, episode 8. Return to citation NVIDIA: initial public offering prospectus, 1999. Return to citation NVIDIA: investor FAQs, IPO date. Return to citation NVIDIA: GeForce 256 retrospective, October 11, 2024. Return to citation NVIDIA: CUDA developer documentation. Return to citation Krizhevsky et al.: ImageNet Classification with Deep Convolutional Neural Networks, 2012. Return to citation NVIDIA: DGX-1 launch announcement, April 5, 2016. Return to citation NVIDIA: DGX-1 delivery to OpenAI, August 2016 (republication archive). Return to citation NVIDIA: fiscal 2009 annual report. Return to citation NVIDIA: completion of Mellanox acquisition, April 27, 2020. Return to citation NVIDIA and SoftBank: termination of Arm acquisition, February 2022. Return to citation NVIDIA: Blackwell platform announcement, March 18, 2024. Return to citation Stanford eCorner: The Power of Corporate Culture, January 29, 2003. Return to citation Stanford GSB: Huang on first-principles thinking, April 25, 2024. Return to citation HKUST: Huang–Harry Shum fireside chat, November 23, 2024. Return to citation Stanford: fiscal 2008 financial review, Huang gift commitment. Return to citation Stanford Engineering: Huang Center dedication. Return to citation Oneida Baptist Institute: Fall 2019 On Campus newsletter. Return to citation Oregon State: $50 million Huang gift, October 14, 2022. Return to citation HKUST Engineering: Top Engineering Scholars Award gift, September 2, 2025. Return to citation QEPrize: 2025 Modern Machine Learning laureates. Return to citation IEEE: 2026 Medal of Honor ceremony account. Return to citation White House: PCAST appointments, March 25, 2026. Return to citation NVIDIA: 2026 proxy statement. Return to citation Stanford Engineering: Huang interview following the 2010 Huang Center dedication. Return to citation ","date":"2026-10-05T12:00:00+08:00","image":"/post/jensen-huang-biography/cover-hd.jpg","permalink":"/post/jensen-huang-biography/","title":"Jensen Huang Biography"},{"content":"An evidence-based guide to commercial space history, the value chain and representative companies, market structure, and the path from engineering capability to delivery and cash flow. Understanding the Value Chain, Competitive Landscape and Business Realities Through Rocket Recovery\nDEX Research | Sources verified as of October 3, 2026. Industry revenue and company financial data cover 2025. Publication dates, statistical periods and the scope of each source are provided in the separate appendix.\nThis report examines commercial space and the materials, precision manufacturing, critical components, testing and systems integration that support it. Its coverage of advanced engineering is limited to space-related businesses. The analysis draws on publicly available sources from around the world and incorporates examples from China. U.S. government procurement and regulatory examples explain specific mechanisms; the applicable arrangements must still be distinguished by country and mission.\nReal demand for services and actual procurement already exist in commercial space, but technical breakthroughs, growth in industry revenue and company profitability represent different levels of progress. For an engineering capability to generate revenue repeatedly, it must also move through customer procurement, contract execution, delivery and payment collection. This report follows that path to explain how the industry works. The companies cited illustrate business roles rather than market share rankings.\nPart 1 Industry Story After a Rocket Lands In December 2015, a SpaceX Falcon 9 launched from Florida on a communications satellite deployment mission. The rocket\u0026rsquo;s first stage returned to Earth and completed a vertical landing. On March 30, 2017, a Falcon 9 first stage that had previously flown on a mission was used for another orbital launch. [1]\nThese two moments validated two connected capabilities: bringing hardware back and having previously flown hardware undertake another mission. The hardware reused here was the rocket\u0026rsquo;s first stage; the validation did not cover the entire Falcon 9. For engineering teams, recovery and reflight provided evidence from actual missions. For business operators, the calculation of costs and returns was only beginning.\nBuilding new hardware requires spending on materials, processing and assembly. Reuse can reduce some of that repeated spending, but it also adds recovery, transportation, inspection, repair and turnaround work. Whether the engines, structures and control systems are fit for the next mission requires engineering assessment. After the hardware returns to Earth, this work must still be completed before it can carry out another customer mission.\nMission frequency further changes these economics. Factories, launch facilities, mission control and permanent teams all incur ongoing expenses. If there are too few orders ready for execution, or payload, launch site and mission preparations cannot be synchronized, the total cost of each mission may remain high even when hardware manufacturing costs are reduced. A publicly quoted price is the price offered to customers for a service; the company\u0026rsquo;s actual cost depends on the resources needed to complete the mission.\nCustomers also make their own calculations. A delay in bringing a communications satellite into service affects its capacity available for sale; a scientific mission may face constraints imposed by its orbit and observation conditions. Customers are buying a transportation outcome under agreed orbital, timing and delivery conditions. Missions differ in their requirements for price, reliability and scheduling flexibility. This is also why large rideshare launches and dedicated small-satellite launches can coexist.\nTransactions continue after the rocket lifts off. Satellites must be deployed and commissioned in orbit, operators must connect ground networks, and terminals must be installed for users. Earth observation data needs to be processed before it can enter workflows for insurance, asset monitoring or public administration. Each stage can generate revenue, and each also requires resources and entails delivery responsibilities.\nCommercial space is therefore a set of interconnected businesses. Launch determines how payloads reach space, advanced engineering determines whether equipment can be manufactured consistently and perform its missions, and downstream services determine why customers continue paying. To understand how these businesses developed, we first review the industry\u0026rsquo;s history, then examine its current value chain and forms of competition.\nPart 2 Industry History INDUSTRY MAP\nSix stages in the development of commercial space Trace how communications demand, specialized launch, public procurement, reuse and serial manufacturing accumulated into today’s industry.\nFit to view Expand all Collapse Full screen Read the full text outline Commercial space1960s · Comms [11]Key milestones1962 · Telstar 1 transatlantic TV demo 1964 · INTELSAT agreements 1965 · Early Bird to commercial use 1980s · Launch [12]Key milestones1980 · Arianespace founded May 1984 · Spacenet F1 commercial debut Customers buy launch services 2006\u0026#43; · COTS [2][24]Key milestonesCOTS · Development milestones 2007 · Rocketplane Kistler: funding gap ends deal Dec 2008 · CRS before demos finish Development ≠ service procurement 2014 · Crew [28]Key milestonesSep 2014 · Boeing / SpaceX CCtCap Company-owned and operated NASA requirements and certification 2015–17 · Reuse [1]Key milestonesDec 2015 · First-stage landing Mar 2017 · SES-10 reflight First-stage reuse ≠ proven profit 2020s · Scale [13][14][23]Key milestones2022 · Airbus / OneWeb batch production Jul 2023 · Zhuque-2 Y2 orbital success Oct 2, 2026 · Airbus delivers 32 OneWeb satellites Delivery ≠ launch or in-orbit service Based on the accompanying report and its evidence appendix, with sources verified through October 3, 2026. Diagram preparation does not extend that research cutoff. These stages summarize the report’s historical examples; they are neither an exhaustive chronology nor a universal sequence for every country. Earlier businesses continue alongside newer technologies. The 2015 and 2017 events concern the Falcon 9 first stage, not reuse of the entire rocket. The October 2026 Airbus announcement establishes manufacturing delivery and shipment preparation, not completed launch or service entry.\nSources: [11] NASA · communications satellite history · [12] Arianespace · commercial launch history · [24] NASA · COTS final report · [28] NASA · 2014 CCtCap awards · [1] NASA · December 2015 first-stage landing · [1] SES · 2017 results and SES-10 reflight · [13] Airbus · OneWeb serial manufacturing · [14] CNSA · Zhuque-2 Y2 mission · [23] Airbus · 32 OneWeb satellites delivered. Reviewed 2026-10-03.\nI The 1960s Communications Demand Predates New Rocket Companies The history of commercial space predates today\u0026rsquo;s reusable rockets. On July 10, 1962, Telstar 1 was launched to demonstrate transatlantic television communications. NASA\u0026rsquo;s historical records state that AT\u0026amp;T built the satellite and NASA provided the launch under a cost-reimbursement arrangement. On April 6, 1965, COMSAT\u0026rsquo;s Early Bird was launched, taking satellite communications further toward commercial service. [11]\nThis history shows that one early path to commercializing orbital equipment was to meet communications demand that already existed on Earth. Satellite manufacturing, launch and ground facilities were investments needed to deliver the service, while operators recovered their spending through communications business. What customers ultimately bought was usable connectivity rather than the satellite\u0026rsquo;s technical specifications.\nInternational communications also required organizational cooperation. NASA\u0026rsquo;s historical material records that the agreements establishing INTELSAT were signed in August 1964. [11] Network development required technology, ground access and operating arrangements to be brought together. This system-wide requirement continues today, although service types, network architectures and participating companies have changed.\nII The 1980s Specialized Launch Becomes a Service Customers Can Buy Arianespace was established in 1980 and carried out its first commercial mission in May 1984, placing the Spacenet F1 communications satellite into orbit. [12] Commercial customers were purchasing transportation from specialized launch providers before the emergence of a new generation of private rocket companies.\nSpecialization allowed satellite customers to purchase launch services without building their own launch systems. Launch providers prepared and carried out transportation, while customers organized investments around their communications or observation businesses. Specialized suppliers took on complex engineering, and commercial transportation became a distinct business within the value chain. Later entrants changed technical approaches, cost structures and forms of competition within an existing market.\nIII From 2006 Government Supports Development and Purchases Services NASA launched the Commercial Orbital Transportation Services (COTS) program in 2006 to support companies developing transportation capabilities. Its funded Space Act Agreements used milestones agreed in advance: companies received the corresponding milestone payments after meeting the relevant objectives. Companies invested their own funds and bore cost overruns incurred in meeting those objectives, while NASA provided technical assistance and verified progress. COTS itself was not a contract to procure the final transportation service under the Federal Acquisition Regulation. [2][24]\nGovernment support for development did not eliminate companies\u0026rsquo; financing risks. The first round of partners included SpaceX and Rocketplane Kistler. NASA terminated its agreement with the latter in October 2007 for reasons that included its failure to secure the required capital. Following the second round of selection, NASA signed an agreement with Orbital Sciences in February 2008. [24] The ability to complete engineering objectives and the availability of funding to sustain operations before milestone payments arrive constrain each other.\nNASA subsequently purchased cargo transportation to the International Space Station through Commercial Resupply Services (CRS) contracts. The first round of contracts was awarded in December 2008, before the COTS demonstrations had been completed. Development support and service procurement used different legal arrangements and overlapped in time. [2][24] A customer can specify future mission needs in advance, but companies must still develop the capabilities needed to carry out the transportation.\nThis mechanism assigned some responsibility for design, development and operations to companies while retaining public funding, technical assistance and government demand. It connected development with the market, but companies still bore the risk of a failed approach and cost overruns. The market structure of commercial space developed through this interaction between business operations and public procurement.\nIV 2014 Commercial Transportation Expands to Crewed Missions On September 16, 2014, NASA announced Commercial Crew Transportation Capability (CCtCap) contract awards to Boeing and SpaceX. These firm-fixed-price procurement arrangements covered certification, testing and the corresponding transportation missions. The companies owned and operated the transportation systems, while NASA set mission requirements and conducted safety and performance certification. [28]\nCrew transportation expanded the scope of companies\u0026rsquo; services and increased their validation responsibilities. Commercialization changed ownership and procurement arrangements, while safety and performance certification remained a prerequisite for carrying out missions. Potential contract value describes the scale of future work, technical certification confirms the relevant capability, and actual missions and revenue reflect subsequent execution. Planned dates and actual completion dates must also be recorded separately.\nV 2015 to 2017 From Recovery to Reflight The first-stage landing in 2015 and reflight in 2017 provided evidence of recovery and the ability to perform another orbital mission, respectively. [1] They expanded the commercial possibilities of reuse technology. Long-term business returns depend on subsequent maintenance, turnaround and mission execution; profit per mission or the percentage of costs saved requires supporting cost data.\nMoving from a single demonstration to consistent operations requires hardware condition assessment, mission preparation and the supply chain to work together. Increasing the number of reuses offers an opportunity to improve business performance only if turnaround spending, reliability and order execution can all be kept under control. Competition therefore extends from the ability to manufacture hardware for a single mission to the organizational ability to carry out missions repeatedly.\nVI The 2020s Constellation Deployment Drives Serial Manufacturing In February 2022, Airbus described the production system for OneWeb satellites: suppliers delivered components in volume, satellites were manufactured on high-speed assembly lines, and digital production and inspection tools were used. [13] Constellations require large amounts of equipment to operate together, making manufacturing consistency, testing cadence and network replenishment plans business considerations.\nBatch manufacturing can also reproduce the same design or process problem across multiple products. Standardization helps repeat procurement and assembly, provided that interfaces, quality control, and change management develop alongside it. Building a qualified prototype and continuously delivering a batch of reliable products are related but distinct capabilities.\nChinese companies are also advancing their launch capabilities. On July 12, 2023, LandSpace\u0026rsquo;s Zhuque-2 Y2 successfully reached orbit. The China National Space Administration recorded it as the world\u0026rsquo;s first rocket powered by liquid oxygen and methane to successfully deliver a payload to its intended orbit. [14] This event demonstrated mission progress for a particular launch technology; the ability to provide services consistently still needs to be assessed through subsequent missions and business results.\nOn October 2, 2026, Airbus announced the delivery of the first 32 next-generation OneWeb satellites, ready for shipment to the United States for subsequent launch. [23] The announcement documented progress in manufacturing delivery; transportation, launch, in-orbit commissioning and entry into service remained subsequent steps. Industrialization therefore consists of a sequence of deliveries, rather than just a planned satellite count.\nThese historical stages gradually accumulated to form today\u0026rsquo;s industry structure. Communications services, government procurement, specialized manufacturing and new launch technologies continue to coexist, with different customer needs driving the development of different engineering capabilities. Commercial space needs to be understood in the context of these interconnected businesses.\nPart 3 Industry Value Chain The value chain is divided by business function, and a company can participate in several segments. The following sections explain how materials, components and engineering services support spacecraft manufacturing, then examine how manufacturing and launch create network capabilities and why downstream customers pay. A company\u0026rsquo;s appearance in different segments does not mean that the same revenue can be counted more than once.\nINDUSTRY MAP\nThe commercial space and advanced engineering value chain Explore nine business functions, the engineering capabilities behind them, and representative companies with documented roles.\nFit to view Expand all Collapse Full screen Read the full text outline Commercial space and supporting engineeringUpstream · Materials, components and engineeringMaterials \u0026amp; precision manufacturing [15]Consistent materials and engineered structures Hexcel · composite materials Beyond Gravity · fairings, adapters and separation systems Propulsion, control \u0026amp; space electronics [3][16]Power, attitude control, propulsion and data handling Moog · propulsion, fluid control and avionics Rocket Lab · star trackers, reaction wheels and solar power Testing, verification \u0026amp; integration [17][19]Mechanical, thermal-vacuum and compatibility tests Airbus · testing services and facilities Thales Alenia Space · assembly, integration and testing Rocket Lab · spacecraft-level test facilities Midstream · Manufacturing, launch and operationsLaunch services [18]Payload delivery to an agreed orbit and schedule SpaceX · Falcon and rideshare arrangements Rocket Lab · Electron and Kick Stage LandSpace · Zhuque development, manufacturing and launch Satellite manufacturing [19]Bus functions, payload integration and complete satellites Airbus · complete satellites or bus-only solutions Thales Alenia Space · communications and observation systems Rocket Lab · configurable spacecraft and mission services Constellation operations [6][20]Spacecraft, ground access and network management SpaceX / Starlink · low Earth orbit broadband network Eutelsat · GEO business and OneWeb LEO network Iridium · cross-linked LEO voice, data and IoT network Downstream · Connectivity, data and mission servicesSatellite communications [20]Connectivity, terminals, installation and customer support Starlink · broadband connectivity services Eutelsat · capacity and managed service packages Iridium · voice, messaging, data and IoT services Remote sensing \u0026amp; data services [21]Observation data, processing and usable analysis Planet · optical imagery, mosaics and analytics BlackSky · constellation and Spectra platform ICEYE · SAR observations and Flood Insights Government \u0026amp; research missions [22]Public demand spans the entire value chain SpaceX · Dragon resupply transportation Lockheed Martin · Orion spacecraft manufacturing Rocket Lab · NASA small science mission launches Based on the accompanying report and its evidence appendix, with sources verified through October 3, 2026. Diagram preparation does not extend that research cutoff. DEX editorial taxonomy, not a verified supplier-contract network, market-share ranking or exhaustive company list. Companies may span several stages; their revenue must not be counted twice. Government and research customers generate demand across the chain. GEO means geostationary orbit; LEO means low Earth orbit; IoT means Internet of Things; SAR means synthetic aperture radar. Advanced engineering here is limited to space-related business.\nSources: [15] Hexcel · space materials · [15] Beyond Gravity · structures and separation · [16] Moog · space products · [17] Airbus · test services · [17] Thales Alenia Space · assembly, integration and testing · [18] SpaceX · Falcon User’s Guide · [18] Rocket Lab · Electron · [18] LandSpace · launch activities · [19] Airbus · satellite and bus solutions · [19] Rocket Lab · spacecraft and integration · [20] Starlink · 2024 progress report · [20] Iridium · network · [20] Eutelsat · managed connectivity · [21] Planet · products · [21] BlackSky · constellation and Spectra · [21] ICEYE · SAR data · [21] ICEYE · Flood Insights · [22] NASA · resupply transportation · [22] NASA · Orion and prime contractor · [22] Rocket Lab · NASA Aspera launch award. Reviewed 2026-10-03.\nI Upstream Materials Components and Engineering Capabilities 1 Materials and Precision Manufacturing A spacecraft must first withstand the launch environment before it can operate in orbit. Its structure must withstand vibration and changes in loading, while its materials must accommodate temperature changes. Lightweight design therefore has to meet requirements for weight, strength and dimensional stability together. Hexcel\u0026rsquo;s space product information lists applications for composites in fairings, boosters and satellite structures, and also highlights the requirements imposed by temperature changes and dimensional stability. [15]\nHexcel and Beyond Gravity occupy different positions in this segment. Hexcel supplies material systems including carbon fiber, prepregs, honeycomb cores and bonding materials; Beyond Gravity provides fairings, interstage structures, payload adapters and separation systems. Fairings protect payloads during passage through the atmosphere, while separation systems first maintain the connection and then release payloads as the mission requires. Material suppliers deliver materials that meet specifications, and structure suppliers further turn those materials into products that perform load-bearing, mounting or release functions. [15]\nThe challenge in moving from qualified samples to batch delivery is maintaining consistency in subsequent batches. Buyers need to trace material origins, processing steps and the batches affected by anomalies; manufacturers must translate drawings into tooling, processing parameters, inspection methods and complete records. Expanding production therefore involves both equipment and process systems: additional equipment needs reliable processing and inspection procedures before it can produce qualified output.\nMaterials and structural components typically enter procurement and production schedules as part of customer projects. A long record of verification helps suppliers secure orders for subsequent batches and makes their process experience a barrier to competition. At the same time, specialized equipment needs enough production work to spread its costs; when customer projects are delayed, idle capacity still ties up funds. Business results in this segment depend on whether technical capabilities can continue to translate into procurement, delivery and customer payments.\n2 Propulsion Control and Space Electronics Once in orbit, a satellite still needs to maintain its condition and carry out its mission. The power system supplies electricity, avionics process commands and data, the attitude control system maintains the pointing of cameras or antennas, and the propulsion system performs the required maneuvers. Rocket Lab\u0026rsquo;s space systems products include solar cells and arrays, reaction wheels, star trackers, radios and software; Moog provides propulsion, fluid control, avionics, power and mechanism products. [3][16]\nStar trackers and reaction wheels perform measurement and actuation, respectively. A star tracker determines a satellite\u0026rsquo;s attitude by observing stars. Rocket Lab\u0026rsquo;s ST-16HV has an onboard star catalog and processor and can output attitude and angular velocity; a reaction wheel receives commands for rotational speed, angular momentum or torque and helps change or maintain pointing. Control software translates measurements into action commands, with sensors and actuators working together to control pointing. [16]\nWhen components are integrated into a satellite, they must be compatible with other systems in terms of voltage, communications interfaces, heat dissipation and mass. Moog\u0026rsquo;s avionics products can be used for command and data handling, power and payload applications; Rocket Lab\u0026rsquo;s components are used in its own spacecraft and also supplied for other missions. Component suppliers can operate independent product lines around a particular function, while satellite manufacturers can incorporate their own components into bus solutions. [3][16][19]\nAn existing flight record helps customers assess product risk, while new missions still require verification of interfaces and compatibility. Suppliers\u0026rsquo; production schedules are affected by product selection, procurement batches and customer progress. Even small components can face lengthy verification cycles. When demand is concentrated in a few constellations, changes to customer designs or a slowdown in deployment flow through to component orders and factory scheduling.\n3 Testing Verification and Systems Integration Testing turns design requirements into results that can be checked. Vibration, acoustic and shock testing address the mechanical environment; thermal vacuum testing examines performance under temperature and low-pressure conditions; electromagnetic compatibility testing examines interference between equipment. Airbus publicly offers these environmental tests and also undertakes test facility engineering, equipment supply, training and the construction of assembly, integration and testing centers. [17]\nSystems integration must also address what happens after individual components have qualified. Power connections, command transmission, sensor feedback and communications links need to be verified in combination. Rocket Lab\u0026rsquo;s spacecraft manufacturing facilities have thermal vacuum, vibration and electromagnetic compatibility testing capabilities for complete satellites; Thales Alenia Space\u0026rsquo;s Rome facility connects design, assembly, integration and testing, using modular clean rooms and digital production tools. [17][19]\nTest results establish whether equipment meets requirements under specified conditions and provide engineering evidence for design changes and release for delivery. Their applicability corresponds to the conditions tested. Customers may purchase testing services, test equipment or facility construction; verification within a spacecraft manufacturer is engineering work needed to deliver the product. These different business models have different revenue sources and carry different equipment and staffing costs.\nBatch manufacturing makes testing a capacity constraint for the entire production line. As assembly accelerates, thermal vacuum, vibration and other testing must keep pace; otherwise, assembled satellites will still queue for testing. Thales Alenia Space\u0026rsquo;s description of its factory capabilities includes plans for batch production of smaller satellites, which describe planned capacity. [17] Moving from plans to sustained delivery requires assembly, test facilities and staffing to work together.\nII Midstream Manufacturing Launch and Operations 1 Launch Services A launch company delivers the service of taking a payload to the agreed orbit. Customers care not only about how much the rocket can carry, but also about orbit, timing, payload interfaces and the mission success record. Satellites of the same mass can require different transportation conditions when their target orbits differ. Published nominal payload capabilities apply under specific conditions; the capacity available for a particular mission is also constrained by its orbit and flight plan.\nSpaceX provides transportation services through its Falcon series and lists multiple-payload and dedicated rideshare arrangements in its Falcon User\u0026rsquo;s Guide; Rocket Lab\u0026rsquo;s Electron focuses on small satellite launches, with its product page emphasizing orbital choice and schedule control; LandSpace is developing research, manufacturing, testing and launch activities around its Zhuque series of liquid oxygen–methane rockets. The three companies have different focuses in payload range, mission arrangements and business stage. [18]\nRideshare allows several payloads to share a launch, requiring customers to accommodate common launch and interface arrangements. A dedicated mission can be organized around one customer\u0026rsquo;s timing and orbital requirements. Whether the customer is willing to pay for that arrangement depends on the business value of reaching orbit on schedule. Rocket Lab also offers a Kick Stage for precise deployment; SpaceX\u0026rsquo;s guide lists different dispensers and mission interfaces. Payload compatibility and deployment into orbit are both parts of the transportation service. [18]\nLaunch delivery depends on a sequence of activities working together: preparing the rocket, coordinating the launch site, checking payload interfaces and executing mission procedures. A delay in a customer\u0026rsquo;s payload can also change the launch schedule. Reuse adds recovery, maintenance and turnaround management; savings in hardware manufacturing expenditure must be assessed alongside these inputs and the actual mission frequency. Repeatedly carrying out customer missions is what allows manufacturing and operational capabilities to become an ongoing business.\n2 Satellite Manufacturing A satellite consists of a bus and a payload. The bus provides basic functions such as structural support, power, thermal control, and attitude and orbit control; the payload performs specific tasks such as communications or observation. Airbus publicly offers complete satellites or bus-only solutions; Rocket Lab\u0026rsquo;s spacecraft buses integrate power, propulsion, star trackers, reaction wheels, radios and software. Customers can purchase a complete satellite or purchase a bus and then integrate a payload suited to their mission. [19]\nReusing an existing bus design can preserve common components and manufacturing processes, but a new payload still requires compatibility work. Higher payload power consumption requires adjustments to power and heat dissipation; greater pointing accuracy requires the attitude control system to meet new requirements. Batch manufacturing therefore involves coordinating common designs with mission needs: repeating procurement and assembly where possible while retaining the configurations needed to perform specific missions.\nAirbus participates in manufacturing for constellations such as OneWeb, Thales Alenia Space provides communications, observation and other space systems, and Rocket Lab provides configurable spacecraft and related mission services. Rocket Lab\u0026rsquo;s publicly described services include design, manufacturing, integration, testing, launch arrangements and in-orbit management, and individual contracts can cover different parts of this work. Bus supply, payload integration and contracting for a complete mission therefore entail different delivery scopes and responsibilities. [19]\nManufacturers organize development, procurement, assembly and testing under their contracts, then fulfill their obligations according to delivery and acceptance requirements. For fixed-price projects, a rise in costs does not necessarily bring a corresponding increase in the contract price, and overruns can erode profit. The gap between supplier payments and customer receipts also ties up manufacturers\u0026rsquo; funds. Contract signing, completion of manufacturing, acceptance and entry into in-orbit service reflect different stages of contract execution, product delivery and customer use.\n3 Constellation Operations Constellation operators must organize dispersed spacecraft into a continuously operating network. SpaceX provides low Earth orbit broadband through Starlink; Eutelsat combines its geostationary orbit (GEO) business with OneWeb\u0026rsquo;s low Earth orbit network; Iridium operates a cross-linked low Earth orbit network, providing voice, data and Internet of Things connectivity. Different network architectures and product capabilities serve different customer needs. [20]\nSatellites must work with ground access, user terminals, routing and network management to create services that can be sold. Demand in a particular region must also match the capacity available there. Iridium publicly describes how its cross-linked satellites work with ground gateways, as well as its operations team\u0026rsquo;s monitoring, software upgrades and anomaly handling. Eutelsat both sells network capacity and provides managed services, illustrating different ways that operators can deliver network capabilities to customers. [20]\nOnce a network enters operation, maintenance and renewal become ongoing work. As satellites reach the end of their lives, companies need to replenish or replace them; user terminals and ground systems also need maintenance. Capacity at peak demand, anomaly handling and equipment renewal all affect the service customers receive. Constellation construction is therefore a process of continuing investment, with operating and replenishment expenditure remaining after the initial deployment.\nOperating revenue comes from customers purchasing capacity and services. Expanding coverage can bring new customers, but may first require more investment in satellites, ground facilities and operations; added capacity must actually be used and purchased to improve business results. The construction stage primarily turns funding into network capabilities, while the operating stage must turn those capabilities into service contracts, user payments and renewals. The two stages involve different funding needs and business tasks.\nIII Downstream Communications Data and Mission Services 1 Satellite Communications The downstream segment turns network capabilities into connectivity services for specific users. Operators often also sell services, so the same company can appear in both the midstream and downstream segments: the former explains how the network operates, while the latter explains what customers buy. Starlink is a SpaceX business brand that provides broadband connectivity; Eutelsat offers capacity and service packages to enterprise customers and others; Iridium provides voice, messaging, data and Internet of Things connectivity. [20]\nConnectivity requirements vary by use. Households need everyday internet access, maritime and aviation users need connections while moving, and businesses may need remote-site access or backup links. In November 2023, Eutelsat OneWeb and ICT integrator NEC XON announced a multiyear distribution agreement for sub-Saharan Africa covering installation, training, and bandwidth and service-level arrangements. [20] This case shows how a network operator can work through local integrators to connect satellite capacity with terminal delivery and customer support.\nBroadband and Internet of Things connectivity serve different tasks. Video, multiuser online applications and file transfers require corresponding network capacity; asset tracking and sensor messages may place greater emphasis on coverage, power consumption and message reliability. Iridium offers different voice, messaging and IoT products, illustrating how a single network supports multiple uses. Different uses entail different terminals, service requirements and procurement needs. [20]\nSubscriptions and renewals can generate recurring revenue, but service providers must also bear the costs of terminals, installation, customer support, distribution and network maintenance. Enterprise customers agree on service scope and duration through contracts, while individuals decide whether to renew based on their experience. A service\u0026rsquo;s ability to keep meeting customers\u0026rsquo; needs affects retention; the revenue and service costs associated with each customer together determine whether the subscription business can generate a profit.\n2 Remote Sensing and Data Services Remote sensing companies can develop products at different stages, including image acquisition, processing and the delivery of analysis. Planet provides optical imagery, image mosaics and analytics products; BlackSky combines its own low Earth orbit constellation with the Spectra tasking and analytics platform; ICEYE provides synthetic aperture radar observations and uses radar data for information services such as those for floods. Companies can deliver either observation data or information products developed from that data. [21]\nOptical and radar observations provide different information. Optical imagery records features in visible light and other spectral bands, with effective observation affected by illumination, clouds and other conditions; PlanetScope documentation specifies solar elevation requirements. Synthetic aperture radar (SAR) actively transmits radar signals and measures their returns, providing complementary observations at night and under cloud cover, and its images require different interpretation methods. Each technology has its own operating conditions, and the two can be combined according to the information a mission requires. [21]\nWhat customers really need is a result they can act on. Insurers assessing a disaster\u0026rsquo;s impact need information on location, extent and losses; energy companies monitoring assets need to identify changes promptly. ICEYE\u0026rsquo;s Flood Insights combines its own radar imagery with third-party data, algorithms and expert analysis to provide information such as flood depth and extent; Planet and BlackSky also connect observation data to workflows through analytics and platform tools. [21]\nRevisit frequency determines observation opportunities, resolution affects the detail that can be identified, and delivery speed determines when customers receive results. For optical observation, a satellite\u0026rsquo;s return to a target area must still coincide with suitable illumination and cloud conditions; once an image is acquired, it also needs processing and interpretation. The commercial value of remote sensing depends on whether the entire process can deliver usable results promptly and continue to meet customers\u0026rsquo; monitoring or decision-making needs.\n3 Government and Research Missions Government and research customers are a source of demand across the value chain, purchasing launch, transportation, observation, buses or complete missions. SpaceX uses Dragon to carry out NASA\u0026rsquo;s space station resupply missions; Lockheed Martin manufactures the Orion spacecraft for NASA; Rocket Lab carries out launches for NASA\u0026rsquo;s small science missions. Purchasing transportation on a per-mission basis and commissioning a company to develop and manufacture a spacecraft entail different procurement scopes, execution responsibilities and revenue arrangements. [22]\nNASA\u0026rsquo;s PREFIRE mission illustrates a specific division of work: Blue Canyon Technologies built the CubeSats, JPL supplied the spectrometers, Rocket Lab provided Electron launches, and the University of Wisconsin–Madison is responsible for processing the data. Satellite buses, science payloads, transportation and research are handled by different participants. A complete science mission can generate purchases of several products and services along the value chain. [22]\nPublic procurement can help companies establish delivery records, while requiring suppliers to organize their work around mission objectives, quality and schedules. Procurement quantities, time frames and contract scopes affect companies\u0026rsquo; production arrangements and revenue timing. Government agencies act here as customers or collaborating institutions; companies provide the agreed mission capabilities, and the formation and continuity of demand depend on the corresponding project arrangements.\nOrders, revenue and cash record different stages of business progress. Orders reflect agreed purchases and missions; revenue is recognized for performance under the applicable rules; cash reflects actual receipts and payments. Contract execution requires funding, and a company may still have to wait for the remaining payment after completing a mission. Whether public demand can support an ongoing business depends both on how much work a company completes and on its fulfillment costs and payment arrangements.\nIV Competition Across the Value Chain 1 Vertically Integrated Companies SpaceX connects its launch operations with the Starlink network business, while Rocket Lab provides both launch services and space systems.[3][18][20] A company\u0026rsquo;s own operations can generate internal demand, and component and spacecraft teams can coordinate designs directly. Vertical integration can therefore improve coordination between systems and delivery schedules. Its value ultimately depends on mission execution and costs, rather than simply the number of activities a company covers.\nIntegration also expands funding and project management responsibilities. Companies must sustain factories, mission teams, and multiple product lines at the same time, while delays in new projects can tie up capital for extended periods. Internal deployment creates engineering work, while purchases by external customers generate corresponding business revenue. Launches, complete spacecraft, components, and services each involve different execution processes. The businesses driving group growth also determine its subsequent delivery and funding requirements.\n2 Major Mission Contractors and New Entrants On December 19, 2025, the U.S. Space Development Agency (SDA) announced four agreements totaling approximately $3.5 billion to deliver and operate 72 Tranche 3 Tracking Layer satellites. Teams led by Lockheed Martin, Rocket Lab, Northrop Grumman, and L3Harris are each responsible for 18 satellites, with launches planned for U.S. fiscal year 2029.[5]\nThis procurement establishes identifiable demand and shows that new entrants can participate in the same program as established contractors. Competition centers on the ability to deliver complex missions, encompassing design, manufacturing, testing, and operating responsibilities. Winning an agreement creates work to be performed; cost control and timely execution determine whether that work produces business results.\nThe approximately $3.5 billion covers satellite delivery and operations, a scope extending beyond hardware manufacturing alone. This is a specific U.S. defense procurement, serving a different market from civilian broadband demand. Public customers can support initial orders, while companies must also manage budget arrangements, contractual scope, and acceptance requirements.\n3 Network Operators and Service Channels Eutelsat combines its geostationary orbit (GEO) business with OneWeb\u0026rsquo;s low Earth orbit (LEO) network in a multi-orbit approach. Starlink and Iridium also provide services addressing broadband demand and voice, data, and Internet of Things requirements, respectively.[6][20] Orbital architecture affects how connections are provided, while service competition also involves terminals, sales channels, ground networks, and customer support.\nOperators must match regional demand with available capacity. Business customers may place greater importance on network integration and agreed service levels, while individual users express their preferences through their experience and renewal decisions. Expanding coverage can create sales opportunities, but revenue depends on capacity being purchased and used. The costs of acquiring and serving customers also affect whether that revenue produces a profit.\n4 Specialist Component and Engineering Suppliers On March 12, 2025, Rocket Lab announced an Airbus contract to provide 200 solar panels for 100 satellites in Eutelsat\u0026rsquo;s next-generation OneWeb program. The announcement did not disclose the contract value.[7] This is a verifiable procurement relationship: the operator defines network requirements, the satellite manufacturer organizes spacecraft delivery, and the component supplier provides specific products.\nSpecialist suppliers can participate in multiple programs through different spacecraft manufacturers without operating a complete network themselves. Their competitive strengths may include product suitability, flight heritage, consistent quality, and delivery capabilities, while production volumes vary with customers\u0026rsquo; program schedules. Qualification records establish a basis for participating in procurement, and continuing orders depend on subsequent demand. The announcement of 32 satellite deliveries in October 2026 and the 2025 solar panel contract document different stages; the available announcements do not establish that the latter contract has been fully completed.[7][23]\nV The Boundaries of Market Size and Company Revenue SIA\u0026rsquo;s public summary, released on May 13, 2026 and based on research conducted by BryceTech, estimates global commercial satellite industry revenue at approximately $303 billion in 2025. The principal segments listed are ground equipment at $165.2 billion, satellite services at $105.0 billion, satellite manufacturing at $20.4 billion, and commercial launch at $12.4 billion. This report uses the public summary, and the figures are rounded.[4]\nThese figures show that commercial satellite industry revenue is not concentrated primarily in rocket launches. Ground equipment and established satellite services account for a substantial share. The $303 billion includes a range of mature businesses, while the products and customers that individual companies pursue differ. It is also not a separate market estimate for the space-related advanced engineering activities discussed in this report.\nMarket size must be matched to a company\u0026rsquo;s actual products. Component suppliers address manufacturers\u0026rsquo; procurement requirements, launch providers compete for transportation missions, and operators sell network capacity or services. Group revenue at companies spanning several segments includes multiple businesses, so revenue corresponding to a particular market must first be separated before calculating market share. This report compares competitive approaches by business role; the representative companies listed are not ranked by market share.\nREVENUE MIX · Full year 2025\nGlobal commercial satellite industry revenue mix Commercial satellite industry under the SIA / BryceTech definition, including ground equipment, satellite services, satellite manufacturing and commercial launch. This is not company market share or the total market for commercial space and advanced engineering.\nWorldwide · Share of commercial satellite industry revenue (%)\nFull screen View the data table Global commercial satellite industry revenue mix · Full year 2025 · Share of commercial satellite industry revenue (%) Industry segmentShare Ground equipment54.5%Satellite services34.7%Satellite manufacturing6.7%Commercial launch4.1% Download data (CSV) Public-summary figures: approximately US$303 billion in total, comprising US$165.2 billion of ground equipment, US$105.0 billion of satellite services, US$20.4 billion of satellite manufacturing and US$12.4 billion of commercial launch. Each share = segment revenue / 303 × 100, rounded to one decimal place. The displayed shares total 100.0%; there is no residual Other category. The source revenue values are rounded and include mature ground equipment and service businesses. This composition is not a company ranking, an estimate of an individual entrant’s addressable market, or a separate size estimate for space-related advanced engineering. The paid full report was not obtained. Sources were verified through October 3, 2026; this is a historical 2025 snapshot.\nSource: Satellite Industry Association / BryceTech — 29th Annual State of the Satellite Industry Report · public summary (2026-05-13). Reviewed 2026-10-03.\nVI Rocket Lab Financial Disclosures Rocket Lab\u0026rsquo;s 2025 annual report disclosed approximately $602 million in annual revenue, including approximately $199 million from launch services and $403 million from space systems. Calculated from the disclosed figures, space systems accounted for approximately 66.9% of the company\u0026rsquo;s annual revenue. This is the company\u0026rsquo;s own revenue mix. Space systems includes spacecraft, components, and related services, a broader scope than component sales alone.[3][9]\nThis example shows that a commercial space company can develop multiple sources of revenue. Alongside Electron launches, Rocket Lab already has a substantial space systems business. Spacecraft procurement, component sales, and launch services involve different execution cycles and cost arrangements. The work driving growth also determines the production, delivery, and funding commitments the company must organize.\nThe company disclosed approximately $1.847 billion in backlog at the end of 2025, excluding unexercised customer options under its definition.[3][9] These orders describe work still to be performed and provide visibility into future business, while remaining subject to execution and termination provisions. The $602 million represents revenue recognized in 2025; cash and profit each have their own accounting measures. The stage of contract execution, payment arrangements, and costs incurred determine subsequent business results.\nRevenue records business activity recognized under accounting rules. Profitability and funding conditions must also be assessed through gross profit, research and development spending, operating expenses, and cash flow. Rocket Lab provides a public financial case that can be examined in detail. Its revenue mix, losses, and customer concentration describe this particular company; other commercial space businesses require assessment in the context of their own operations.\nREVENUE MIX · Full year 2025\nRocket Lab revenue mix Rocket Lab consolidated revenue for the year ended December 31, 2025. Space Systems includes spacecraft, systems, components and related services. This is one company’s revenue composition, not industry market share or component revenue alone.\nWorldwide company revenue · Share of Rocket Lab revenue (%)\nFull screen View the data table Rocket Lab revenue mix · Full year 2025 · Share of Rocket Lab revenue (%) Business segmentShare Space Systems66.9%Launch Services33.1% Download data (CSV) Original annual-report figures are in thousands of U.S. dollars: Space Systems 402,757; Launch Services 199,042; total 601,799. Converted to millions, these are US$402.757 million, US$199.042 million and US$601.799 million. Shares = each segment / 601,799 × 100, rounded to one decimal place, totaling 100.0%. Revenue is distinct from profit, backlog and cash receipts. These figures describe Rocket Lab and do not rank commercial space companies. Sources were verified through October 3, 2026; the chart does not extend that research cutoff.\nSource: Rocket Lab / U.S. Securities and Exchange Commission — Rocket Lab 2025 Form 10-K (2026-02-26). Reviewed 2026-10-03.\nPart 4 Industry Challenges The challenge in commercial space is turning a technical success into a business that can keep delivering. Development, manufacturing, launch and operations each incur costs, while customer payments may not arrive at the same time as those expenditures. Following a project\u0026rsquo;s progression reveals the interdependence of delivery, funding, regulatory access and long-term operations, as well as the areas companies can improve.\nINDUSTRY MAP\nFrom engineering capability to sustained business Separate technical validation, customer orders, execution, delivery, revenue and cash when assessing a commercial space business.\nFit to view Expand all Collapse Full screen Read the full text outline Business pathway1 · CapabilityOperating questionsTest evidence has defined limits First-stage recovery ≠ reflight Demonstration ≠ proven profit 2 · ProcurementOperating questionsCustomer · mission or service Scope, price, timing, acceptance Backlog ≠ cash; work remains 3 · ExecutionOperating questionsFund development through testing Fixed-price overruns can cut profit Supply, tests and launch readiness 4 · DeliveryOperating questionsDelivery ≠ launch or commissioning Launch = agreed transport outcome Networks and data must meet needs 5 · Revenue / cashOperating questionsRevenue · performance and rules Cash · agreed payment schedule Supplier/customer timing may differ 6 · ContinuityOperating questionsRepeat orders, renewals, delivery Execution, overhead, development costs Licensing, upkeep and disposal Based on the accompanying report and its evidence appendix, with sources verified through October 3, 2026. Diagram preparation does not extend that research cutoff. DEX editorial synthesis of the operating mechanisms in Parts 1, 3 and 4. Branches organize questions and do not imply a universal accounting or payment sequence. Technical capability, signed orders, recognized revenue, cash receipts and profit are different measures; none can substitute for the others. Licensing conditions and contractual terms depend on the country and mission.\nSources: [1] SES · evidence of first-stage reflight · [3] Rocket Lab · 2025 Form 10-K · [9] Rocket Lab · 2025 financial results · [13] Airbus · batch manufacturing · [17] Airbus · testing and verification · [23] Airbus · manufacturing delivery status · [24] NASA · funding and milestone mechanisms · [10] ESA · lifetime operating responsibilities. Reviewed 2026-10-03.\nI Cash May Run Out Before Development Is Complete New models require upfront spending on personnel, equipment, procurement and testing, while revenue is generally recognized only when contractual requirements are met. Anomalies during testing may require design changes, replacement orders for parts and further verification. As the engineering timeline lengthens, funding must continue to support the team. Receiving development support does not ensure that a company can sustain itself until the next payment.\nNASA\u0026rsquo;s COTS history offers a concrete lesson. After becoming a partner in the first round, Rocketplane Kistler failed to secure sufficient capital, and NASA terminated its agreement in October 2007. The case shows that a project\u0026rsquo;s continuation depends not only on the technical plan, but also on whether the company can raise and maintain funding before payment milestones are reached. [24]\nRocket Lab\u0026rsquo;s 2025 financial statements show a full-year consolidated net loss of approximately $198 million and a net operating cash outflow of approximately $166 million. [3][9] The former reflects the consolidated earnings result; the latter reflects actual cash movements from operating activities. Each has its own accounting basis, and neither establishes the independent profit or loss of every business segment. As a company expands, revenue growth and the funding needed to operate must be planned together.\nDevelopment and funding plans need to advance together. Staged verification, funding reserves for testing anomalies and customer payment terms that align as closely as possible with expenditure can ease cash pressure. Development failure and interruptions in financing remain possible risks. These risks can also reach suppliers, whose collection of payments after delivery may be affected by their customers\u0026rsquo; financial condition.\nII Reuse Does Not Mean Fixed Costs Have Been Spread Across Enough Missions First-stage recovery and reflight demonstrated the possibility of using hardware for further missions. [1] The benefits of reuse also depend on inspection and maintenance work, component life, turnaround time and the number of missions actually flown. Retaining hardware can reduce some repeated manufacturing, while recovery facilities, personnel, transportation and inspections add investment. Both sides determine the resulting savings.\nFactories, launch facilities and operating teams incur fixed costs. If there are too few missions, or if a rocket spends too long waiting for customer payloads, sites and licenses, those costs continue but must be spread across fewer missions. Manufacturing capacity needs to match available launch opportunities and actual mission volumes to reduce the fixed cost borne by each mission.\nLaunch companies can shorten preparation time for repeated tasks, make designs easier to inspect and maintain, and coordinate payload and site arrangements. Customers, in turn, decide whether to purchase a dedicated launch based on the business value of reaching orbit on schedule and having a choice of orbit. Whether improvements produce a long-term cost advantage depends on delivery reliability, turnaround resources and execution results across ongoing missions. A published price, by itself, reflects only the customer\u0026rsquo;s purchase price.\nIII Contract Value Revenue and Cash Remain Separated by Execution Fixed-price contracts establish the scope of work and price in advance, leaving the company to bear changes in execution costs. In its 2025 Form 10-K, Rocket Lab disclosed that most of its sales that year came from fixed-price contracts and that unreimbursed cost overruns affect profitability. Estimates of costs to complete are particularly uncertain for contracts that include development work. [3]\nThis risk can spread through the supply chain. Prime contractors must manage schedules and customer requirements, while component suppliers must meet quality and delivery conditions. Additional testing, delays or repeat procurement can increase expenditure without a corresponding increase in contract value. Clarifying the scope of work, change procedures and acceptance criteria before signing, then continually updating cost estimates during execution, can help reduce misjudgments. More accurate pricing also requires sufficient engineering and procurement information.\nRocket Lab\u0026rsquo;s backlog at the end of 2025 was approximately $1.847 billion, representing work still to be performed under enforceable agreements and excluding unexercised customer options. [3] Subsequent revenue requires performance and recognition under the applicable rules, while actual cash receipts depend on payment arrangements. Expected revenue recognition dates describe plans for future execution and differ from revenue already earned or payment deadlines.\nThe company also disclosed that launch and spacecraft build contracts typically allow customers to terminate with advance notice and payment of a termination fee. An unnamed government customer accounted for 28% of total revenue in 2025, showing the importance of a single customer. The disclosure did not identify the specific agency, and the figure does not represent all government revenue. [3] Order quality therefore depends on cancellation conditions, customer funding, acceptance and payment progress. Diversifying customers can reduce concentration risk, but developing business with new customers still takes time and may not promptly fill the revenue gap left by changes in a large project.\nIV Capacity Expansion Must Keep Pace with Quality and Supply Chain Capabilities Volume manufacturing requires every subsequent batch to meet design and interface requirements. A delay in a single component can affect the assembly of a complete satellite, while a design change may require existing parts to be verified again. After expanding equipment and personnel, a company must also establish adequate qualified supply and testing capabilities before added capacity can translate into products delivered on schedule.\nThe OneWeb production system described by Airbus includes suppliers providing components in volume and a high-speed assembly line. Airbus\u0026rsquo;s testing services also cover vibration, acoustics, thermal vacuum and electromagnetic compatibility. These businesses illustrate the supply and verification work behind volume assembly. [13][17] The factory capacity announced by Thales Alenia Space describes potential production capability; actual output still depends on orders and production execution. [17]\nExpansion needs to bring together identified demand, qualified suppliers and testing capabilities. Recording manufacturing batches, managing design changes, verifying interfaces early and assessing alternative sources for critical parts can help reduce rework and delays. Alternative parts may require further verification, and excess inventory also ties up cash. Quality, schedules and funding therefore need to be coordinated within the same production plan.\nV The Chinese Market Must Complete the Transition from Policy to Sustained Delivery China\u0026rsquo;s 2024 Government Work Report identified commercial space as one of the new growth drivers. [8] On July 12, 2023, LandSpace\u0026rsquo;s Zhuque-2 Y2 successfully reached orbit, demonstrating progress in liquid oxygen–methane launch technology by a private company. [14] Policy provides a direction for development, and mission success provides engineering evidence. Company orders and profitability still need to emerge through subsequent procurement and execution.\nIndustrialization requires turning technical capabilities into customer purchases, production, testing, missions and cash receipts. Launch companies need to conduct missions consistently, satellite companies need reliable deliveries in volume, and suppliers need to align added capacity with actual demand. Participation in a project can provide opportunities for verification; long-term procurement and capacity utilization depend on whether subsequent business continues.\nCompanies can first verify their capabilities around defined missions, then arrange equipment and personnel investment according to orders. Mission records, contract scope, repeat purchases, actual deliveries and financial condition progressively reveal progress toward industrialization. Customer budgets, acceptance and launch conditions may still change. The effect of policy ultimately depends on whether the conditions for development support the continuation of concrete business activity.\nVI Rockets Spectrum and Target Markets Have Different Access Requirements Once engineering preparations are complete, a mission must still meet applicable licensing requirements. The FAA states that U.S. commercial launch or reentry operators conducting activities abroad, and non-U.S. commercial launch or reentry vehicles operating within the United States, fall within its licensing requirements. Activities abroad may also need to meet host-country rules. An FAA announcement in March 2026 explained the transition of launch and reentry vehicle licensing to the Part 450 framework. That framework concerns vehicle licenses; other applicable approvals still need to be obtained separately. [26]\nCommunications businesses also involve national licensing and international spectrum coordination. The ITU\u0026rsquo;s official explanation states that national administrations license satellite systems and participate in network coordination, notification and the registration of frequency assignments under the Radio Regulations. [27] International coordination and registration primarily address matters such as radio interference. These are separate from launch licensing and the physical occupation of orbits.\nLaunch companies, constellation operators and terminal service providers need to arrange applicable approvals and operating conditions according to the country, frequency band and mission. Incorporating these requirements into design and delivery plans early can reduce waiting after engineering work is complete. Filed plans, actual deployment and the lawful provision of services are different stages in a project\u0026rsquo;s progression. Once a network has technical coverage, access to a target market still requires compliance with local conditions.\nVII Responsibilities Across the Satellite Life Cycle After a constellation\u0026rsquo;s initial deployment, operators still need to monitor orbits, avoid collisions, address failures and replenish satellites. ESA\u0026rsquo;s Space Environment Report 2026 is based on data through the end of 2025 and states that end-of-life disposal remains insufficient to ensure long-term sustainability. [10] The problem documented in the report requires operators to extend their responsibilities across the entire satellite life cycle.\nFor relevant satellite systems authorized by the FCC or seeking access to the U.S. market, the FCC has a rule requiring disposal no later than five years after the end of the mission for applicable spacecraft that end their missions in, or pass through, the low Earth orbit region below 2,000 kilometers and use uncontrolled atmospheric reentry for disposal. Transitional arrangements and case-by-case waivers are also available. [25] This requirement has specific applicability conditions and jurisdictional scope, which operating plans need to take into account when determining disposal responsibilities.\nOperators and suppliers need to incorporate collision avoidance capability, propellant, redundancy against failures, disposal plans and funding for constellation replenishment into design and budgets early. ESA\u0026rsquo;s scenario modeling indicates that debris generated by collisions could continue to increase even if new launches cease. This is a simulation result based on specific assumptions. [10] In-orbit services therefore have corresponding technical demand, while the profitability of companies providing them still depends on customer procurement, service delivery and costs.\nI pay closer attention to companies that can complete missions repeatedly, deliver reliably and receive payment. They need to control execution costs while also taking responsibility for licensing and in-orbit operations. Technical progress gives a business capabilities; customer demand and sustained execution determine whether those capabilities can support it over the long term. This is the starting point for how I understand the opportunities in commercial space and advanced engineering.\nAppendix Supporting Evidence and Sources Sources verified through October 3, 2026. The materials below support historical, business, order, financial, and regulatory facts in the article. Discussions of profitability mechanisms and competitive advantages are the author’s analysis. Company product pages document publicly described activities; specific supplier relationships are supported by contract announcements. Retrieval dates are noted for dynamic pages without clear publication dates. Annual statistics, historical announcements, and future plans retain their original time frames.\n[1] NASA and SES Falcon 9 First-Stage Recovery and Reflight NASA\u0026rsquo;s Astronomy Picture of the Day entry dated December 28, 2015, describes the controlled landing of a Falcon 9 first stage on land during the previous week, while the second stage continued deploying communications satellites. SES published its full-year 2017 results on February 23, 2018. The \u0026lsquo;Future satellite capacity and fleet update\u0026rsquo; section on printed page 11 states that SES-10 launched on March 30, 2017, using a previously flown Falcon 9 first stage.\nThe NASA source directly supports a landing in December 2015 and the mission type; the article does not use it to infer an exact day or payload count. The SES source identifies a customer and a reflight mission. Neither establishes reuse of the entire rocket, the first reuse of any spacecraft, or long-term profitability.\nNASA APOD landing entry · SES 2017 results PDF\n[2] NASA COTS Development and Commercial Resupply Services NASA\u0026rsquo;s COTS program summary and Commercial Resupply Services overview support the history of commercial transportation development beginning in 2006 and subsequent purchases of space station resupply services. The program summary was published in 2013 and describes the development and demonstration stage at that time. The resupply page explains NASA\u0026rsquo;s role as a transportation customer. Legal arrangements and payment mechanisms are detailed in [24].\nCOTS development support and CRS service procurement were different arrangements that overlapped in time. The first CRS awards did not wait until the demonstrations were complete. The U.S. example should not be presented as a common institutional model for every country.\nNASA COTS program summary · NASA commercial resupply overview\n[3] Rocket Lab 2025 Form 10-K Rocket Lab\u0026rsquo;s annual report filed with the SEC covers the year ended December 31, 2025, and was signed on February 26, 2026. Item 1 describes launch, space systems, and component businesses. Item 1A discusses fixed-price execution risk. Financial notes 20 and 21 disclose business segments and customer concentration, respectively; the report also includes backlog and cash flow information. Sales in 2025 were predominantly fixed-price. Cost overruns can reduce profitability, and development estimates carry uncertainty. This risk disclosure does not establish that every contract lost money.\nThe report supports analysis of the company\u0026rsquo;s business structure and operating challenges. Space Systems includes spacecraft and systems, rather than components alone. One government customer accounted for 28% of revenue; that customer cannot automatically be identified as SDA, and the percentage is not the share of all government revenue.\nSEC annual report\n[4] SIA Public Summary of the 2026 Satellite Industry Report On May 13, 2026, SIA released the 29th edition of its report, produced by BryceTech. The public summary covers 2025. Commercial satellite industry revenue was approximately US$303 billion. Ground equipment, satellite services, satellite manufacturing, and commercial launch generated approximately US$165.2 billion, US$105.0 billion, US$20.4 billion, and US$12.4 billion, respectively. See \u0026lsquo;Key Revenue and Industry Segment Takeaways\u0026rsquo; in the announcement.\nThis article uses the public summary; the paid full report was not obtained. These are industry revenues that include substantial ground equipment and established services. They are not a separate market estimate for commercial space and advanced engineering, nor an estimate of the market accessible to new entrants. Segment figures are approximate.\nSIA public summary\n[5] SDA Procurement of 72 Tranche 3 Tracking Layer Satellites The U.S. Space Development Agency announced four OTA agreements on December 19, 2025, with a combined value of approximately US$3.5 billion. Teams led by Lockheed Martin, Rocket Lab, Northrop Grumman, and L3Harris are each to deliver and operate 18 satellites, for a total of 72, with launches planned for U.S. fiscal year 2029. See the opening of the announcement and the list of four potential agreement values.\nThis is a specific defense program, not a measure of civilian industry growth. The figures are potential agreement values covering mission responsibilities, not cash received or unit prices for manufacturing alone.\nSDA procurement announcement\n[6] Eutelsat Integration of GEO and OneWeb Businesses Eutelsat issued a unified-brand announcement on September 4, 2025, describing the integration of its GEO and OneWeb LEO businesses within one corporate organization. It supports the discussion of a multi-orbit operating approach. Additional network and product sources appear in [20]. GEO and LEO refer here to geostationary orbit and low Earth orbit, respectively.\nBrand and organizational integration establish the company\u0026rsquo;s chosen approach. They do not establish that integration necessarily reduced costs, delivered superior performance across all services, or produced profitable synergies. Different orbits must be assessed against specific capacity, customer, and mission requirements.\nEutelsat unified-brand newsroom announcement\n[7] Rocket Lab Airbus OneWeb Solar Panel Contract Rocket Lab announced on March 12, 2025, that it would supply 200 solar panels for 100 Eutelsat OneWeb LEO satellites manufactured by Airbus. The panels include carbon composite substrates, solar cells, and photovoltaic assemblies, with production planned in Albuquerque. See the first four paragraphs. This is an identified procurement relationship connecting an operator, a satellite manufacturer, and a component supplier.\nThe announcement does not disclose the contract value, profit, or cash receipts. Two hundred panels do not mean 200 satellites. An order does not establish complete delivery or operation in orbit.\nRocket Lab contract announcement\n[8] China 2024 Government Work Report The Government Work Report was delivered on March 5, 2024, at the second session of the 14th National People\u0026rsquo;s Congress. It identified commercial space among new growth engines to be developed. The original draft used a Ministry of Education republication; a National Development and Reform Commission republication is also provided. See the discussion of emerging and future industries under building a modern industrial system in \u0026lsquo;Tasks for 2024.\u0026rsquo;\nThe policy statement supports an assessment of development priorities. It does not award a contract to any company discussed in this article. It cannot establish orders, deliveries, revenue, or profit. The report date and the date of a website\u0026rsquo;s republication must also be distinguished.\nFull text republished by the Ministry of Education · NDRC republication\n[9] Rocket Lab Full-Year 2025 Financial Results The February 26, 2026, financial release and the annual report in [3] show approximately US$602 million of 2025 revenue, a US$198 million net loss, and US$166 million of net cash used in operating activities. Year-end backlog was approximately US$1.847 billion. Note 20 of the annual report reports approximately US$199 million of launch revenue and US$403 million of Space Systems revenue. See the annual income and cash flow statements, backlog disclosures, and segment note. The annual report defines backlog as remaining work under enforceable contracts. Typical contracts allow termination with notice and a fee; expected revenue recognition is not a cash collection schedule.\nThe original tables are in thousands of U.S. dollars; the article converts and rounds the figures. Backlog excludes unexercised options and is neither revenue nor cash. Approximately 37% recognition within 12 months was a year-end expectation. The revenue mix is calculated from the original values in thousands: 402,757 ÷ 601,799 × 100% ≈ 66.9%. It is not industry market share. The release labels its financial tables unaudited; the annual report provides a cross-check.\nRocket Lab full-year financial release\n[10] ESA Space Environment Report 2026 ESA published the overview on September 14, 2026. The full report is Issue/Revision 10.1, issued September 8, 2026, with statistical data mainly through the end of 2025. It addresses congested orbits, collisions, and end-of-life disposal. Section 7.2 of the full report examines future scenarios, including the possibility of debris increasing even after new launches cease.\nThese are not full-year 2026 observations. Section 7.2 uses the DELTA-4 model, with 100 Monte Carlo simulations per scenario over a 100-year horizon. Results depend on assumptions about launches, explosions, disposal, and other factors; they are not certain forecasts. The report does not establish that debris removal is profitable. ESA project standards are not global law.\nESA report overview · ESA full report PDF\n[11] NASA Early Satellite Communications History NASA\u0026rsquo;s historical material records the launches of Telstar 1 on July 10, 1962, and Early Bird on April 6, 1965. The official Telstar retrospective was published July 10, 2012. See the beginning of that article and \u0026lsquo;The Billion Dollar Technology\u0026rsquo; and \u0026lsquo;The Global Village\u0026rsquo; in Communications Satellites for the transition from television demonstrations to international communications services. NASA also records the INTELSAT organizational agreement of August 20, 1964, adding context on international coordination and commercial communications.\nTelstar should not be described as the \u0026lsquo;first active communications satellite\u0026rsquo; without defining the claim. Historical prices, cumulative network figures, and references to what was \u0026lsquo;current\u0026rsquo; in these articles do not describe the industry\u0026rsquo;s position in 2026.\nNASA satellite communications history · NASA Telstar retrospective\n[12] Arianespace Dedicated Commercial Launch Services Arianespace\u0026rsquo;s 40th-anniversary retrospective states that the company was founded in 1980 and conducted its first commercial mission in May 1984, placing the Spacenet F1 communications satellite in orbit. See the opening and the paragraph on the first commercial mission. The source shows that specialized commercial launch services predated later entrants associated with NewSpace.\nThe page does not provide a verifiable exact publication date, so none is assigned. The article uses the mission month to avoid date differences across time zones. Cumulative mission numbers and the launcher lineup in the historical retrospective are not used to assess the current market.\nArianespace official retrospective\n[13] Airbus OneWeb Serial Manufacturing Airbus described OneWeb\u0026rsquo;s manufacturing process on February 8, 2022, including components supplied in batches, rapid satellite assembly, digital twins, automated transportation, and intelligent inspection. See the manufacturing-process and digital-tool sections. The source supports the discussion of organizational changes required to move from prototypes to serial production.\nThe stated rate of two satellites per day was a production capacity or speed disclosed at that time, not a continuing delivery rate for 2026. Cost claims without a clear comparison baseline and promotional claims about effects on astronomy are not treated as facts in this article.\nAirbus manufacturing feature\n[14] CNSA Zhuque-2 Y2 Reaches Orbit The China National Space Administration announced on July 12, 2023, that LandSpace\u0026rsquo;s Zhuque-2 Y2 liquid-oxygen/methane launch vehicle successfully delivered its payload to the intended orbit. This is a historical example of developing Chinese commercial launch capabilities. The event date and propellant choice come from the official announcement. Current LandSpace products and activities are covered in [18].\nThe evidence establishes successful orbital delivery, not reusability, lower unit costs, or company profitability. That Zhuque-2 mission must not be conflated with Zhuque-3 recovery designs or targets for later vehicles.\nCNSA mission announcement\n[15] Hexcel and Beyond Gravity Materials and Structures Hexcel\u0026rsquo;s Defense \u0026amp; Space page lists composite applications in launch vehicles and satellites. Beyond Gravity lists payload fairings, interstage structures, payload adapters, and separation systems. See Hexcel\u0026rsquo;s \u0026lsquo;Launchers, Missiles \u0026amp; Space\u0026rsquo; and \u0026lsquo;Satellites\u0026rsquo; sections and Beyond Gravity\u0026rsquo;s launcher product description. These sources support the distinction between supplying materials and supplying engineered structures.\nThe pages do not state clear publication dates; they were retrieved on October 3, 2026. Product catalogs do not establish that other companies in this article are customers. Claims of leadership and cumulative program counts cannot be converted directly into commercial space market share or customer numbers.\nHexcel space materials · Beyond Gravity business overview\n[16] Moog and Rocket Lab Propulsion Avionics and Attitude Control Moog\u0026rsquo;s product page documents propulsion, fluid control, avionics, power, and mechanisms. Rocket Lab\u0026rsquo;s reaction wheel page describes speed, angular momentum, and torque commands. Version 2 of the ST-16HV datasheet, dated April 11, 2023, lists a star catalog, processor, and attitude outputs. JPL\u0026rsquo;s SWOT material explains determining attitude through stellar observations, controlling pointing with reaction wheels, and adjusting orbit with thrusters.\nThese sources distinguish sensor measurements from actuator commands. Data for one model should not be generalized to every product, and design targets should not be presented as universally demonstrated performance. Attitude control and orbital maneuvering are different functions. Rocket Lab\u0026rsquo;s broader component portfolio is covered in [3].\nMoog space products · Rocket Lab reaction wheels · ST-16HV datasheet · NASA JPL: SWOT spacecraft\n[17] Airbus and Thales Alenia Space Testing and Integration Airbus Test Services lists vibration, acoustic, shock, thermal-vacuum, EMC/RF, and AIT-center services. Thales Alenia Space introduced its Rome Space Smart Factory on October 7, 2025, describing assembly, integration, testing, and modular cleanrooms. The product and factory materials support the distinction between testing services and a manufacturer\u0026rsquo;s internal spacecraft verification.\nTesting assesses compliance with requirements under specified conditions; it does not guarantee no future failures. Figures such as capacity exceeding 100 satellites per year describe capability, not completed sales. Owning testing facilities does not establish high utilization, stable cash flow, or high profitability.\nAirbus testing services · Thales Alenia Space factory announcement\n[18] SpaceX Rocket Lab and LandSpace Launch Services Electron\u0026rsquo;s official product page describes dedicated small-satellite launches, orbit and schedule options, and Kick Stage deployment. LandSpace\u0026rsquo;s website describes Zhuque research, manufacturing, testing, and launch activities. The article uses the 2025 Falcon User\u0026rsquo;s Guide: its changelog is dated March 2025, while the filename includes 2025-05-09. Section 3.6, printed page 14, covers multiple payloads, dedicated rideshare missions, and third-party dispensers.\nThese materials establish differences between transportation offerings, not rankings. Maximum advertised payload capability varies with orbit and mission conditions and should not be treated as capacity under a common recovery profile. One test does not establish mature reuse, and advertised starting prices are not complete costs for every mission.\nRocket Lab Electron · LandSpace products · SpaceX Falcon User’s Guide\n[19] Airbus Thales and Rocket Lab Spacecraft Platforms Airbus\u0026rsquo;s constellation page offers complete satellites and platform-only solutions. Thales describes communications, observation, and other systems. The \u0026lsquo;Assembly, Integration \u0026amp; Test\u0026rsquo; section of Rocket Lab\u0026rsquo;s spacecraft page lists spacecraft-level thermal-vacuum, vibration, and electromagnetic compatibility facilities. Chapter 2, Section 2.1 of NASA\u0026rsquo;s small spacecraft technology review describes a bus\u0026rsquo;s support functions, including power, thermal control, attitude control, and communications, and distinguishes bus procurement from payload hosting.\nThe sources clarify bus and payload responsibilities. Purchasing a bus does not eliminate interface or environmental verification. A bus order should not be assumed to include the customer\u0026rsquo;s payload or a complete mission. NASA\u0026rsquo;s market review is a January 2026 snapshot, and its provider list is not a market-share ranking.\nAirbus constellation solutions · Thales Alenia Space activities · Rocket Lab spacecraft · NASA complete spacecraft platforms review\n[20] Starlink Eutelsat and Iridium Networks and Connectivity SpaceX\u0026rsquo;s Starlink 2024 progress report documents its LEO broadband business. Iridium\u0026rsquo;s services and Network pages describe voice, data, IoT, intersatellite links, and operational support. Eutelsat launched ADVANCE managed connectivity on June 24, 2021, including capacity, terminals, and network interconnection. Its multiyear distribution agreement with NEC XON, announced November 15, 2023, specifies installation, training, bandwidth, and service-level arrangements. Both announcements come from Eutelsat\u0026rsquo;s official newsroom and support the discussion of product scope and channel responsibilities.\nA company can appear in both the midstream and downstream sections without its revenue being counted twice. A service menu does not establish availability in every region or paying customers. Residential broadband and narrowband IoT are different businesses. Altitudes and performance figures on older pages are not specifications for every newer model.\nStarlink 2024 progress report PDF · Iridium services · Iridium network · Eutelsat ADVANCE service announcement · Eutelsat–NEC XON distribution agreement\n[21] Planet BlackSky and ICEYE Observation and Analysis Planet\u0026rsquo;s products and PlanetScope documentation describe optical multispectral imagery, analytics, and lighting requirements. BlackSky describes its constellation and Spectra tasking and analytics platform. ICEYE describes active SAR observations and Flood Insights, which incorporates third-party data, algorithms, and expert analysis. These sources support the distinction between optical imagery, SAR, and the delivery of application services.\nA revisit does not guarantee a usable, cloud-free observation, and image resolution does not establish the quality of a business conclusion. SAR complements optical observations at night and through clouds, but should not be described as free of every operating constraint. Advertised delivery speeds are not contractual guarantees for every mission or evidence of realized customer benefits.\nPlanet products · BlackSky corporate overview · ICEYE satellite data · ICEYE insurance solutions · PlanetScope technical documentation · ICEYE SAR data · ICEYE Flood Insights\n[22] NASA and Contractors Public and Research Missions NASA\u0026rsquo;s resupply page documents Dragon cargo transportation, while its Orion reference identifies Lockheed Martin as the prime contractor. Rocket Lab announced the Aspera award on May 14, 2025. NASA\u0026rsquo;s May 29, 2024, PREFIRE announcement identifies Blue Canyon as the CubeSat manufacturer, JPL as the spectrometer provider, Rocket Lab as the launch provider, and the University of Wisconsin-Madison as the data processor. PREFIRE\u0026rsquo;s stated scientific goal is to observe heat escaping from polar regions into space, making the research objective, equipment responsibilities, and contractors\u0026rsquo; scope separately traceable.\nThe sources distinguish platform, payload, transportation, and research roles. An award is not a completed launch. NASA is a customer agency rather than a competing company. Deep-space research missions do not establish a mature mass-consumer business, and different procurement scopes should not be treated as the same service revenue.\nNASA commercial resupply · NASA Orion reference · Rocket Lab Aspera award · NASA PREFIRE roles and launch announcement\n[23] Airbus Delivery of 32 OneWeb Satellites Airbus announced on October 2, 2026, that it had delivered 32 next-generation OneWeb LEO satellites to Eutelsat. They were being prepared for shipment from the Toulouse production facility for a subsequent launch in the United States. They form the first batch of a program for 669 next-generation satellites. See the first three paragraphs. The source distinguishes orders, spacecraft delivery, and network replenishment.\nAs of the announcement, the supported status is delivered and being prepared for shipment, not launched or providing services in orbit. The 669 satellites are a program total, not a completed quantity. The announcement also does not map individual spacecraft to the solar panel procurement batch in [7].\nAirbus October 2, 2026, delivery announcement\n[24] NASA 2014 COTS Final Report NASA/SP-2014-617, May 2014. Printed pages 12–14 and 20–23 explain Space Act Agreements and milestone payments. Pages 31–33 discuss termination after Rocketplane Kistler\u0026rsquo;s funding shortfall and the selection of Orbital. Pages 82–83 describe the advance CRS awards on December 23, 2008, their overlap with COTS demonstrations, and service procurement under the Federal Acquisition Regulation. This article uses the complete report and distinguishes the two arrangements.\nThe agreements required companies to contribute their own funding and bear overruns in meeting the milestones. Printed page 38 also records additional funding in 2010; fixed milestone amounts therefore did not mean the program budget could never change. Termination of the Rocketplane Kistler agreement in 2007 was a specific historical case and should not be generalized to every project.\nNASA COTS final report PDF · NASA Cygnus historical retrospective\n[25] FCC Five-Year Disposal Rule for Specified LEO Space Stations The official Federal Register notice, 89 FR 65217–65223, dated August 9, 2024, includes 47 CFR 25.283(e). For covered space stations ending their mission in, or passing through, the region below 2,000 kilometers and planning disposal through uncontrolled atmospheric reentry, disposal must occur as soon as practicable after mission end and no later than five years afterward. The relevant amendments took effect September 9, with compliance requirements beginning September 29.\nThe scope includes relevant FCC authorizations and access to the U.S. market by systems licensed outside the United States; it is not a rule for all satellites worldwide. Satellites already in orbit when the rule was adopted are exempt. Previously licensed but unlaunched systems have a two-year transition, and individual waivers are available. For maneuverable spacecraft, mission end is when collision-avoidance maneuvers can no longer be performed; for others, it is completion of the primary mission.\nFederal Register rule text\n[26] FAA Transition to Part 450 Vehicle Licensing The FAA\u0026rsquo;s March 17, 2026, announcement on streamlining commercial space licensing describes the transition of launch and reentry vehicle licenses to Part 450. The legacy-license transition deadline was March 9, 2026. A license can cover an appropriate set of operations, vehicle configurations, and sites. The International page further explains licensing scope for U.S. commercial launch or reentry operators outside the United States, and for relevant non-U.S. commercial launch or reentry vehicles operating within the United States.\nThis is not a global licensing system, and Part 450 does not govern every space activity. Launch-site licensing is addressed by other regulations. An FAA vehicle license is not a spectrum or downstream service authorization. A rule proposed in 2023 should not be presented as an effective regulation.\nFAA 2026 licensing announcement · FAA international jurisdiction overview\n[27] ITU Satellite Frequency Coordination and Registration ITU\u0026rsquo;s \u0026lsquo;Regulation of Satellite Systems\u0026rsquo; backgrounder was updated in September 2026. See \u0026lsquo;Policy and regulatory considerations\u0026rsquo; and \u0026lsquo;International coordination and registration.\u0026rsquo; National administrations license systems and participate in network coordination, notification, and registration of frequency assignments under the Radio Regulations. Relevant status is recorded in the MIFR, and applicable systems are subject to deployment milestones.\nThe international framework aims to prevent harmful radio interference; deployment milestones also discourage resource hoarding. These are distinct from domestic licensing and physical collision avoidance. ITU does not issue a launch license for each satellite or grant exclusive ownership of an orbital altitude. Frequency filings are not counts of operational spacecraft or paying users.\nITU official backgrounder\n[28] NASA 2014 CCtCap Crew Transportation Procurement NASA announced fixed-price CCtCap contracts with Boeing and SpaceX on September 16, 2014, with maximum potential values of US$4.2 billion and US$2.6 billion, respectively. They cover certification, at least one crewed test, and two to six missions after certification. The announcement and Commercial Crew Essentials explain that the companies own and operate the systems, while NASA reviews requirements and certifies safety.\nThe figures are potential values of contracts awarded in 2014, not recognized revenue or prices solely for operational flights. Commercialization did not remove safety verification. The 2017 return-to-flight target in the original announcement was a historical plan, not an achieved date.\nNASA 2014 CCtCap announcement · NASA Commercial Crew Essentials\n","date":"2026-10-04T00:00:00Z","image":"/post/commercial-space-advanced-engineering/cover.jpg","permalink":"/post/commercial-space-advanced-engineering/","title":"Commercial Space and Advanced Engineering Industry Report"},{"content":"How AI computing became an infrastructure business, who supplies the value chain, and what determines delivery and investment returns. Sources current to October 1, 2026. This report covers the chips, servers, data centers and cloud services used for AI training and inference. Financial figures retain each company\u0026rsquo;s reporting period and business scope. Supporting materials appear in a separate appendix outside Parts 1 to 4.\nAI services depend on a chain of suppliers and operators that turn equipment into usable computing capacity. This report follows an order through that chain, traces how the industry developed, identifies representative companies, and examines the constraints on delivery and investment returns.\nPart 1: Story In September 2026, Dell reported results for the fiscal quarter ended July 31. AI-optimized server revenue was $16.4 billion, new orders totaled $60.9 billion, and the quarter ended with a $95 billion backlog.[23] Revenue represents business recognized during the quarter, orders reflect newly booked demand, and backlog is the stock still awaiting delivery. Converting that backlog into revenue requires production and delivery to be completed.\nMuch has to happen between an order and a working installation. Server vendors need accelerators, memory and networking components to arrive, then must test the assembled systems. Customers need sites with suitable power and cooling. A delay at one stage can push back the deployment schedule. For the customer, receiving the equipment is only one step in the project.\nOnce a cluster is running, the operating questions change. Users care less about how many chips it contains than whether its answers are useful and arrive quickly enough. Service providers must handle more paid requests while meeting those expectations, because depreciation, electricity bills and staffing costs continue. The capabilities promised at purchase have to be delivered in everyday operation.\nAI infrastructure connects businesses that can otherwise look quite different. Chip design, manufacturing, system delivery and cloud services all respond to the same underlying demand. Understanding the industry means following an order through to the point where computing capacity is used and customers continue paying for it. The formation of this value chain begins with much earlier changes in research and engineering.\nPart 2: Industry History 1. From research questions to commercial experiments In 1943, McCulloch and Pitts proposed a mathematical model of neural activity. In 1950, Alan Turing examined machine intelligence and introduced the \u0026lsquo;imitation game\u0026rsquo;. The term \u0026lsquo;artificial intelligence\u0026rsquo; appeared in the 1955 proposal for the Dartmouth research project, which took place in the summer of 1956. The first question was whether ideas about intelligence could be turned into programs a computer could execute.[1]\nEarly programs made progress on well-defined problems but often failed when faced with the complexity of the real world. In the 1970s, disappointed research expectations were followed by cuts in funding. Expert systems, which rose to prominence in the 1980s, encoded domain knowledge as rules but were subsequently constrained by maintenance costs and limited adaptability. Specialized Lisp computers also faced competition from cheaper general-purpose workstations. Both downturns exposed the distance between a technical demonstration and a commercially sustainable application.[2]\nIn 1997, IBM\u0026rsquo;s Deep Blue defeated Garry Kasparov in a six-game match. It combined specialized search chips, parallel computing, evaluation functions and databases of chess games to solve a clearly defined problem. The principle of designing a system around a task can still be seen in later AI infrastructure.[3]\n2. GPUs and cloud computing improve access GPUs can perform many similar operations at once, making them well suited to much of the computation in neural networks. NVIDIA introduced the CUDA architecture in 2006 and made related development tools available in 2007. Researchers gained not only faster chips but also programming tools they could put to work. Around the same time, AWS launched its S3 storage and EC2 computing services in 2006, allowing developers to rent infrastructure.[4][5]\nAlexNet provided a clear example of the change in 2012. The individual network described in the paper took about five to six days to train on two GTX 580 GPUs. The team\u0026rsquo;s ensemble of models achieved a top-5 test error rate of 15.3% in the ImageNet competition, compared with 26.2% for the runner-up. This error rate measures the share of cases in which the correct category did not appear among the five highest-ranked answers. GPUs, labeled data, network design and training methods all contributed to the breakthrough.[6]\n3. Larger training runs change system design In 2018, OpenAI analyzed large training runs carried out since 2012. It found that the amount of compute used in the largest runs in its sample had doubled roughly every 3.4 months, increasing by more than 300,000 times overall. This was a historical change in resources devoted to training. It was not a growth rate for model capabilities, and it cannot simply be extrapolated to the present.[7]\nHardware and algorithms were both changing. NVIDIA introduced Tensor Cores with its Volta architecture in 2017 to accelerate matrix operations. Google\u0026rsquo;s first-generation TPU, deployed from 2015, was designed primarily for inference: applying an already-trained model to new inputs. The 2017 Transformer paper demonstrated greater training parallelism in its machine-translation experiments and provided an architectural foundation for later language models.[8][9][10]\nGPT-3, introduced in 2020, had 175 billion parameters. ChatGPT\u0026rsquo;s release in 2022 brought conversational models to a much wider audience. Demand acquired an additional dimension: systems had to handle a continuous stream of user requests as well as complete training runs.[11]\n4. Computing becomes a data center engineering problem Meta\u0026rsquo;s 2024 Llama 3 report described using up to approximately 16,000 H100 GPUs to train its 405-billion-parameter model and discussed networking, storage and fault recovery in detail. As the number of machines grows, communication delays and downtime become more costly.[12]\nCooling must keep pace with increasing system density. The GB200 NVL72 uses a liquid-cooled rack, while the H100 product range still includes air-cooled configurations. The choice depends on power consumption and deployment conditions. Software also changes how efficiently equipment is used: the PagedAttention study demonstrated that better GPU memory management could improve inference throughput under the conditions tested.[13][14]\nPower supply further constrains where a project can be built and when it can enter service. In its 2026 report, the International Energy Agency estimated global data center electricity consumption at approximately 485 terawatt-hours in 2025 and forecast approximately 950 terawatt-hours in 2030. These figures include non-AI uses. Operators must then answer a set of practical questions: when will the equipment go live, how many customers can it serve, and how long will it take to recover the investment?[15]\nPart 3: Value Chain INDUSTRY MAP\nThe AI computing infrastructure value chain Follow the capabilities needed to turn chips into usable computing services.\nFit to view Expand all Collapse Full screen Read the full text outline AI computing infrastructureUpstream · Design and componentsAccelerators and softwareGPUs, development tools and model execution Representative companiesNVIDIA AMD Huawei Cloud provider chipsSpecialized accelerators for cloud workloads Representative companiesAWS · Trainium Google · TPU Fabrication and advanced packagingWafer production and system integration Representative companiesTSMC ASE HBM memoryData capacity and transfer bandwidth Representative companiesSK hynix Micron Samsung Midstream · Systems and facilitiesNetworking and interconnectsCommunication between servers and computing nodes Representative companiesNVIDIA Broadcom Servers and integrationAssembly, testing and delivery of complete systems Representative companiesDell Supermicro Power and coolingPower distribution, thermal management and site readiness Representative companiesVertiv Schneider Electric Downstream · Computing servicesCloud computing servicesCapacity, service availability and cluster operations Representative companiesAWS Microsoft Azure Google Cloud CoreWeave Customer usesModel training Inference and business applications DEX editorial map based on Part 3 and the supporting appendix. Company examples are illustrative and may span several stages. Branches group business roles; they do not establish supplier contracts, market shares or an exhaustive list.\nSources: NVIDIA · computing and networking business · TSMC · CoWoS advanced packaging · Dell · AI server revenue and delivery pipeline · IEA · energy and infrastructure constraints. Reviewed 2026-10-01.\n1. Working back from the customer\u0026rsquo;s bill The value chain ends with the user of computing capacity. A business might rent GPUs by the hour or pay for model services according to usage. At sufficient scale, it may build its own cluster. Cloud providers collect service fees and spend a portion on servers, networking, electricity and operations. Server suppliers buy accelerators, memory and other components, while chip designers purchase manufacturing services from foundries and packaging providers. This is a simplified account of a typical division of work. Actual contracts may also involve large customers buying components directly and commissioning their integration.\nThe same end-customer demand generates revenue for several companies along the chain. Chip sales, server sales and cloud revenue therefore cannot simply be added together to calculate market size. When examining a particular stage, identifying the customer, the product or service delivered and the point at which revenue is recognized is more useful than starting with a single figure for the entire industry.\nThe table identifies representative companies by what they deliver. A company may participate at several stages; the list is not a market-share ranking.\nValue chain stage Representative companies Business role and sources Accelerators and software NVIDIA , AMD , Huawei NVIDIA supplies GPUs and CUDA; AMD offers Instinct products; Huawei provides the Ascend platform and CANN.[4][18][19][27] Cloud provider chips AWS , Google Develop Trainium and TPU respectively, configuring compute for internal and customer workloads.[21][22] Fabrication and advanced packaging TSMC , ASE TSMC provides wafer fabrication and CoWoS; ASE supplies advanced packaging including 2.5D/3D integration.[16][29] HBM memory SK hynix , Micron , Samsung Supply high-bandwidth memory, providing accelerators with data capacity and transfer bandwidth.[17][30][31] Networking and interconnects NVIDIA , Broadcom Supply data center networking products and chips that connect servers and computing nodes.[18][20] Servers and integration Dell , Supermicro Integrate accelerators and other components into AI servers, with air-cooled and liquid-cooled system formats.[23][32] Power and cooling Vertiv , Schneider Electric Supply power, thermal management and supporting data center infrastructure.[24][33] Cloud computing services AWS , Microsoft Azure , Google Cloud , CoreWeave Deliver computing capacity and related services to customers and operate the underlying clusters.[5][25][26][34] 2. Chip design and software Accelerator design determines how computing units, on-chip memory and interconnects work together. Software determines whether models can use those resources. CUDA offers one way to understand this relationship: development tools, libraries and accumulated code affect the engineering work required to move to different hardware.[4] On that basis, this report argues that a chip should be assessed not only on performance and price, but also on the time needed to migrate models, debug them and achieve stable operation.\nGeneral-purpose GPUs can accommodate changing models and customer requirements. Large cloud providers are also in a position to develop specialized chips for more clearly defined workloads. AWS announced general availability of Trainium3 UltraServers in December 2025. Google introduced the TPU 8t and 8i for different workloads in April 2026. The former was an availability announcement; the latter was a product announcement. Neither establishes how many systems have actually been deployed.[21][22]\nThis creates two different commercial choices: selling chips and using chips to provide services. Whether an in-house chip pays off depends on how widely development costs can be spread, as well as performance, utilization and operating costs after migration. Cloud providers have an incentive to reduce the cost of serving each unit of demand. External customers are more concerned with whether their workloads run successfully.\n3. Manufacturing advanced packaging and memory Once a chip is designed, it must pass through wafer fabrication, packaging and testing. Advanced packaging is taking on more work. TSMC\u0026rsquo;s CoWoS, for example, connects logic chips and high-bandwidth memory within the same packaging system. In its second-quarter 2026 earnings call, TSMC said packaging capacity remained tight. More wafer supply does not automatically mean complete accelerators can be delivered on time.[16]\nHBM is high-bandwidth memory placed close to the computing chip. Capacity determines how much data it can hold, while bandwidth affects how quickly data moves. One cannot substitute for the other. Micron\u0026rsquo;s published specifications for its 12-high HBM4 product, for example, list 36 GB of capacity and bandwidth exceeding 2.8 TB/s. These are component specifications; they cannot be converted directly into the amount of text a server generates per second.[17]\nConstraints at this stage travel through the supply chain. Delays in computing chips, HBM or packaging can each hold up system shipments. Capacity expansion, meanwhile, requires upfront spending. The corresponding additional revenue generally cannot be recognized until the new capacity is ready and products have been delivered. Assessing business conditions therefore requires looking at demand, construction progress and yields together, rather than relying on expansion announcements alone.\n4. Networks and server integration Large training clusters must exchange data between computing nodes while continuously reading training material from storage.[12] Assessing a system therefore requires measuring how much time is spent on communication and data access. A single chip\u0026rsquo;s peak computing performance does not capture these issues.\nServer vendors and system integrators combine accelerators, CPUs, memory, networking and cooling into deliverable products. They also handle testing, deployment and after-sales service. Revenue can grow rapidly, but the funding needed for procurement, inventory commitments and delivery obligations rise with it.\n5. Power cooling and cloud operations Data centers need power distribution, backup power, cooling equipment and routine maintenance. Buying servers does not by itself provide usable computing capacity: sites, grid connections and installation must also be ready. Vertiv supplies infrastructure including power and thermal-management systems. It reported approximately $3.274 billion in sales in the second quarter of 2026. That figure indicates the scale of one supplier\u0026rsquo;s business; it cannot all be classified as revenue from liquid cooling for AI.[24]\nOnce systems are operating, utilization and pricing become central. Training customers may occupy large amounts of equipment for a defined period, while inference services must respond to changing request volumes and latency requirements. Under the same requirements for answer quality and waiting time, the number of requests each machine can handle is a measure of efficiency closer to what operators need. Software optimization and standardized testing provide ways to make such comparisons.[14][28]\nThe operating economics can be expressed as a straightforward calculation: service revenue must cover equipment depreciation, financing, electricity, premises, networking and operating costs. Even as new chips become faster, older equipment may still serve less demanding tasks. But falling rental prices or low utilization will lengthen the payback period. End customers\u0026rsquo; willingness to pay determines whether expansion further up the chain can produce recurring revenue.\nPart 4: Industry Challenges 1. Supply constraints and delivery schedules Advanced manufacturing, packaging, memory and grid connections constrain different stages of delivery. TSMC\u0026rsquo;s disclosure of tight packaging capacity shows that constraints can arise after wafer fabrication. The power bottlenecks discussed by the IEA show that construction obstacles remain even after equipment leaves the factory.[16][15] This report\u0026rsquo;s assessment is that a shortage at a particular stage may give its suppliers greater pricing power, but that advantage must be reassessed as supply expands and customer demand changes.\nCapacity takes time to build, while chip generations and customer workloads can change during construction. A shortage today does not guarantee strong utilization once an expansion is complete. Buyers need to assess delivery dates and readiness across the whole project; suppliers need to judge whether demand will persist when their new capacity comes online.\n2. Revenue growth and investment returns The table below selects public figures from different stages of the value chain to show where demand is turning into revenue. Companies differ in fiscal periods, product scope and revenue recognition. These figures therefore do not rank shares of a single market. All amounts are in billions of US dollars and represent disclosed quarterly revenue; orders and forecasts are excluded.\nCompany / business Period ended Quarterly revenue ($bn) Scope and source NVIDIA Data Center 2026-07-26 89.0 Includes computing and networking; not GPU revenue alone.[18] AMD Data Center 2026-06-27 6.7 Includes EPYC CPUs and Instinct GPUs.[19] Broadcom AI semiconductors 2026-08-02 16.7 Includes custom accelerators and AI networking.[20] TSMC , company-wide 2026-06-30 40.20 Includes non-AI demand, such as smartphones.[16] Dell AI-optimized servers 2026-07-31 16.4 Recognized server revenue.[23] Vertiv , company-wide 2026-06-30 Approx. 3.274 Includes infrastructure and services for non-AI uses.[24] Google Cloud 2026-06-30 24.768 Includes revenue from GCP, Workspace and other products.[25] CoreWeave , company-wide 2026-06-30 2.575 Company revenue, including cloud services.[26] NVIDIA offers a view of revenue concentration within one company. In the quarter ended July 26, 2026, Data Center revenue was approximately $89.0 billion against total company revenue of $96.221 billion. Data Center therefore represented approximately 92.5% of NVIDIA\u0026rsquo;s revenue, with other businesses accounting for the remaining 7.5%. These calculated percentages describe NVIDIA\u0026rsquo;s revenue mix; Data Center includes computing and networking, and the figures do not measure AI revenue alone or NVIDIA\u0026rsquo;s share of the industry.[18]\nREVENUE MIX · FY2027 Q2\nNVIDIA revenue mix Quarter ended July 26, 2026. Data Center as a share of NVIDIA\u0026#39;s total quarterly revenue. This measures one company\u0026#39;s revenue composition, not its industry market share or AI-only sales.\nWorldwide company revenue · Share of NVIDIA revenue (%)\nFull screen View the data table NVIDIA revenue mix · FY2027 Q2 · Share of NVIDIA revenue (%) Business groupingShare Data Center92.5%All other businesses7.5% Download data (CSV) Calculated from approximately $89.0 billion in Data Center revenue and $96.221 billion in total revenue. Data Center share = 89.0 / 96.221 × 100, rounded to one decimal place; the other category is the residual to 100%. Data Center includes computing and networking and is not synonymous with AI. The source rounds Data Center revenue, so the percentages are approximate.\nSource: NVIDIA — Financial results for second quarter fiscal 2027 (2026-08-26). Reviewed 2026-10-01.\nDifferences in scale do not establish which business model is more profitable. Chip design, wafer fabrication, complete-system delivery and cloud operations carry different costs. Assessing the quality of a business also requires examining gross profit, capital expenditure, cash flow and asset utilization. In particular, business revenue that includes CPUs or enterprise productivity services cannot all be counted as AI accelerator or model-service revenue.\nContract value is no substitute for profitability. In the second quarter of 2026, CoreWeave reported revenue of $2.575 billion alongside a GAAP net loss of $0.626 billion. Its approximately $104 billion quarter-end revenue backlog included remaining performance obligations and estimated future revenue under other contracts. Realization remained subject to conditions including delivery and service availability.[26] These figures place growth and cost pressures on the same set of accounts.\n3. Platform choice and software migration The first approach is to build a hardware and software platform around accelerators. NVIDIA\u0026rsquo;s business now spans computing and networking, while AMD supplies both data center CPUs and accelerators. The second is for cloud providers to develop chips around their own workloads, then supply computing capacity to internal operations or external customers. Products from AWS and Google illustrate this approach. The third is to provide custom accelerators and networking chips for large customers. Broadcom\u0026rsquo;s disclosed AI semiconductor revenue falls into this category.[18][19][20][21][22]\nAll three approaches can appear on a single customer\u0026rsquo;s purchasing list. An established platform is attractive when models are still changing and there is little time for migration. Specialized designs become more attractive when workloads are stable enough and usage is sufficiently large. This report therefore expects competition to center on workloads, software compatibility and total cost of use. A single chip metric is unlikely to determine the outcome.\nCompetition in China’s market also involves software. Huawei\u0026rsquo;s documentation positions CANN, its heterogeneous computing architecture, between AI frameworks and Ascend hardware, covering functions such as compilation, operators and runtime execution.[27] Assessing a platform therefore requires looking beyond chip specifications to whether existing models can be migrated, which operators need rewriting and how easily the tools support debugging.\n4. Hardware specifications and actual service costs A useful system comparison starts with the same task. The model, precision, quality target and latency requirements need to be specified before throughput is compared. MLPerf’s inference benchmarks use defined scenarios and quality requirements for this reason.[28] A chip specification or a result from one workload cannot establish the operating performance of every deployment.\nFor operators, the next step is to connect those measurements to the bill. The same number of installed machines can produce different costs per completed request if utilization, downtime or software efficiency differs. Evaluation therefore needs to include deployment work, power, maintenance and the amount of capacity customers actually use.\n5. Whether paid usage can sustain expansion On the demand side, watch whether paid usage keeps pace with capacity additions. On the supply side, track whether advanced packaging, memory and power projects are delivered on schedule. For operating performance, examine utilization, cash flow and investment payback periods. These measures are connected: delivery delays can hold back real customer demand, while fully installed equipment can still face falling prices and low utilization.\nAI infrastructure now links model development with manufacturing, engineering and ongoing operations. Companies that deliver reliably, enable customers to use their products or services successfully, and cover costs while generating recurring cash flow at their own stage of the chain have an opportunity to turn a wave of purchasing into a lasting business. That is this report\u0026rsquo;s overall assessment of the value chain; actual delivery and financial results will be needed to test it.\nAppendix: Supporting Materials Source numbers correspond to the citations in the article. The appendix is separate from Parts 1 to 4. Citation scope and access notes were reviewed on October 3, 2026; the financial and statistical periods stated in the report are unchanged.\n[1] Early AI documents McCulloch \u0026amp; Pitts (1943) — A Logical Calculus of the Ideas Immanent in Nervous Activity\nTuring (1950) — Computing Machinery and Intelligence\nMcCarthy et al. (1955) — Dartmouth research proposal\nDartmouth — Artificial Intelligence Coined at Dartmouth\nLocation: McCulloch and Pitts, publisher bibliographic record and public abstract; Turing, pp. 433–434; opening of the Dartmouth proposal; Dartmouth history page. Original papers and institutional records. The McCulloch–Pitts full text requires subscription access; the citation here is limited to its metadata and abstract, not a claim that the full paper was inspected.\n[2] AI winters and expert systems Ted E. Senator (2026) Implications for AI Research: Applying Lessons from the Expert Systems Boom and Bust to the Current Large-Language Model Boom (PDF) Location: PDF pp. 1–2. A retrospective scholarly paper.\n[3] Deep Blue Campbell, Hoane \u0026amp; Hsu (2002) — Deep Blue Location: IBM paper abstract. An account by the project researchers.\n[4] GPUs and CUDA NVIDIA CUDA Programming Guide — Introduction\nNVIDIA CPU vs GPU — What’s the Difference\nLocation: guide introduction and official historical account. Distinguishes the 2006 architecture introduction from the 2007 software release.\n[5] Early AWS services AWS Our Origins Location: paragraphs on the launch of S3 and EC2. Official historical account.\n[6] AlexNet Krizhevsky, Sutskever \u0026amp; Hinton (2012) — ImageNet Classification with Deep Convolutional Neural Networks Location: Sections 3 and 6, Table 2. Training hardware refers to an individual network; the competition score is for an ensemble.\n[7] Historical training compute Amodei \u0026amp; Hernandez / OpenAI (2018) — AI and Compute Location: opening and Overview. Original 2018 analysis of its historical sample.\n[8] Volta and Tensor Cores NVIDIA (2017) NVIDIA Launches Revolutionary Volta GPU Platform\nNVIDIA Technical Blog (2017) Programming Tensor Cores in CUDA 9\nLocation: the May 10, 2017 announcement establishes the Volta launch and introduction of Tensor Cores. The October 17, 2017 technical article, in its original opening paragraphs, describes the matrix-multiply-and-accumulate function. The technical explanation is supported by the original technical-article body, not inferred from the launch release or the technical page’s AI-generated summary.\n[9] First-generation TPU Jouppi et al. (2017) — In-Datacenter Performance Analysis of a Tensor Processing Unit Location: paper abstract. Supports deployment in 2015 and its inference role.\n[10] Transformer Vaswani et al. (2017) — Attention Is All You Need Location: abstract and Section 4. Findings relate to the paper’s machine-translation experiments.\n[11] GPT-3 and ChatGPT Brown et al. (2020) — Language Models are Few-Shot Learners\nOpenAI (2022) — Introducing ChatGPT\nLocation: GPT-3 abstract and ChatGPT launch page. A model paper and the product announcement of 2022-11-30, respectively.\n[12] Llama 3 training cluster Meta Llama Team (2024) — The Llama 3 Herd of Models Location: Sections 3.3.1–3.3.4. An engineering example from a particular training project.\n[13] Cooling configurations NVIDIA — GB200 NVL72 product information\nNVIDIA — H100 product specifications\nLocation: GB200 NVL72 rack description and H100 Form Factor specification. Evidence for the named product configurations.\n[14] PagedAttention Kwon et al. (2023) — Efficient Memory Management for Large Language Model Serving with PagedAttention Location: abstract and evaluation. Throughput findings depend on the tested models, workloads and latency constraints.\n[15] Data center electricity IEA (2026) — Key Questions on Energy and AI: Executive Summary Institutional report, 2026-04-16; executive summary. 485 TWh is a 2025 estimate and 950 TWh a 2030 forecast; both include non-AI uses.\n[16] TSMC: fabrication and packaging TSMC — CoWoS technology\nTSMC — 2Q26 results presentation (2026-07-16)\nTSMC — 2Q26 earnings-call transcript\nLocation: technology page; presentation p. 4; call transcript p. 10. Revenue is company-wide; the packaging constraint is management’s assessment at that time.\n[17] HBM product specifications Micron — HBM4 Location: specifications for the 12-high product. Capacity and bandwidth are supplier component specifications, not measured server performance.\n[18] NVIDIA quarterly revenue NVIDIA — Q2 fiscal 2027 results (2026-08-26) Location: Data Center discussion and the quarter ended 2026-07-26. The revenue covers computing, networking and other products in that business.\n[19] AMD quarterly revenue AMD — Q2 2026 financial results (2026-08-04) Location: Data Center discussion; quarter ended 2026-06-27. The segment includes server CPUs and GPUs.\n[20] Broadcom quarterly revenue Broadcom — Q3 fiscal 2026 results (2026-09-02) Location: AI semiconductor revenue discussion; quarter ended 2026-08-02. Uses reported revenue, not guidance for the following quarter.\n[21] AWS custom silicon AWS — Amazon EC2 Trn3 UltraServers availability (2025-12-02) Official availability announcement. Supports availability, not deployment volume or market share.\n[22] Google TPU products Google — Introducing TPU 8t and TPU 8i (2026-04-22) Official product introduction. Supports workload specialization, not general availability or installed volumes.\n[23] Dell: revenue, orders and backlog Dell — Q2 fiscal 2027 financial results (2026-09-01) Location: AI server commentary and the quarter ended 2026-07-31. Revenue, orders and backlog are cited as separate measures.\n[24] Vertiv: power and thermal management Vertiv — Q2 2026 results, SEC Exhibit 99.1 (2026-07-29) Location: net sales and company description; quarter ended 2026-06-30. Figures are company-wide.\n[25] Google Cloud revenue scope Alphabet — Q2 2026 earnings release (2026-07-22)\nAlphabet — Form 10-Q, quarter ended 2026-06-30\nLocation: release segment-revenue table; Google Cloud in the 10-Q revenue-recognition discussion. Includes cloud services, subscriptions and product sales.\n[26] CoreWeave: growth and operating costs CoreWeave — Q2 2026 earnings release (2026-08-11) Location: highlights, revenue-backlog definition and income statement; quarter ended 2026-06-30. Net loss is GAAP; backlog is not cash received.\n[27] Ascend software architecture Huawei — CANN Community Edition 8.5.0 documentation Location: CANN architecture and functionality. Technical documentation, not comparable market-share evidence.\n[28] Conditions for inference comparisons MLCommons — MLPerf Inference: Datacenter Location: benchmark description, scenarios and quality requirements. Used for comparison principles, not vendor rankings.\n[29] ASE advanced packaging ASE — VIPack™ Location: VIPack overview and six packaging technology pillars, especially the passages on 2.5D/3D architectures and HBM interconnects.\n[30] SK hynix high bandwidth memory SK hynix Begins Volume Production of Industry’s First HBM3E (2024-03-19) Location: Opening paragraph of the 19 March 2024 release, the HBM definition and the discussion of AI processor–memory interconnections.\n[31] Samsung high bandwidth memory Samsung Semiconductor — HBM Location: HBM overview describing TSV stacking, AI training and HPC; used to establish business scope, not customer qualification or market share.\n[32] Supermicro GPU servers Supermicro — GPU Servers for AI, Deep / Machine Learning \u0026amp; HPC Location: The Liquid Cooled GPU Systems and Air Cooled GPU Systems categories under Supermicro GPU Servers.\n[33] Schneider Electric data center infrastructure Schneider Electric — AI Data Centers e-guide (2026-02-11) Location: Opening portfolio description on the guide download page; version 1.5, document 998-2372985_AI_Ready_DC.\n[34] Microsoft Azure accelerated computing Microsoft Azure — Virtual Machine series Location: The N Family — GPU accelerated virtual machines section, including the roles of the ND, NC and NV series.\nBrand icon sources and licenses are listed in the asset credits.\n","date":"2026-10-01T21:45:00+08:00","image":"/post/ai-computing-infrastructure/cover.png","permalink":"/post/ai-computing-infrastructure/","title":"AI and Computing Infrastructure: Industry History, Value Chain, and Challenges"},{"content":"Send a message to DEX Research. Corrections, collaboration, feedback, and inquiries are welcome. Get in touch Use the form below to send a message. I read every submission and will reply when appropriate.\nTypical topics:\nReport corrections or source suggestions Dataset feedback Research collaboration Sponsorship / partnership inquiries General questions about the methodology Name Email Message Send Message You can also reach me directly:\nEmail: dex222444@gmail.com Proton Mail: qizhangdong325@proton.me ","date":"2026-09-30T00:00:00Z","permalink":"/contact/","title":"Contact \u0026 Messages"},{"content":"An evidence-based guide to semiconductor history, design and manufacturing, market structure, AI demand, policy, and supply-chain risks. Scope and Data Notes In this report, the “semiconductor industry” includes chip design, electronic design automation (EDA) software and semiconductor intellectual property (IP), manufacturing equipment and materials, wafer fabrication, packaging and testing, and major end uses. Historical events are dated to when they occurred. Company financials, capacity, and process developments generally reflect information through December 31, 2025; policy information is updated through September 27, 2026. Citation scope and access notes were reviewed on October 3, 2026; this source review does not change the stated data cutoffs. Market-share figures are cited only when the source specifies the market boundary, time period, and measurement basis. Definitions of “foundry,” “AI accelerator,” and “advanced process” vary across organizations, so figures using different definitions should not be compared directly.\nExecutive Summary Semiconductors are not a single market. They are a cross-border industrial network spanning design tools, architecture and circuit IP, manufacturing equipment, critical materials, wafer fabrication, packaging and testing, and end systems. Development cycles, capital needs, and business models differ substantially across these activities. Software and IP businesses depend on research, ecosystem compatibility, and long-term licensing relationships. Equipment and materials require lengthy process qualification. Wafer fabrication relies on sustained capital investment, yield learning, and capacity utilization. Packaging and testing are expanding from conventional back-end services into system-level integration. A single “smile curve” or industry-wide gross-margin range cannot adequately describe this structure.\nThree forces have shaped the industry over the past eight decades. The first is progress in devices and manufacturing, from the transistor, integrated circuit, and planar process to FinFETs, gate-all-around transistors, and extreme ultraviolet (EUV) lithography. The second is a change in industrial organization. The rise of the dedicated foundry allowed design and manufacturing to be performed by different companies, supporting a specialized ecosystem of fabless designers, foundries, outsourced assembly and test providers (OSATs), and tool suppliers. The third is a shift in demand, from mainframes and consumer electronics to personal computers, mobile communications, cloud computing, automotive electronics, and today\u0026rsquo;s AI infrastructure.\nBy the end of 2025, AI training and inference demand was driving investment in advanced logic, advanced packaging, high-bandwidth memory (HBM), and high-speed interconnects. That growth, however, was not reaching every semiconductor category equally. Mature-node chips, analog and power devices, consumer memory, and industrial semiconductors remained subject to their own inventory cycles and end-market demand. Governments were also using incentives, research programs, and export controls to strengthen supply security. As a result, supply-chain decisions increasingly balance cost and scale against compliance, regional capacity, and resilience.\n1. How the Industry Developed 1.1 From Vacuum Tubes to Silicon Transistors Electronic computers of the 1940s relied heavily on vacuum tubes. Tubes could amplify and switch electrical signals, but their size, power consumption, heat, and limited service life constrained miniaturization and reliability. In 1947, a Bell Laboratories team developed the point-contact transistor, establishing solid-state devices as a promising alternative.H18\nEarly transistors were mainly made of germanium. In 1954, Morris Tanenbaum at Bell Laboratories produced a silicon transistor. At Texas Instruments, Gordon Teal organized the research laboratory and recruited a technical team led by Willis Adcock that developed commercial silicon transistors. These devices offered a wider operating-temperature range than germanium devices.H01 A later manufacturing advantage was the use of an adherent, electrically insulating oxide layer to separate surface interconnections, as described in Robert Noyce\u0026rsquo;s device-and-lead patent.H17\nIn 1957, eight engineers left Shockley Semiconductor Laboratory to establish Fairchild Semiconductor. Fairchild and the companies that grew out of it became an important part of Silicon Valley\u0026rsquo;s semiconductor startup network. The defensible conclusion is that this event accelerated the circulation of technical talent, venture capital, and new firms; it was not the sole origin of Silicon Valley\u0026rsquo;s entrepreneurial culture.\nOne of the key Bell Laboratories patents associated with the point-contact transistor is John Bardeen and Walter Brattain\u0026rsquo;s US 2,524,035, Three-Electrode Circuit Element Utilizing Semiconductive Materials.H15 Its patent grant date is distinct from the 1947 laboratory demonstration.\n1.2 Integrated Circuits, the Planar Process, and Moore\u0026rsquo;s Law Replacing vacuum tubes with individual transistors did not solve the problems of connecting large numbers of components or manufacturing them at scale. Integrated circuits and the planar process emerged in the late 1950s. Through oxidation, photolithography, diffusion, and metal interconnection, the planar process made it possible to form and connect multiple devices on the surface of a single silicon wafer. It laid the foundation for high-volume monolithic integrated circuits.\nEarly integrated circuits took different technical approaches. Jack Kilby\u0026rsquo;s relevant patent is US 3,138,743, Miniaturized Electronic Circuits. Robert Noyce\u0026rsquo;s is US 2,981,877, Semiconductor Device-and-Lead Structure.H16H17\nIn 1965, Gordon Moore used the limited data then available to predict that the number of components on an integrated circuit would roughly double every year for the next decade. In 1975, he revised the cadence to approximately every two years.H02 What became known as “Moore\u0026rsquo;s Law” was both an empirical observation and a reference point for coordinating technology roadmaps across design, equipment, materials, and manufacturing. It does not imply that the price of every chip automatically falls. Whether the cost per function declines also depends on die area, yield, design complexity, packaging, and utilization.\n1.3 Microprocessors, Memory Competition, and US–Japan Adjustments Intel introduced the 4004 in 1971 after developing it for a calculator.H03 It was a commercial four-bit microprocessor containing approximately 2,300 transistors, specifications given in Intel\u0026rsquo;s 50th-anniversary infographic.H20 Its significance lay in showing that a general-purpose programmable processor could be sold as a standardized product. The personal-computer market subsequently emerged through the combined development of eight- and 16-bit processors, memory, software, and complete computer systems. The 4004 alone did not “directly launch the PC era.”\nFrom the late 1970s through the 1980s, DRAM became a focal point of competition between Japanese and US companies. Japan\u0026rsquo;s Ministry of International Trade and Industry supported a VLSI research program, while manufacturers\u0026rsquo; production capabilities, quality control, and domestic electronics demand also contributed to their growth.H14 A historical study by the US International Trade Commission reports that Japanese firms\u0026rsquo; share of the global DRAM market rose from less than 30% in 1978 to nearly 75% in 1986.H04 Those dated figures are more precise than a general claim of “nearly 80% in the mid-1980s.”\nIntel exited DRAM in 1985, the year it introduced the 386 processor.H21 Its consumer-facing Intel Inside cooperative marketing program formally began in 1991.H05 The 1986 US–Japan Semiconductor Agreement primarily addressed access to the Japanese market and anti-dumping concerns. Later arrangements referred to an industry expectation that foreign suppliers would reach a 20% share of the Japanese market, not a binding floor reserved for US chips.H06\n1.4 Dedicated Foundries and Vertical Specialization Vertically integrated manufacturers dominated the industry\u0026rsquo;s early years, often handling product definition, design, wafer fabrication, packaging, and testing within one company. It would nevertheless be inaccurate to say that all companies followed the integrated device manufacturer (IDM) model. Specialization expanded as process development and fab construction became more expensive.\nTSMC was founded in 1987 and built its business around a dedicated foundry model: it manufactured customers\u0026rsquo; designs without selling its own branded chips.H07 This model enabled design companies to bring products to market without building advanced fabs, while foundries aggregated demand from multiple customers to spread process R\u0026amp;D and capacity investment. It created more room for fabless companies such as Qualcomm, NVIDIA, and Broadcom. AMD moved toward a fabless model much later, after spinning off manufacturing assets to GlobalFoundries in 2009–2010.H08\n1.5 Immersion Lithography, FinFETs, and EUV In the early 2000s, the industry faced growing pressure to improve the resolution of 193 nm argon-fluoride (ArF) lithography. A 157 nm exposure path had been explored, but it posed challenges for materials and optical systems. Immersion lithography placed ultrapure water between the projection lens and wafer, increasing numerical aperture and improving resolution and depth of focus while retaining the 193 nm light source. The wavelength remained 193 nm; resolution improved through the larger numerical aperture. ASML\u0026rsquo;s December 2003 announcement reported TSMC\u0026rsquo;s order for the industry\u0026rsquo;s first immersion lithography tool. This historical reference was corroborated in search-indexed text; its original URL now redirects to a general news index rather than the announcement.H19 Commercial production still required collaborative work across fabs, optics, light sources, photoresists, and research institutions.H09S01\nIn transistor architecture, Hitachi researchers demonstrated the DELTA precursor in 1989; a University of California, Berkeley team led by Chenming Hu subsequently developed and named the FinFET.H13 Berkeley\u0026rsquo;s institutional history also credits Jeff Bokor and Tsu-Jae King as collaborators. These overview sources do not establish a precise date for the naming.H10 Intel began high-volume production of its 22 nm tri-gate transistor in 2012.H11 FinFET is therefore best understood as the product of sustained research and industrialization by multiple teams, rather than the invention of a single researcher.\nEUV lithography uses 13.5 nm light. ASML delivered its first production-oriented EUV system in 2013, and customers gradually adopted EUV for advanced logic and memory production later in the 2010s. The first High-NA EUV system was delivered in 2023.H12 Prices, configurations, and revenue-recognition practices differ significantly across system generations; any quoted equipment price must specify the model, year, currency, and accounting basis.\n1.6 Mobile Computing, AI, and Heterogeneous Integration Smartphones increased demand for highly integrated, low-power systems on a chip (SoCs), radio-frequency front ends, image sensors, and mobile memory. They also helped drive advanced manufacturing from planar transistors toward FinFETs. More recently, generative AI has shifted attention toward parallel computing, HBM, high-speed networks, and advanced packaging. GPUs are well suited to massively parallel workloads, but CPUs, GPUs, purpose-built accelerators, and network processors generally work together in a system. “GPUs replace CPUs” is too simple a description.\nAs the cost of designing and producing a single large die rises, chiplets and advanced packaging have become important ways to scale systems. A chiplet architecture does more than mechanically split a large logic chip. It assigns compute, input/output, cache, or analog functions to separately designed and manufactured dies, then integrates them through standard or proprietary interconnects. “More than Moore” is broader: it also encompasses the extension of functions such as sensing, radio frequency, power, optoelectronics, and heterogeneous materials. It should not be treated as synonymous with chiplets.\n2. The Semiconductor Value Chain INDUSTRY MAP\nThe semiconductor value chain See how design, production capabilities and end markets fit together.\nFit to view Expand all Collapse Full screen Read the full text outline SemiconductorsUpstream · Design \u0026amp; inputsChip designCPUs, GPUs and accelerators Analog, power and RF EDA \u0026amp; reusable IPDesign and verification tools Processor and interface IP Equipment \u0026amp; materialsLithography and process tools Wafers, gases and chemicals Midstream · ProductionWafer fabricationDedicated foundries Integrated device manufacturers Assembly \u0026amp; packagingConventional packaging Advanced packaging and chiplets TestingWafer probing Final test and reliability Downstream · End marketsComputingData centers and AI PCs and smartphones Connected systemsAutomotive Telecommunications Specialized applicationsIndustrial and energy Defense and aerospace DEX editorial map based on the accompanying report. Examples are illustrative, not exhaustive or ranked. Companies can operate across several stages; connections show categories, not verified supplier contracts.\nSources: Semiconductor industry primer — production stages. Reviewed 2026-09-29.\n2.1 Chip Design, EDA, and Semiconductor IP Chip design begins with product requirements and system architecture, then proceeds through logic design, functional verification, synthesis, placement and routing, timing closure, physical verification, and tape-out preparation. EDA software links design rules, foundry process design kits, and manufacturing constraints. Its value comes from algorithms, complete tool flows, process compatibility, and years of accumulated validation data.\nSemiconductor IP consists of designed and verified modules that can be reused in a chip, including processor cores, memory controllers, PCIe, DDR, USB, SerDes, and security blocks. An instruction set architecture (ISA) must be distinguished from processor IP. Arm licenses both architectures and processor-core IP. RISC-V is an open-standard ISA, not a processor core that can be manufactured directly; companies must still develop or license a specific implementation.S02 x86 is a proprietary ISA ecosystem, with Intel and AMD as its principal product suppliers.\nDigital devices include CPUs, GPUs, microcontrollers, FPGAs, SoCs, network processors, and AI accelerators. Analog and mixed-signal chips manage power, data conversion, amplification, and sensor interfaces. RF and optoelectronic devices handle wireless transmission and reception, filtering, power amplification, and conversion between electrical and optical signals. These categories differ in design cycle, software dependence, product life, and process needs. An advanced node is not the only measure of a chip\u0026rsquo;s value.\n2.2 Manufacturing Equipment and Materials Front-end equipment includes lithography, etching, thin-film deposition, ion implantation, thermal processing, cleaning, chemical-mechanical polishing (CMP), metrology, and defect inspection systems. Back-end equipment includes thinning and dicing, die attach, bonding, molding, probe systems, automatic test equipment (ATE), and sorting equipment. Atomic layer deposition (ALD) is especially useful for thickness control and conformal coverage in high-aspect-ratio structures. “High selectivity” applies to particular selective deposition processes, not to every ALD tool.\nCritical materials include silicon wafers and compound-semiconductor substrates, photoresists, masks, electronic gases, wet chemicals, deposition precursors, CMP consumables, sputtering targets, package substrates, and bonding materials. Purity specifications vary by material and process; they cannot all be summarized as “nine nines.” In gas classification, phosphine, arsine, and diborane can be used for doping. Nitrogen trifluoride is primarily used for chamber cleaning and some etching processes, rather than as a typical dopant gas.\nQualification for high-volume production often takes substantial time. A supplier must demonstrate more than the specifications of a single tool or material batch: customers also need stable performance across lots, defect control, service capability, and compatibility with their process platform. This joint optimization helps explain high concentration in some niches. It does not mean that every segment has only one supplier.\n2.3 Wafer Fabrication Wafer manufacturers are commonly divided into IDMs and foundries. An IDM sells its own products and performs at least some manufacturing; a dedicated foundry primarily manufactures customer designs. In practice, the boundary is not absolute. Some IDMs offer foundry services to external customers, while some systems companies take a direct role in chip design and supply-chain management.\nA typical front-end process repeatedly applies film formation, photoresist coating, exposure, development, etching, ion implantation, thermal processing, cleaning, and CMP to form transistors and multiple interconnect layers on a wafer. After front-end fabrication, a foundry delivers a processed wafer or diced dies, not a “bare wafer.” A bare wafer is generally a substrate on which device structures have not yet been formed.\nProcess-node names identify generations of manufacturing platforms; they no longer correspond to a single directly measurable physical dimension. “2 nm” or “Intel 18A” therefore does not mean that every transistor feature measures 2 nm or 1.8 nm. Process capability should be assessed through transistor architecture, density, performance, power, yield, design rules, and production status. Intel 18A uses RibbonFET gate-all-around transistors and PowerVia backside power delivery. In 2025, Intel disclosed that the first 18A client product had entered production and that it planned to begin high-volume production that year.S03S06\n2.4 Packaging and Testing Conventional packaging protects the die, provides electrical and mechanical connections, and supports assembly into a system. Advanced packaging also enables dense interconnects, more bandwidth, power management, and heterogeneous integration. Flip-chip packaging connects a die to its substrate through bumps. In 2.5D packaging, a silicon interposer or redistribution structure can connect multiple side-by-side dies. In 3D packaging, dies are stacked using hybrid bonding, through-silicon vias (TSVs), or other vertical interconnects. CoWoS is a 2.5D and related advanced-packaging platform; it should not be conflated with every form of 3D stacking.S07\nHBM typically stacks DRAM dies above a base die and connects them through TSVs; the HBM package can then be integrated with a processor through advanced packaging.S08 3D NAND, by contrast, stacks memory cells vertically within a NAND device. It is a device structure and manufacturing process, not a synonym for TSV-based die stacking.S04\nTesting includes wafer-level probing, final testing after packaging, and reliability evaluation for particular uses. Automotive integrated circuits commonly undergo failure-mechanism-based stress tests and customer qualification under specifications such as AEC-Q100. AEC-Q100 Rev J states that AEC operates no certification board: suppliers perform qualification and submit the data for users to verify compliance. Qualification should therefore not be described as an AEC-issued certification.S05\n2.5 End Markets Data centers use CPUs, GPUs and dedicated accelerators, HBM, network switches and optical interconnects, power-management devices, and security chips. Smartphones and PCs balance performance, energy use, wireless connectivity, and cost. Automotive electronics encompass microcontrollers, analog and power devices, cockpit and driver-assistance processors, sensors, and battery-management chips; they also require lengthy qualification and supply commitments. Industrial, renewable-energy, telecommunications, and defense applications place particular weight on reliability, long-term supply, environmental tolerance, or specific security requirements.\nAdvanced nodes are not essential for every application. Power management, analog, RF, sensors, and many automotive and industrial products continue to use mature processes extensively. An industry assessment should therefore track both advanced-node investment and mature-node inventories, utilization, and replacement demand.\n3. Market Structure 3.1 Concentration and Interdependence Some semiconductor segments are highly concentrated because R\u0026amp;D is expensive, customer qualification takes time, process knowledge is difficult to replicate quickly, and software and hardware ecosystems raise switching costs. That does not establish that the top three suppliers hold 70%–90% in “almost every” subsector. A sound market-share statement first defines the market—for example, complete EUV systems, discrete data-center GPUs, DRAM, foundry services, or OSAT—and then specifies the geography, period, and whether it measures revenue or shipments.\nRegional specialization likewise cannot be reduced to closed “blocs.” US companies are strong in EDA, processor and accelerator design, and parts of the equipment market. Europe is prominent in lithography, optics, and certain automotive and industrial chips. Japan has important materials, equipment, and image-sensor companies. South Korea has large-scale memory producers. Taiwan plays a central role in foundry and packaging. Mainland China is expanding its mature-process, packaging, equipment, and materials capabilities. Cross-border investment, customer relationships, and supply dependencies remain extensive.\n3.2 Chip Design and AI Computing General-purpose processors, mobile SoCs, analog chips, and AI accelerators each have different competitive structures. NVIDIA leads in data-center GPUs and their software ecosystem, but a claim that it holds 80%–90% of “AI training and inference chips” lacks a consistent market boundary. In its review of NVIDIA\u0026rsquo;s proposed acquisition of Run:ai, the European Commission\u0026rsquo;s decision reported NVIDIA\u0026rsquo;s volume share of the defined global discrete data-center GPU market in bracketed ranges: [80–90]% in each of 2021–2023 and [70–80]% in the first half of 2024. These are estimated ranges for the specified periods, not precise shares or a full-year 2024 result. The decision records NVIDIA\u0026rsquo;s warning, as the notifying party, that volume estimates inferred from revenue and average purchase prices were less reliable than value shares.M01 The case illustrates why market share must be reported with its product scope, date, and method.\nAMD and Intel offer GPUs or other accelerators, while cloud providers develop in-house or custom ASICs such as TPUs and Trainium. In-house chips can improve performance, cost, or supply control for specific workloads, but they do not automatically displace commercial GPUs. Their results depend on software tools, utilization, model fit, networking, and deployment scale.\n3.3 Foundries and Advanced Processes MARKET SHARE · Q4 2025\nGlobal wafer foundry revenue share Wafer foundry revenue under TrendForce\u0026#39;s market definition. Samsung excludes System LSI. This is not total semiconductor revenue or the expanded Foundry 2.0 market.\nWorldwide · Share of foundry revenue (%)\nFull screen View the data table Global wafer foundry revenue share · Q4 2025 · Share of foundry revenue (%) CompanyShare TSMC70.4%Samsung Foundry7.1%SMIC5.2%UMC4.2%GlobalFoundries3.8%Other foundries9.3% Download data (CSV) Five largest suppliers shown. Other foundries = 100% minus the five published shares and includes both other ranked and unranked suppliers. Percentages retain the source rounding. Historical quarter; not full-year 2025, all chip sales, or the broader Foundry 2.0 definition.\nSource: TrendForce — AI Demand Drives 4Q25 Global Top 10 Foundries Revenue Up 2.6% QoQ; Samsung Gains Share and Tower Moves Up in Rankings (2026-03-12). Reviewed 2026-09-29.\nTSMC is the leading dedicated foundry. In its 2025 annual report, the company defined “Foundry 2.0” broadly to include logic wafer fabrication, packaging, testing, masks, and non-memory IDM activity, and estimated that market at US$305 billion in 2025. This is substantially broader than conventional dedicated foundry services; a Foundry 2.0 share should not be directly compared with a third-party pure-foundry share. TSMC also reported that its 3 nm process accounted for 24% of its own wafer revenue in 2025 and that its 2 nm process entered volume production in the fourth quarter of that year.M02 Those figures describe TSMC\u0026rsquo;s revenue mix and manufacturing progress, not the entire industry\u0026rsquo;s 3 nm or 2 nm market share.\nSamsung operates in memory, logic products, and foundry services, so its process investment must be considered alongside both internal IDM demand and external foundry customers. Intel offers manufacturing and packaging to external customers through Intel Foundry; the scale of 18A production and external customer adoption should be updated against subsequent earnings reports and product deliveries. SMIC, UMC, and GlobalFoundries also have different product mixes, process platforms, customer industries, and expansion priorities.\n3.4 Memory and Advanced Packaging The major DRAM suppliers include Samsung Electronics, SK hynix, and Micron. NAND participants also include Kioxia, Western Digital/SanDisk-related operations, and Solidigm. Memory is highly cyclical: prices respond to inventory, capital expenditure, product transitions, and end-market demand. HBM growth has prompted suppliers to invest in advanced DRAM, TSVs, and packaging. Any assertion that “two companies hold the overwhelming majority of HBM” should identify the quarter and whether it measures revenue or bit shipments, while accounting for changes at suppliers including Micron.\nFoundries, memory makers, IDMs, and OSATs all participate in advanced packaging. ASE, Amkor, and JCET are major OSAT providers, while TSMC, Samsung, and Intel combine advanced packaging with front-end processes. Control is not simply shifting in one direction from OSATs to foundries. Platforms compete and overlap in interposers, hybrid bonding, HBM integration, testing, and volume delivery.\n3.5 Equipment, Materials, and Profitability ASML is currently the only company able to supply complete EUV lithography systems commercially. DUV, metrology, inspection, and other manufacturing-equipment markets have different competitors. ASML\u0026rsquo;s 2025 annual report records €32.7 billion in total net sales, a gross margin of 52.8%, and revenue recognition for 48 EUV systems during its 2025 fiscal year.M03 These figures illustrate the scale and technical barriers of the EUV business. They do not support a claim that every equipment monopoly earns a 60%–80% gross margin.\nTSMC reported a gross margin of 59.9% for 2025. Revenue recognition, depreciation, and cost structures differ among EDA, IP, equipment, materials, foundries, and packaging and test providers.M02 Profitability should therefore be analyzed using specific companies and a consistent fiscal year and accounting basis. At minimum, software licenses, equipment sales, materials, manufacturing, and testing should be distinguished rather than assigned fixed margins across the value chain.\n4. Principal Risks and Constraints 4.1 Industry Cycles and Concentrated AI Demand AI infrastructure is increasing demand for advanced logic, HBM, advanced packaging, networks, and power devices, but semiconductors remain subject to inventory and capital-spending cycles. If cloud providers slow investment, relevant suppliers could face order revisions and lower utilization. If AI-related revenue and computing demand continue to grow, advanced capacity could remain tight in the near term. These are conditional scenarios, not grounds for declaring an inevitable “ROI cliff” or “profit collapse.”\nAI capacity does not crowd out every traditional chip category in equal measure. Advanced GPUs and automotive microcontrollers generally use different nodes and production lines, limiting direct substitution. HBM expansion may redirect some DRAM resources, but consumer-memory prices also depend on inventory, demand, and suppliers\u0026rsquo; capital discipline. Risk analysis needs to distinguish products and processes.\n4.2 Industrial Policy, Export Controls, and Regionalization The US CHIPS and Science Act allocated US$50 billion for the Department of Commerce to administer semiconductor incentives and R\u0026amp;D programs. That figure represents statutory program funding, not cash already paid to companies.R01 The European Chips Act took effect on September 21, 2023. In its release that day, the European Commission stated the EU\u0026rsquo;s policy goal of raising its share of the global semiconductor market to 20% by 2030; that number is a historical policy target, neither an achieved share nor a firm forecast.R02 In June 2026, the European Commission proposed a Chips Act 2.0 to build on the original law. The proposal should be distinguished from the 2023 act already in force.R05\nExport controls are changing customer screening and delivery procedures for equipment, software, HBM, and advanced computing chips. In January 2025, the US Bureau of Industry and Security updated advanced-computing controls and foundry due-diligence requirements; related rules also changed definitions of advanced-node integrated circuits and the Entity List.R03 Businesses consequently face licensing, end-user, resale, technical-service, and geographic compliance risks. Policies can change, so a rule in force at one point should not be treated as a permanent industrial boundary.\nRegional incentives can add local capabilities and geographic redundancy, but they can also raise construction costs, reduce utilization, intensify competition for talent, and complicate cross-border operations. Whether a project amounts to “duplicative capacity” depends on actual demand, its technology generation, and long-term utilization. Not every localization project can be assumed in advance to destroy economies of scale.\n4.3 Technical Complexity and Recovery of Capital Investment Advanced processes face short-channel effects, interconnect delay, power density, heat, stochastic defects, and growing design complexity. Gate-all-around transistors, backside power delivery, EUV, High-NA EUV, and advanced packaging offer new ways to scale, while increasing R\u0026amp;D, equipment, mask, design-migration, and yield-ramp costs. A node label is not a physical-limit gauge. The characters “2 nm” alone cannot establish that quantum tunneling has become the decisive obstacle for every product.\nInvestment in a fab or critical tool varies substantially with the project boundary, cleanroom, equipment mix, capacity, and location. A claim that a fab costs US$20–30 billion, or that a certain tool has a particular price, should identify a specific project or system, announcement date, currency, and whether infrastructure is included. Totals from different projects should not be substituted for one another.\n4.4 Supply Concentration and Operational Continuity Concentrated supply does create single-point risks. ASML is currently the sole commercial supplier of complete EUV systems, and some equipment subsystems, material formulations, and advanced-packaging capacities are concentrated among a small number of firms. Yet photoresists, metrology and inspection, and most process materials generally have multiple suppliers—even if an alternative cannot be qualified and substituted quickly. More useful risk measures include qualification time for replacements, inventory coverage, geographic concentration, capacity flexibility, and the cost of an alternative process. A blanket claim that there is “no second option anywhere in the world” obscures these differences.\nCompanies can reduce exposure through developing second sources, stocking critical spare parts, diversifying production sites, signing long-term purchase agreements, and conducting joint qualification. Because semiconductor tools and materials must be qualified against specific processes, establishing an alternative often takes months or longer. That work should begin during normal operations, before a supply interruption.\n4.5 Electricity, Water, and Infrastructure Advanced fabs require reliable electricity, ultrapure water, gases, and waste-treatment systems. AI data centers are increasing demand for high-density computing, cooling, and grid connections. The International Energy Agency estimates that data centers used about 415 TWh of electricity worldwide in 2024, or about 1.5% of global electricity consumption. In its 2025 base case, the IEA projects roughly 945 TWh by 2030.R04 These are global model estimates; they do not mean that every regional grid will reach its limits at the same time.\nPower constraints vary sharply by location, depending on grid-connection queues, generation mix, transmission and distribution capacity, and data-center clustering. Semiconductor companies should evaluate power reliability, water availability, extreme weather, and carbon costs when selecting sites. Data-center customers should also incorporate server utilization, model efficiency, and cooling methods into capacity planning.\n4.6 Talent and Organizational Capability Semiconductor production requires specialists in devices, materials, chemistry, mechanics, optics, software, quality, and equipment maintenance. New fabs need more than additional graduates: they need experienced teams that can introduce processes into volume production, improve yield, and keep tools running. Talent risk should not be described as an equally “severe shortage” in every region. It should be measured by project location, job category, hiring time, and turnover.\nIf industrial policy subsidizes buildings and machines without vocational training, research platforms, supplier engineering capacity, and arrangements for international talent mobility, new capital may be slow to turn into stable output. Companies should include training periods, succession for critical roles, replication across fabs, and supplier-service capacity in their expansion plans.\nConclusion Technological accumulation, specialization, and cross-border collaboration define the semiconductor industry. Advances in transistor architecture, lithography, materials, design tools, fabrication, and packaging depend on one another. No single segment determines the outcome on its own. Dedicated foundries separated design and manufacturing, while advanced packaging is bringing front-end and back-end work closer together again. AI has increased demand for advanced computing, but has also deepened dependence on HBM, interconnects, electricity, and software ecosystems.\nAssessing a company or market requires comparable definitions and data. A market-share figure needs a time period, geography, product boundary, and measurement basis. An equipment price needs a model and year. A process node cannot be read as a literal physical dimension. Gross margins should be compared only under consistent accounting conventions. Unpublished yields, customer confidence, or future capacity should not be presented as established facts.\nOver the next several years, competition will center on four capabilities: advancing device and system technologies; maintaining efficient volume production despite heavy capital spending; building auditable cross-border supply networks; and securing power, talent, and critical materials. Regionalization will increase the weight of compliance and redundancy in investment decisions without fully replacing global specialization. Companies that assess technology roadmaps, customer demand, capital discipline, and supply security together will be better placed to navigate both growth and cyclical volatility.\nReferences Historical Sources [H01] Computer History Museum, “1954: Silicon Transistors Offer Superior Operating Characteristics.” https://www.computerhistory.org/siliconengine/silicon-transistors-offer-superior-operating-characteristics/ Scope: Tanenbaum\u0026rsquo;s 1954 device, Teal\u0026rsquo;s laboratory-organizing role, Adcock\u0026rsquo;s team leadership, commercial silicon transistors, and temperature performance. The later oxide-layer discussion uses [H17]; this 1954 page is not evidence for the separate 1957 Fairchild account.\n[H02] Intel, “Moore’s Law.” https://www.intel.com/content/www/us/en/newsroom/resources/moores-law.html\n[H03] Intel, “The Chip that Changed the World.” https://www.intel.com/content/www/us/en/newsroom/opinion/chip-that-changed-world.html Canonical destination of the former newsroom link. Supports the calculator origin and 1971 introduction; the four-bit and 2,300-transistor specifications are sourced separately to [H20].\n[H04] U.S. International Trade Commission, “The South Korea-Japan Trade Dispute in Context: Semiconductor Manufacturing, Chemicals and Concentrated Supply Chains.” https://usitc.gov/sites/default/files/publications/332/working_papers/semiconductor_working_paper_corrected_103119.pdf\n[H05] Intel, “Ingredient Branding: End User Marketing and Intel Inside.” https://www.intel.com/content/www/us/en/history/virtual-vault/articles/end-user-marketing-intel-inside.html Scope: the 1991 campaign launch and cooperative advertising model. It does not establish the separate 1985 DRAM exit, which is sourced to [H21].\n[H06] Office of the United States Trade Representative, “1996 National Trade Estimate—Japan: Semiconductors.” https://ustr.gov/archive/Document_Library/Reports_Publications/1996/1996_National_Trade_Estimate/1996_National_Trade_Estimate-Japan.html\n[H07] TSMC, “2025 Annual Report—About TSMC.” https://investor.tsmc.com/static/annualReports/2025/english/index.html\n[H08] AMD, “AMD Reports Fourth Quarter and Annual Results,” January 21, 2010. https://ir.amd.com/financial-information/sec-filings/content/0001193125-10-009806/dex991.htm\n[H09] ASML, “How Immersion Lithography Saved Moore’s Law,” 2023. https://www.asml.com/en/company/stories/2023/how-immersion-lithography-saved-moores-law\n[H10] University of California, Berkeley EECS, “History.” https://eecs.berkeley.edu/about/history/ Location: semiconductor-history paragraph naming Bokor, Hu, and King as FinFET collaborators. This institutional overview does not date the naming; [H13] supports the earlier Hitachi precursor and subsequent Berkeley development.\n[H11] Intel, “Moore’s Law: Fun Facts.” https://www.intel.com/content/www/us/en/history/history-moores-law-fun-facts-factsheet.html\n[H12] ASML, “EUV Lithography Systems.” https://www.asml.com/en/products/euv-lithography-systems\n[H13] IEEE Technology Navigator, “FinFETs.” https://technav.ieee.org/topic/finfets/ Location: “What Are FinFETs?” Supports the 1989 Hitachi DELTA precursor and the subsequent Berkeley development and naming, but not a precise late-1990s naming date.\n[H14] Ministry of Economy, Trade and Industry of Japan, “2018 White Paper on International Economy and Trade—VLSI Project History.” https://www.meti.go.jp/report/tsuhaku2018/2018honbun/i2220000.html\n[H15] John Bardeen and Walter H. Brattain, US 2,524,035, “Three-Electrode Circuit Element Utilizing Semiconductive Materials.” https://patents.google.com/patent/US2524035A/en\n[H16] Jack S. Kilby, US 3,138,743, “Miniaturized Electronic Circuits.” https://patents.google.com/patent/US3138743A/en\n[H17] Robert N. Noyce, US 2,981,877, “Semiconductor Device-and-Lead Structure.” https://patents.google.com/patent/US2981877A/en\n[H18] Computer History Museum, “1947: Invention of the Point-Contact Transistor.” https://www.computerhistory.org/siliconengine/invention-of-the-point-contact-transistor/\n[H19] ASML, “TSMC Selects ASML for Industry’s First Immersion Tool Order,” December 3, 2003. https://www.asml.com/en/news/press-releases/2003/tsmc-selects-asml-for-industry-first-immersion-tool-order Historical reference with an access limitation: the title and TSMC order statement were corroborated in search-indexed text, but the original URL redirected to ASML\u0026rsquo;s generic press-release index on October 3, 2026. This is not a currently accessible live copy of the release, and no verified equivalent live replacement was found. The accessible 2023 retrospective [H09] provides immersion-history context; it does not independently establish the full 2003 order announcement.\n[H20] Intel, “Celebrating the 50th Anniversary of the Intel 4004,” 2021 infographic (PDF), p. 1. https://download.intel.com/newsroom/2021/data-center/4004-infographic.pdf Location: 1971 comparison column. Supports the four-bit instruction-set description and 2,300-transistor count.\n[H21] Intel, “Semiconductors and Intel: An Introduction” (PDF), p. 18, “Intel’s history in 4 fast eras.” https://www.intel.com/content/dam/www/central-libraries/us/en/documents/semiconductors-and-intel-introduction.pdf Location: 1985–1995 timeline. Supports the 1985 DRAM exit and 386 introduction; it is a separate source from the Intel Inside marketing history.\nTechnology and Value-Chain Sources [S01] ASML, “Lenses and Mirrors—Lithography Principles.” https://www.asml.com/technology/lithography-principles/lenses-and-mirrors\n[S02] RISC-V International, “About RISC-V.” https://riscv.org/about/\n[S03] Intel, “Intel 18A Process Technology Simply Explained,” January 30, 2025. https://newsroom.intel.com/intel-foundry/intel-18a-process-technology-simply-explained\n[S04] Samsung Semiconductor, “3D V-NAND Flash Memory.” https://semiconductor.samsung.com/support/tools-resources/dictionary/semiconductor-glossary-3d-v-nand-flash-memory/ Scope: vertically stacked NAND memory cells and their distinction from a single-layer arrangement. This glossary does not establish HBM\u0026rsquo;s DRAM/base-die structure or CoWoS packaging; those claims use [S08] and [S07].\n[S05] Automotive Electronics Council, “AEC-Q100: Failure Mechanism Based Stress Test Qualification for Integrated Circuits,” Rev J, August 11, 2023. AEC-Q100 Rev J — manufacturer-hosted copy at Guerrilla RF (PDF). Location: §§1.3.1–1.3.3, printed p. 2 (PDF p. 8), on qualification, the absence of an AEC certification board, and user approval. The AEC publisher documents index could not be retrieved during the October 3, 2026 review; that access failure does not establish deletion. The inspected copy is the AEC standard hosted by a manufacturer, not the publisher\u0026rsquo;s live index, and does not establish which revision is currently latest.\n[S06] Intel, “Postcard from Intel Technology Tour Arizona: Panther Lake Draws in Cameras and Crowds,” October 10, 2025. https://www.intel.com/content/www/us/en/newsroom/news/client-computing/postcard-itt-panther-lake-draws-cameras-and-crowds.html\n[S07] TSMC, “CoWoS.” https://3dfabric.tsmc.com/english/dedicatedFoundry/technology/cowos.htm Location: technology overview and CoWoS-S/R/L descriptions. Supports the 2.5D integration of logic and HBM using silicon or redistribution-layer interposers; it is not a source for every form of 3D bonding.\n[S08] SK hynix, “SK hynix Partners with TSMC to Strengthen HBM Technological Leadership,” April 19, 2024. https://news.skhynix.com/en/sk-hynix-partners-with-tsmc-to-strengthen-hbm-technological-leadership/ Location: base-die paragraph and TSV/CoWoS explanatory notes. Supports the DRAM/base-die stack, TSV interconnections, and integration with a processor; cited for technical structure, not for promotional leadership claims or later production outcomes.\nMarket and Company Sources [M01] European Commission, Case M.11766, NVIDIA/Run:ai merger decision, December 20, 2024. https://ec.europa.eu/competition/mergers/cases1/202516/M_11766_10599589_2740_3.pdf Location: §4.2.1, Table 2 and paragraph 92, printed pp. 21–22 (PDF pp. 22–23). The market is worldwide discrete data-center GPUs by volume; the bracketed ranges cover 2021–2023 and H1 2024. Paragraph 92 records the notifying party NVIDIA\u0026rsquo;s caution about the reliability of volume estimates derived from revenue and average purchase prices. That caution is attributed to NVIDIA, not presented as an independently established Commission finding.\n[M02] TSMC, “2025 Annual Report.” https://investor.tsmc.com/static/annualReports/2025/english/index.html\n[M03] ASML, “2025 Annual Report.” https://www.asml.com/en/investors/annual-report/2025\nPolicy and Risk Sources [R01] NIST, “Funding Updates.” https://www.nist.gov/chips/funding-updates Location: opening program-funding paragraph. Official fallback confirming Commerce\u0026rsquo;s administration of US$50 billion in semiconductor incentives and R\u0026amp;D funding; this is an allocation, not cash already disbursed. Original provenance: U.S. Department of Commerce, “Semiconductor Industry—CHIPS for America”. Direct access to that Commerce page returned HTTP 403 during the October 3, 2026 review; it is access-blocked, not established to be deleted. The funding source does not independently establish the report\u0026rsquo;s regionalization cost analysis.\n[R02] European Commission, “Digital Sovereignty: European Chips Act Enters into Force,” September 21, 2023. https://digital-strategy.ec.europa.eu/en/news/digital-sovereignty-european-chips-act-enters-force Location: opening and paragraph stating the 20%-by-2030 goal. This dated release supports commencement and the historical policy target. The current European Chips Act policy page remains useful for policy context but no longer states that target in the version reviewed on October 3, 2026; it is not substituted for the dated evidence.\n[R03] U.S. Bureau of Industry and Security, “Commerce Strengthens Restrictions on Advanced Computing Semiconductors,” January 15, 2025. https://www.bis.gov/press-release/commerce-strengthens-restrictions-advanced-computing-semiconductors-enhance-foundry-due-diligence-prevent\n[R04] International Energy Agency, “Energy and AI,” April 10, 2025. https://www.iea.org/reports/energy-and-ai\n[R05] European Commission, “Proposal for the Chips Act 2.0,” June 3, 2026. https://digital-strategy.ec.europa.eu/en/library/proposal-chips-act-20\n","date":"2026-09-28T00:00:00Z","image":"/post/semiconductor-industry-report/cover.webp","permalink":"/post/semiconductor-industry-report/","title":"Semiconductors and Chips: Industry History, Value Chain, Markets, and Risks"},{"content":"A sourced starting map of AI inputs, model platforms and applications, with business-model questions and metrics to track.  From computing inputs to models and paid workflows. artificial intelligence models inference enterprise automation 人工智能 Enterprise AI software and services; excludes a standalone valuation of chipmakers or cloud infrastructure. Who gets paid when an AI capability becomes a useful workflow? Global business-model lens; US framework and vendor examples. Follow customer revenue after inference, support and integration costs. A growing user count alone does not establish attractive unit economics. Compute \u0026 data Cloud providers, data owners and tooling suppliers Supply the resources used to build and run AI systems. Compute consumption, data licensing or tooling contracts. Can data rights and serving costs support the intended use? Models \u0026 deployment Model developers and deployment platforms Turn resources into usable inference and evaluation services. Usage-based APIs, capacity commitments or licenses. Does reliability hold under real workloads, not just benchmarks? Applications \u0026 adoption Software vendors, integrators and enterprise teams Embed a capability into a customer's operating process. Subscriptions, implementation projects or usage fees. Will verified savings translate into repeat paid use? Which workflows have measurable time savings? Do pilots become recurring paid deployments? Model errors, data rights and accountability may block adoption. Falling model prices may help customers without improving every supplier's margin. Cost per completed task Include inference, retries and human review; compare like-for-like tasks. Paid retention Track continued paid use, not registrations or free trials. Task success rate Use a defined test set and disclose human intervention. Amazon Bedrock publishes input/output token pricing for several models, providing a concrete example of usage-based inference billing. NIST's AI RMF is voluntary guidance for managing risk across AI design, development, use and evaluation. AWS Amazon Bedrock Pricing Undated; live pricing page Company documentation Billing-model example only; no price or market-size estimate is reproduced. NIST AI Risk Management Framework AI RMF 1.0 released 2023-01-26 Government framework Risk-management scope; not a revenue forecast or an official value-chain taxonomy. AI risks and industry turbulence","date":"2026-09-22T00:00:00Z","permalink":"/industry-breakdowns/ai-software/","title":"AI Software \u0026 Services — Industry Breakdown"},{"content":"Explore battery materials, systems and project operation, with research questions and clearly scoped IEA evidence.  From minerals and cells to installed storage systems. battery lithium stationary grid minerals cells 储能 电池 Battery supply chains, with a stationary-storage operating lens. EV batteries and grid-storage revenue pools are not interchangeable. Who earns the return: material suppliers, manufacturers or operators? Global supply chain; project economics depend on local power markets. Do not equate cheaper cells with better project returns. Connection, installation, financing, degradation and market rules belong in the same model. Materials \u0026 components Miners, refiners and component suppliers Provide battery-grade inputs and components. Material and component sales under spot or contract terms. Do quality, supply security and chemistry choices align? Cells \u0026 systems Cell makers, pack makers and system integrators Turn components into a tested, controllable storage system. Cell/system sales, integration and warranty services. Can manufacturing quality and warranty obligations be sustained? Projects \u0026 operation Developers, utilities and storage operators Connect and operate assets to serve a particular power market. Contracted or market-based power-system services. Will local revenues cover degradation and full project costs? Where does variable generation create a need for flexibility? Can storage participate and get paid in the relevant local market? Safety, degradation and warranties may change lifecycle costs. Supply concentration and changing project revenues can offset lower hardware prices. Installed system cost Use consistent currency, date and system boundary; distinguish USD/kW from USD/kWh. Usable capacity \u0026 efficiency Measure at the same operating conditions and asset age. Contracted revenue share Separate contracted cash flows from merchant assumptions. The IEA describes concentration across battery supply chains and explains storage's role in flexibility and grid stability. Its 2024 scenarios are not treated here as current forecasts. International Energy Agency Batteries and Secure Energy Transitions — Executive summary 2024-04-25 Intergovernmental analysis Supply-chain structure and power-system applications; historical scenarios, not fresh market estimates.","date":"2026-09-22T00:00:00Z","permalink":"/industry-breakdowns/battery-storage/","title":"Batteries \u0026 Energy Storage — Industry Breakdown"},{"content":"Connect drug discovery, clinical evidence and commercialization through a carefully scoped US regulatory reference.  From scientific discovery to evidence and commercialization. biotech pharmaceuticals drugs clinical trials medicine 生物制药 医药 Innovative drug development. The US pathway is an illustrative regulatory reference; individual products and other jurisdictions require separate analysis. How does a scientific asset become an approved and commercially viable product? US regulatory reference; not a universal approval pathway. Separate scientific progress, approval and commercial success. Model development spending and potential dilution before assuming future product revenue. Discovery \u0026 preclinical Research teams, biotechs and research suppliers Develop candidates and investigate their properties before human studies. Research services or licensing arrangements to investigate. Does the evidence justify further development and funding? Clinical evidence \u0026 review Sponsors, trial providers and regulators Generate human evidence and evaluate a submission for approval. Trial-service contracts or milestone arrangements to investigate. Are safety, efficacy and manufacturing evidence sufficient? Manufacture \u0026 access Manufacturers, commercial teams and health-system buyers Supply approved products while maintaining quality and safety monitoring. Product sales, royalties or manufacturing services to investigate. Does approval translate into supply, access and viable net revenue? What unmet need would a successful product address? Which evidence would change clinical or purchasing decisions? Clinical failure and manufacturing problems can change the entire thesis. Approval alone does not establish reimbursement or commercial demand. Evidence milestones State trial phase, population and endpoints; avoid treating a milestone as guaranteed. Cash runway Test spending assumptions and funding needs under delays. Net realized revenue Distinguish actual sales from headline prices and potential markets. FDA's drug-development overview describes discovery, preclinical research, clinical research, review and post-market safety monitoring. The three cards group those activities for readability. The 1962 U.S. Drug Amendments strengthened the evidence of effectiveness required for new-drug approval. This is historical U.S. law, not a worldwide approval pathway. US Food and Drug Administration The Drug Development Process Undated overview; reviewed on the date shown above Regulator explainer US development and review stages; not clinical advice or evidence for a particular drug. U.S. Congress / GovInfo Drug Amendments of 1962, Public Law 87-781 1962-10-10 Original statute · PDF Historical U.S. effectiveness requirement and regulatory powers; not a global or modern market-size estimate. Pharmaceutical industry report","date":"2026-09-22T00:00:00Z","permalink":"/industry-breakdowns/biopharmaceuticals/","title":"Biopharmaceuticals — Industry Breakdown"},{"content":"Understand security signals, products and operations, and separate business-model interpretation from NIST framework evidence.  From security inputs to products and operational response. security identity threat intelligence detection response zero trust 网络安全 Enterprise security products and services. The commercial layers below are DEX's map, not the NIST CSF functions or a compliance checklist. Does the customer buy another tool, or a better security outcome? Global enterprise lens; NIST provides the risk-management reference. Test renewal quality and delivery cost alongside product claims. Security spending is not itself evidence that customer risk has fallen. Signals \u0026 foundations Telemetry, identity and intelligence providers Supply the information and controls used by security teams. Data feeds, infrastructure or identity-service contracts. Are the signals relevant, complete and lawful to use? Security products Endpoint, cloud, network and identity vendors Turn inputs into prevention, detection and investigation tools. Subscriptions, licenses or consumption-based fees. Can tools integrate without overwhelming operators? Operations \u0026 response Internal teams, managed providers and responders Operate controls and coordinate incident handling and recovery. Managed-service contracts, retainers or response projects. Can the customer turn alerts into timely, effective action? Which assets and workflows create an uncovered exposure? Would integration or managed delivery solve a staffing bottleneck? Product overlap and tool consolidation may affect renewals. Failures in the vendor's own systems can undermine trust. Net revenue retention Use a consistent cohort and vendor definition. Response effectiveness Define incident severity and measurement boundaries before comparing. Service delivery cost Include analyst time and infrastructure per customer. NIST CSF 2.0 organizes cybersecurity outcomes under Govern, Identify, Protect, Detect, Respond and Recover. These are outcomes, not six sequential commercial stages. NIST SP 800-207 sets out zero-trust architectural principles; it is not a market-sizing or vendor-share study. Menlo Ventures' 2022 market map is an illustrative snapshot of product categories and vendors, not a current estimate of spending or company shares. NIST The NIST Cybersecurity Framework (CSF) 2.0 2024-02-26 Government framework · PDF Risk-management functions only; does not establish vendor revenue, effectiveness or market size. NIST SP 800-207: Zero Trust Architecture 2020-08 Government publication Zero-trust architecture concepts; not an estimate of industry revenue. Menlo Ventures Cybersecurity Market Map 2022-09 Investor research · PDF Illustrative historical vendor categories; not a verified 2026 revenue dataset. Cybersecurity industry report","date":"2026-09-22T00:00:00Z","permalink":"/industry-breakdowns/cybersecurity/","title":"Cybersecurity — Industry Breakdown"},{"content":"Map industrial robot components, integration and lifecycle services, with payback questions and metrics to investigate.  From components to integrated, supported automation. robots automation industrial integration actuators cobots 机器人 Industrial robot systems and their integration. This is a subset of the explorer's Industrial + Service Robotics row; not a humanoid or service-robot forecast. Can a robot deliver a repeatable outcome in a real operating environment? Global industrial lens; ABB is one vendor example, not a market ranking. Evaluate the complete installed cell and support burden, not just the robot's purchase price. Pilot demonstrations are not recurring production deployments. Components \u0026 control Motion, sensing and control suppliers Provide the hardware and software building blocks. Component sales and software licenses. Can parts satisfy precision, durability and integration requirements? Robots \u0026 integration Robot manufacturers and system integrators Combine machines, tooling and software around a defined task. Equipment sales and engineering projects. How much customization is required for each deployment? Operation \u0026 lifecycle Factory operators and service providers Commission, maintain and improve the deployed system. Maintenance, training and modernization contracts. Does uptime and throughput justify the full lifecycle cost? Which repetitive tasks have stable requirements and measurable payback? Can integration be reused across multiple customer sites? Customization and commissioning delays can erode project margins. Reliability, operator training and safe integration need on-site validation. Installed-system payback Include integration, downtime, maintenance and training. Productive uptime Measure time actually completing the intended task. Repeat deployments Separate pilots from repeat orders and operating installations. ABB's robotics services cover installation, support, maintenance and modernization, illustrating commercial activity beyond the initial equipment sale. ABB Robotics Services Undated; product/service page Company documentation One vendor's service offerings; not independent evidence of achieved returns or industry-wide revenue mix.","date":"2026-09-22T00:00:00Z","permalink":"/industry-breakdowns/robotics/","title":"Robotics — Industry Breakdown"},{"content":"Follow semiconductor design, wafer fabrication and packaging, and examine the economics and evidence behind each stage.  From chip design to fabrication, packaging and testing. chips integrated circuits foundry design packaging 半导体 芯片 Semiconductor production, with equipment and materials as enabling inputs. Not a ranking of individual companies or process nodes. Which production constraint determines where value is captured? Global production lens; industry-body primer. Separate design economics from capital-intensive manufacturing. A shortage in one process or package does not imply a shortage across all chips. Design Chip designers and design-tool/IP suppliers Specify the device and prepare it for production. Chip sales, design services or intellectual-property licensing. Can the design meet performance, power and cost requirements? Wafer fabrication Foundries, integrated manufacturers and input suppliers Fabricate circuits on wafers using specialized facilities. Manufacturing services or internally produced chip sales. Are yield and utilization sufficient to cover fixed costs? Assembly, test \u0026 packaging Packaging/test providers and integrated manufacturers Separate, test and package devices for customers. Packaging and testing services or bundled chip sales. Can packaging capacity and reliability meet product requirements? Which applications are creating durable design wins? Where do customer requirements exceed available production capability? Capacity additions can arrive after the demand cycle changes. Geographic dependencies and customer concentration require separate analysis. Manufacturing yield Good devices as a share of output, for a specified product/process. Capacity utilization Track actual loading, not announced capacity alone. Inventory \u0026 lead time Read together to distinguish shortages from order inflation. SIA's primer distinguishes design, front-end fabrication and back-end assembly, test and packaging. The three-stage map here follows that production sequence. Semiconductor Industry Association Semiconductor Industry Primer: The Stages of Production and Business Models 2015-02-25 Industry-body explainer Production stages; foundational reference, not current market sizing. Semiconductor industry report","date":"2026-09-22T00:00:00Z","permalink":"/industry-breakdowns/semiconductors/","title":"Semiconductors — Industry Breakdown"},{"content":"Explore 50 industries, compare available market-size and growth estimates, and find related industry guides and research.  Find an industry to researchOpen an industry for its details, sources, and related research.\nDownload CSV ↓ The battery-cell row has a sourced 2022 correction. Other original screening estimates are unverified. About the data\nSearch Category All categories AI Foundation Digitization Energy Hard Tech Bio \u0026 Health Manufacturing More filtersMaturity All stages Early Growth Mature Minimum growth (CAGR) Any 10%+ 20%+ 25%+ 50 industries\nClear filtersSort byOriginal orderIndustry nameGrowth rate (low to high)Category All industries and research links are available below. Enable JavaScript to filter, sort, and explore growth scenarios.\nProvisional industry screening estimates. Market years are as labeled; unlabeled values were originally presented as approximately 2025. Open an industry to inspect its sources and research. # Industry Key technologies Market size · see year Growth · CAGR 1 AI Software \u0026amp; ServicesAI FoundationOriginal estimates unverified\nUnverified estimate; source and methodology not confirmed. Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: $1.5–2.4T (2030–32) (unverified)\nEnterprise AI spend surging; Gartner ~$1.5T AI spend in 2025\nRead claim-level evidence appendix\nRead industry guideRead report Large models / Agentic AI, Generative AI ~$230–390B 25%–35% 2 SemiconductorsAI FoundationSourced baseline · 2025Original estimates unverified\nUnverified estimate; source and methodology not confirmed. Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: $1.5–3.2T (2030) (unverified)\nAI servers driving growth; BofA sees $3.2T TAM by 2030\nRead claim-level evidence appendix\nSeparate baseline: USD 795.6 billion · 2025. Annual semiconductor product sales. Worldwide. Semiconductor product sales tracked by WSTS. This is not combined revenue across chip design, EDA, manufacturing equipment, materials and fabrication services.. This separately sourced 2025 observation is above the original ~$630–775B screening range. It does not establish that range, the 10%–18% CAGR, or the $1.5–3.2T projection. WSTS product sales also differ from a whole-value-chain revenue total. World Semiconductor Trade Statistics (WSTS)\nRead industry guideRead reportView industry mind mapView share chart Advanced nodes (2nm/3nm), HBM, AI accelerators ~$630–775B 10%–18% 3 Cloud ServicesAI FoundationOriginal estimates unverified\nUnverified estimate; source and methodology not confirmed. Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: $1.6–3.4T (2040) (unverified)\nInfrastructure demand accelerating with AI\nRead claim-level evidence appendix\nHyperscale data centers, AI training/inference cloud ~$220B baseline 12%–17% 4 Electric Vehicles (EVs)EnergyOriginal estimates unverified\nUnverified estimate; source and methodology not confirmed. Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: $1.1–3.7T (2030–35) (unverified)\nChina leads volume; IEA tracks global adoption\nRead claim-level evidence appendix\nSolid-state batteries, 800V platforms, domain controllers ~$900–990B 10%–15% 5 Batteries \u0026amp; Energy StorageEnergySourced estimate · 2022\nThis figure is an estimate. The $98B baseline is for 2022; the 2040 values and 12%–14% CAGR are scenarios for 2022–2040. Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: $810B–1.1T (2040 scenario)\nNumeric scope: primarily lithium-ion and sodium-ion battery cells; not the full energy-storage industry. China new-type storage is a pillar industry.\nRead claim-level evidence appendix\nData year: 2022. CAGR period: 2022–2040. 2022 historical estimate; 2022–2040 modeled scenario, published in 2024. Battery-cell revenues, primarily lithium-ion and sodium-ion, for EVs, stationary BESS and consumer electronics. Excludes lead-acid and traditional household batteries; not complete storage-system or whole-value-chain revenue. McKinsey Global Institute, The next big arenas of competition, printed pp. 137–140 and endnote 209. Numeric source, published 2024-10-23.\nRead industry guide Solid-state, sodium-ion, flow batteries ~$98B (2022; defined battery-cell revenues) 12%–14% (2022–2040 scenario) 6 Shared Autonomous VehiclesHard TechOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: $610B–2.3T (2040) (unverified)\nWaymo, Cruise, Tesla advancing commercialization\nL4/L5 autonomy, Robotaxi, vehicle-road coordination Early stage High (4%–20%) 7 Space EconomyHard TechOriginal estimates unverified\nUnverified estimate; source and methodology not confirmed. Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: $1–1.8T (2035–40) (unverified)\nLaunch cost decline; commercial services dominant\nRead claim-level evidence appendix\nReusable rockets, LEO constellations, satellite internet ~$550–626B 7%–10% 8 CybersecurityDigitizationSourced baseline · 2024Original estimates unverified\nUnverified estimate; source and methodology not confirmed. Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: $590B–1.2T (2040) (unverified)\nAI threats \u0026#43; quantum computing driving demand\nRead claim-level evidence appendix\nSeparate baseline: USD 193.408 billion · 2024. Estimated information security end-user spending. Worldwide. Gartner worldwide information security end-user spending across security software, security services and network security; a historical estimate from the July 2025 forecast vintage, not vendor revenue or an audited industry total.. The source reports an estimated 2024 spending baseline; its 2025 and 2026 figures are forecasts. This does not verify the original undated ~$160–240B range, 8%–20% CAGR, or 2040 projection. Spending and vendor sales may use different market boundaries. Gartner\nRead industry guideRead reportView industry mind mapView share chart Zero Trust, AI security, quantum-safe cryptography ~$160–240B 8%–20% 9 Robotics (Industrial \u0026#43; Service)Hard TechOriginal estimates unverified\nUnverified estimate; source and methodology not confirmed. Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: $110–205B (2030) (unverified)\nChina intelligent robots is a pillar industry\nRead claim-level evidence appendix\nRead industry guide Humanoid robots, collaborative robots, embodied AI ~$50–90B 14%–20% 10 E-commerceDigitizationOriginal estimates unverified Details \u0026amp; researchMaturity: Mature (editorial)\nProjected size: $14–20T (2040) (unverified)\nEmerging markets \u0026#43; new categories\nSocial commerce, instant retail, cross-border platforms Mature large-scale 7%–9% 11 Digital AdvertisingDigitizationOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: $2.1–2.9T (2040) (unverified)\nAI optimization \u0026#43; privacy shifts\nProgrammatic, AI targeting, retail media networks ~$320–580B range 8%–20% 12 Streaming VideoDigitizationOriginal estimates unverified Details \u0026amp; researchMaturity: Mature (editorial)\nProjected size: Multi-hundred $B (unverified)\nContent \u0026#43; advertising models converging\nCloud gaming, interactive streaming, AI content Mature growth Mid-high 13 Video GamesDigitizationOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: Continued strong growth (unverified)\n~40% of global population may be gamers by 2030\nCloud gaming, metaverse games, AI NPCs Hundreds of $B Mid-high 14 Modular ConstructionManufacturingOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: High potential (unverified)\nLabor shortages \u0026#43; efficiency drivers\nPrefabricated modules, digital construction Early expansion High 15 Future Air MobilityHard TechOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: $75–340B (2040) (unverified)\nChina low-altitude economy is a pillar industry\neVTOL, urban air mobility Early commercialization 10%–20% 16 Obesity \u0026amp; Related DrugsBio \u0026amp; HealthOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: $120–280B (2040) (unverified)\nSemaglutide-class drugs driving rapid growth\nGLP-1 agonists, innovative weight-loss drugs ~$24–100B range 9%–35% 17 Nuclear Fission PowerEnergyOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: $65–150B (2040) (unverified)\nEnergy security \u0026#43; low-carbon dual drivers\nSmall modular reactors (SMR), advanced reactors ~$18B baseline 7%–13% 18 Industrial \u0026amp; Consumer BiotechBio \u0026amp; HealthOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: High potential (unverified)\nNon-medical biotech; sustainable materials\nSynthetic biology, enzyme engineering, bio-based materials Early to growth High 19 Integrated Circuits (China focus)AI FoundationOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: Part of 6 pillars \u0026gt; RMB 10T by 2030 (unverified)\nNDRC emerging pillar industry\nDomestic advanced packaging, Chiplet, automotive chips Rapid China growth 20%\u0026#43; 20 Aerospace (China focus)Hard TechOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: Part of 6 pillars (unverified)\nLarge aircraft \u0026#43; commercial space\nDomestic large aircraft, satellites, LEO constellations Rapid expansion High 21 BiopharmaceuticalsBio \u0026amp; HealthSourced context · 2025Original estimates unverified Details \u0026amp; researchMaturity: Mature (editorial)\nProjected size: Continued expansion (unverified)\nAging \u0026#43; innovation; China pillar industry\nSeparate context: USD 1667.671 billion · 2025. Estimated prescription medicine sales (broader-market context). Worldwide. Global retail and hospital prescription medicine sales at ex-manufacturer prices, from IQVIA MIDAS FY 2025 audited sales (April 2026 vintage) as reported by EFPIA. Includes prescription medicines beyond biological products; this is not a biopharmaceutical-only market size.. This broader prescription-medicine total is context for the Biopharmaceuticals row, not a measurement of that narrower segment. It does not verify the row’s qualitative market label, 15%–25% CAGR, or projected growth. EFPIA / IQVIA MIDAS\nRead industry guideRead reportView industry mind mapView share chart Cell \u0026amp; gene therapy, antibody drugs, AI drug discovery Mature high growth 15%–25% 22 Low-Altitude EconomyHard TechOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: China may exceed RMB 3.5T by 2035 (unverified)\nStrong policy support; NDRC pillar\nDrone logistics, eVTOL, low-altitude air traffic management Early boom in China 30%\u0026#43; 23 New-Type Energy StorageEnergyOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: Part of 6 pillars (unverified)\nMandatory storage pairing with wind/solar\nElectrochemical, compressed air, flywheel storage Rapid China growth 25%\u0026#43; 24 Intelligent RobotsHard TechOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: Part of 6 pillars (unverified)\nNDRC priority; embodied AI synergy\nHumanoid, industrial intelligent, service robots Rapid expansion High 25 Quantum TechnologyAI FoundationOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: ~$85B by 2035 possible (unverified)\nChina future industry; technology breakthrough phase\nQuantum computing, communication, sensing Early R\u0026amp;D / commercialization High 26 Bio-manufacturingBio \u0026amp; HealthOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: High-potential future industry (unverified)\nChina future industry; chemical substitution\nSynthetic bio-manufacturing, cell factories Early High 27 Green HydrogenEnergyOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: High growth (unverified)\nChina future industry; dual-carbon goals\nWater electrolysis, hydrogen trucks, green hydrogen chemicals Early commercialization High 28 Nuclear Fusion EnergyEnergyOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: Long-term huge potential (unverified)\nChina future industry\nTokamak, inertial confinement, demo reactors R\u0026amp;D stage Extremely high (long-term) 29 Brain-Computer InterfaceBio \u0026amp; HealthOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: High-potential future (unverified)\nChina future industry\nInvasive / non-invasive BCI, neuromodulation Early clinical / commercial High 30 Embodied AIHard TechOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: High-potential future (unverified)\nChina future industry; humanoid synergy\nMultimodal models \u0026#43; robot bodies, physical-world interaction Early breakout Extremely high 31 6G CommunicationsDigitizationOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: Commercial after 2030 (unverified)\nChina future industry; IoT foundation\nTerahertz, integrated space-air-ground networks R\u0026amp;D / standards High (long-term) 32 Data Centers \u0026amp; Computing PowerAI FoundationOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: Continued high growth (unverified)\nAI training/inference demand; China computing power network\nLiquid cooling, AI servers, intelligent computing centers High-speed expansion High 33 Renewable Energy (Solar \u0026#43; Wind)EnergyOriginal estimates unverified Details \u0026amp; researchMaturity: Mature (editorial)\nProjected size: Multi-trillion $ install base (unverified)\nCore of global energy transition\nPerovskite cells, large offshore turbines, IBC modules Large-scale mature Mid-high 34 Hydrogen Full Value ChainEnergyOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: High growth (unverified)\nKey path for trucking, steel, chemicals decarbonization\nGreen hydrogen production, storage/transport, fuel cells Early to growth High 35 CCUSEnergyOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: High growth potential (unverified)\nIndustrial decarbonization \u0026amp; negative emissions\nCarbon capture, utilization, geological storage Early High 36 Smart Manufacturing / Industry 4.0ManufacturingOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: ~$1.6T (2030) (unverified)\nCore of China’s manufacturing upgrade\nDigital twins, industrial AI, flexible production lines ~$550–650B (2025) ~19% 37 FintechDigitizationOriginal estimates unverified Details \u0026amp; researchMaturity: Mature (editorial)\nProjected size: Continued expansion (unverified)\nGlobal digital payments \u0026amp; financial inclusion\nDigital payments, embedded finance, blockchain settlement Mature high growth Mid-high 38 Silver Economy / Smart Elderly CareBio \u0026amp; HealthOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: Trillion-RMB market (unverified)\nAging population rigid demand\nAge-friendly products, telemedicine, rehab robots Rapid China growth 15%–20% 39 Gene \u0026amp; Cell TherapyBio \u0026amp; HealthOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: High growth (unverified)\nPrecision medicine breakthroughs\nCRISPR gene editing, CAR-T, stem-cell therapy Rapid commercialization 20%\u0026#43; 40 Synthetic BiologyBio \u0026amp; HealthOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: High potential (unverified)\nMaterials, food, energy applications\nGenetic circuit design, microbial cell factories Early expansion High 41 Advanced MaterialsManufacturingOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: China ~RMB 1.2T by 2030 possible (unverified)\nFuture materials priority direction\nGraphene, superconducting materials, next-gen semiconductors Early to growth High 42 Satellite InternetHard TechOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: High growth (unverified)\nGlobal coverage \u0026amp; emergency communications\nLEO constellations, high-throughput satellite communications Rapid deployment High 43 Autonomous Driving \u0026amp; Connected VehiclesHard TechOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: High growth (unverified)\nSynergy with Robotaxi \u0026amp; low-altitude economy\nLiDAR, domain controllers, vehicle-road-cloud integration Growth stage High 44 Medical AI \u0026amp; Digital HealthBio \u0026amp; HealthOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: High growth (unverified)\nAging \u0026#43; efficiency dual drivers\nAI diagnosis, telemedicine, wearable monitoring Rapid expansion High 45 Green Buildings \u0026amp; Energy EfficiencyManufacturingOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: BIPV ~$42B (2029) etc. (unverified)\nDual-carbon \u0026amp; building energy control\nBIPV, smart building systems BIPV ~$17B (2024) ~20% 46 Logistics Automation \u0026amp; Drone DeliveryManufacturingOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: High growth (unverified)\nE-commerce \u0026amp; instant delivery drivers\nWarehouse robots, last-mile delivery drones Rapid expansion High 47 Digital Twin \u0026amp; Industrial MetaverseManufacturingOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: High potential (unverified)\nSmart manufacturing \u0026amp; training applications\nRead reportView industry mind mapView share chart Factory digital twins, virtual-physical simulation Early to growth High 48 Privacy Computing \u0026amp; Data ElementsDigitizationOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: High growth (unverified)\nData element marketization \u0026amp; compliance\nFederated learning, multi-party computation, data trading Early expansion 20%\u0026#43; 49 Advanced ConnectivityDigitizationOriginal estimates unverified Details \u0026amp; researchMaturity: Growth (editorial)\nProjected size: Long-term high growth (unverified)\nIoT \u0026amp; low-latency application foundation\n5G-Advanced, 6G, space-air-ground-sea integration 5G mature, 6G R\u0026amp;D High 50 Sustainable Food \u0026amp; Alternative ProteinsBio \u0026amp; HealthOriginal estimates unverified Details \u0026amp; researchMaturity: Early (editorial)\nProjected size: High potential (unverified)\nPopulation \u0026#43; climate pressure food innovation\nCultured meat, plant-based proteins, precision fermentation Early commercialization High INDUSTRY DETAILS\nClose × Industry details Explore a growth scenario + Adjust the starting value, annual growth rate, and time period to explore a mathematical scenario.\nStarting value (USD billion) Annual growth rate (%) Years Enter values to calculate\nThe result is a mathematical scenario, not a market forecast.\nAbout the data \u0026amp; source review The full screening list contains 50 industries. Numeric coverage appears when the interactive controls load.\nExcept for the explicitly corrected battery-cell series, original market ranges, growth rates, and projections are unverified research leads. Categories and maturity labels are editorial classifications. Separately sourced records have their own years, geography, and market definitions; these records do not validate the original estimates or make different definitions comparable.\nFile updated October 3, 2026. Eight industries were reviewed on that date; the biopharmaceutical context record retains its September 30 review. Review dates are separate from measurement years and publication dates. Market sizes are in USD billions where available. Filters and scenarios use the displayed values, including the corrected 2022 battery-cell baseline. Calculator midpoints are assumptions, not publisher point forecasts. The growth filter uses a midpoint when both CAGR bounds exist and a lower bound for a one-sided value.\nSeparately sourced records The eight-industry evidence appendix records publishers, links, publication dates, data years, definitions, growth periods, forecast status and correction reasons. Seven audited original ranges carry “Unverified estimate; source and methodology not confirmed.” Nearby source values are kept separate and do not validate the ranges.\nBattery correction: the $98B baseline is a 2022 global battery-cell revenue estimate, not a 2025 energy-storage total. Its $810B–1.1T projection and 12–14% CAGR are 2022–2040 scenarios from McKinsey Global Institute\u0026rsquo;s October 23, 2024 report. This figure is an estimate. See the cell-revenue definition and evidence. The broad industry heading remains editorial.\nThe following pre-existing records remain separate so their boundaries stay visible:\nIndustry row Sourced figure What it establishes Semiconductors USD 795.6 billion in 2025 Annual semiconductor product sales in WSTS\u0026rsquo;s March 6, 2026 release. This is above the original screening range and does not establish its CAGR or forecast. Cybersecurity USD 193.408 billion in 2024 Gartner\u0026rsquo;s historical estimate of worldwide information security end-user spending, from its July 29, 2025 release. The source\u0026rsquo;s later-year numbers are forecasts. End-user spending is not necessarily vendor revenue. Biopharmaceuticals USD 1,667.671 billion in 2025 Broader global prescription-medicine sales at ex-manufacturer prices, reported by EFPIA / IQVIA MIDAS, Key Data 2026, page 14. Includes medicines beyond biological products and does not measure the biopharmaceutical-only market. The downloadable CSV includes separate baseline_* columns for these records, plus claim-level evidence links and estimate notes for the eight audited rows. The corrected battery-cell row supplies its numeric source, publication date, data year, definition and CAGR start/end; empty source fields on other rows still mean the original estimate has not been verified. A sourced baseline or broader-market context is not a forecast validation.\nRead the screening reportResearch methodology ","date":"2026-09-21T00:00:00Z","permalink":"/industries/","title":"Industry Explorer"},{"content":"A provisional 50-industry screening list inspired by global arenas and China policy priorities. Includes an eight-industry evidence appendix and a sourced battery-cell year/scope correction; remaining original ranges are unverified. Interactive database: filter the 50 industries and run a CAGR calculator on the Industries page.\nOver the past two decades, a small set of industries captured an outsized share of global growth and market-value creation. McKinsey Global Institute calls them “arenas of competition”: sectors that combine high growth with intense competitive dynamism.\nMcKinsey Global Institute\u0026rsquo;s 2024 analysis identifies 18 potential future arenas and models $29–48 trillion in combined 2040 revenue across them. Those figures describe McKinsey\u0026rsquo;s collective scenario, not the 50 rows below.\nAt the same time, China\u0026rsquo;s NDRC described six emerging pillar industries and six future industries in March 2026. It estimated that output related to the six emerging pillars alone could exceed RMB 10 trillion by 2030; this is not a forecast for all 12 categories together.\nThis report brings those signals together into a provisional screening list of 50 industries—with representative technologies, indicative figures where available, and commercial or policy notes.\nEvidence status (reviewed October 3, 2026): Eight industries have claim-level notes in the evidence appendix. The battery-cell figures have a source-matched correction: $98B refers to 2022, with a 2022–2040 scenario, rather than an approximately 2025 energy-storage total. The other seven audited original ranges remain unverified, as do unaudited original rows. A nearby number from another publisher does not verify a range or make different market definitions comparable. Each audited row\u0026rsquo;s note applies to its market size, CAGR, projection and numerical notes unless the appendix explicitly establishes otherwise. Separately sourced alternatives retain their own years, definitions and forecast status. The Industries page and CSV carry the same labels and links. See the methodology for the verification standard.\n1. McKinsey’s Future Arenas McKinsey groups high-growth, high-dynamism industries into themes such as:\nAI foundation — AI software and services, semiconductors, cloud Digitization — e-commerce, digital advertising, cybersecurity, streaming, games Electrification — EVs, batteries, nuclear fission Hard tech — robotics, space, shared autonomous vehicles, future air mobility, modular construction New bio-frontiers — obesity drugs, industrial and consumer biotech These arenas are defined by steep S-curves, shifting market shares, and outsized contribution to GDP growth.\n2. China’s Policy Pillars China’s NDRC has prioritized:\nSix emerging pillars: integrated circuits, aerospace, biopharmaceuticals, low-altitude economy, new-type energy storage, intelligent robots Six future industries: quantum technology, bio-manufacturing, green hydrogen and fusion, brain-computer interface, embodied AI, 6G Together they form a clear directional map for the next decade of industrial policy and capital allocation.\n3. How the List of 50 Was Built The 50 entries below expand the McKinsey arenas and Chinese pillars into a broader working list, adding adjacent high-growth segments (data centers, fintech, silver economy, advanced materials, etc.) so readers can scan both global and China-focused opportunity spaces in one place.\nRank Industry Representative Technology Market Size (year as labeled; otherwise ~2025) Projected Size Est. CAGR Key Notes 1 AI Software \u0026amp; Services Large models / Agentic AI, Generative AI ~$230–390B $1.5–2.4T (2030–32) 25%–35% Enterprise AI spend surging; Gartner ~$1.5T AI spend in 2025\nUnverified estimate; source and methodology not confirmed. Evidence 2 Semiconductors Advanced nodes (2nm/3nm), HBM, AI accelerators ~$630–775B $1.5–3.2T (2030) 10%–18% AI servers driving growth; BofA sees $3.2T TAM by 2030\nUnverified estimate; source and methodology not confirmed. Evidence 3 Cloud Services Hyperscale data centers, AI training/inference cloud ~$220B baseline $1.6–3.4T (2040) 12%–17% Infrastructure demand accelerating with AI\nUnverified estimate; source and methodology not confirmed. Evidence 4 Electric Vehicles (EVs) Solid-state batteries, 800V platforms, domain controllers ~$900–990B $1.1–3.7T (2030–35) 10%–15% China leads volume; IEA tracks global adoption\nUnverified estimate; source and methodology not confirmed. Evidence 5 Batteries \u0026amp; Energy Storage Solid-state, sodium-ion, flow batteries ~$98B (2022; defined battery-cell revenues) $810B–1.1T (2040 scenario) 12%–14% (2022–2040 scenario) China new-type storage is a pillar industry\nThis figure is an estimate. Cell revenues, not the full storage industry. Evidence 6 Shared Autonomous Vehicles L4/L5 autonomy, Robotaxi, vehicle-road coordination Early stage $610B–2.3T (2040) High (4%–20%) Waymo, Cruise, Tesla advancing commercialization 7 Space Economy Reusable rockets, LEO constellations, satellite internet ~$550–626B $1–1.8T (2035–40) 7%–10% Launch cost decline; commercial services dominant\nUnverified estimate; source and methodology not confirmed. Evidence 8 Cybersecurity Zero Trust, AI security, quantum-safe cryptography ~$160–240B $590B–1.2T (2040) 8%–20% AI threats + quantum computing driving demand\nUnverified estimate; source and methodology not confirmed. Evidence 9 Robotics (Industrial + Service) Humanoid robots, collaborative robots, embodied AI ~$50–90B $110–205B (2030) 14%–20% China intelligent robots is a pillar industry\nUnverified estimate; source and methodology not confirmed. Evidence 10 E-commerce Social commerce, instant retail, cross-border platforms Mature large-scale $14–20T (2040) 7%–9% Emerging markets + new categories 11 Digital Advertising Programmatic, AI targeting, retail media networks ~$320–580B range $2.1–2.9T (2040) 8%–20% AI optimization + privacy shifts 12 Streaming Video Cloud gaming, interactive streaming, AI content Mature growth Multi-hundred $B Mid-high Content + advertising models converging 13 Video Games Cloud gaming, metaverse games, AI NPCs Hundreds of $B Continued strong growth Mid-high ~40% of global population may be gamers by 2030 14 Modular Construction Prefabricated modules, digital construction Early expansion High potential High Labor shortages + efficiency drivers 15 Future Air Mobility eVTOL, urban air mobility Early commercialization $75–340B (2040) 10%–20% China low-altitude economy is a pillar industry 16 Obesity \u0026amp; Related Drugs GLP-1 agonists, innovative weight-loss drugs ~$24–100B range $120–280B (2040) 9%–35% Semaglutide-class drugs driving rapid growth 17 Nuclear Fission Power Small modular reactors (SMR), advanced reactors ~$18B baseline $65–150B (2040) 7%–13% Energy security + low-carbon dual drivers 18 Industrial \u0026amp; Consumer Biotech Synthetic biology, enzyme engineering, bio-based materials Early to growth High potential High Non-medical biotech; sustainable materials 19 Integrated Circuits (China focus) Domestic advanced packaging, Chiplet, automotive chips Rapid China growth Part of 6 pillars \u0026gt; RMB 10T by 2030 20%+ NDRC emerging pillar industry 20 Aerospace (China focus) Domestic large aircraft, satellites, LEO constellations Rapid expansion Part of 6 pillars High Large aircraft + commercial space 21 Biopharmaceuticals Cell \u0026amp; gene therapy, antibody drugs, AI drug discovery Mature high growth Continued expansion 15%–25% Aging + innovation; China pillar industry 22 Low-Altitude Economy Drone logistics, eVTOL, low-altitude air traffic management Early boom in China China may exceed RMB 3.5T by 2035 30%+ Strong policy support; NDRC pillar 23 New-Type Energy Storage Electrochemical, compressed air, flywheel storage Rapid China growth Part of 6 pillars 25%+ Mandatory storage pairing with wind/solar 24 Intelligent Robots Humanoid, industrial intelligent, service robots Rapid expansion Part of 6 pillars High NDRC priority; embodied AI synergy 25 Quantum Technology Quantum computing, communication, sensing Early R\u0026amp;D / commercialization ~$85B by 2035 possible High China future industry; technology breakthrough phase 26 Bio-manufacturing Synthetic bio-manufacturing, cell factories Early High-potential future industry High China future industry; chemical substitution 27 Green Hydrogen Water electrolysis, hydrogen trucks, green hydrogen chemicals Early commercialization High growth High China future industry; dual-carbon goals 28 Nuclear Fusion Energy Tokamak, inertial confinement, demo reactors R\u0026amp;D stage Long-term huge potential Extremely high (long-term) China future industry 29 Brain-Computer Interface Invasive / non-invasive BCI, neuromodulation Early clinical / commercial High-potential future High China future industry 30 Embodied AI Multimodal models + robot bodies, physical-world interaction Early breakout High-potential future Extremely high China future industry; humanoid synergy 31 6G Communications Terahertz, integrated space-air-ground networks R\u0026amp;D / standards Commercial after 2030 High (long-term) China future industry; IoT foundation 32 Data Centers \u0026amp; Computing Power Liquid cooling, AI servers, intelligent computing centers High-speed expansion Continued high growth High AI training/inference demand; China computing power network 33 Renewable Energy (Solar + Wind) Perovskite cells, large offshore turbines, IBC modules Large-scale mature Multi-trillion $ install base Mid-high Core of global energy transition 34 Hydrogen Full Value Chain Green hydrogen production, storage/transport, fuel cells Early to growth High growth High Key path for trucking, steel, chemicals decarbonization 35 CCUS Carbon capture, utilization, geological storage Early High growth potential High Industrial decarbonization \u0026amp; negative emissions 36 Smart Manufacturing / Industry 4.0 Digital twins, industrial AI, flexible production lines ~$550–650B (2025) ~$1.6T (2030) ~19% Core of China’s manufacturing upgrade 37 Fintech Digital payments, embedded finance, blockchain settlement Mature high growth Continued expansion Mid-high Global digital payments \u0026amp; financial inclusion 38 Silver Economy / Smart Elderly Care Age-friendly products, telemedicine, rehab robots Rapid China growth Trillion-RMB market 15%–20% Aging population rigid demand 39 Gene \u0026amp; Cell Therapy CRISPR gene editing, CAR-T, stem-cell therapy Rapid commercialization High growth 20%+ Precision medicine breakthroughs 40 Synthetic Biology Genetic circuit design, microbial cell factories Early expansion High potential High Materials, food, energy applications 41 Advanced Materials Graphene, superconducting materials, next-gen semiconductors Early to growth China ~RMB 1.2T by 2030 possible High Future materials priority direction 42 Satellite Internet LEO constellations, high-throughput satellite communications Rapid deployment High growth High Global coverage \u0026amp; emergency communications 43 Autonomous Driving \u0026amp; Connected Vehicles LiDAR, domain controllers, vehicle-road-cloud integration Growth stage High growth High Synergy with Robotaxi \u0026amp; low-altitude economy 44 Medical AI \u0026amp; Digital Health AI diagnosis, telemedicine, wearable monitoring Rapid expansion High growth High Aging + efficiency dual drivers 45 Green Buildings \u0026amp; Energy Efficiency BIPV, smart building systems BIPV ~$17B (2024) BIPV ~$42B (2029) etc. ~20% Dual-carbon \u0026amp; building energy control 46 Logistics Automation \u0026amp; Drone Delivery Warehouse robots, last-mile delivery drones Rapid expansion High growth High E-commerce \u0026amp; instant delivery drivers 47 Digital Twin \u0026amp; Industrial Metaverse Factory digital twins, virtual-physical simulation Early to growth High potential High Smart manufacturing \u0026amp; training applications 48 Privacy Computing \u0026amp; Data Elements Federated learning, multi-party computation, data trading Early expansion High growth 20%+ Data element marketization \u0026amp; compliance 49 Advanced Connectivity 5G-Advanced, 6G, space-air-ground-sea integration 5G mature, 6G R\u0026amp;D Long-term high growth High IoT \u0026amp; low-latency application foundation 50 Sustainable Food \u0026amp; Alternative Proteins Cultured meat, plant-based proteins, precision fermentation Early commercialization High potential High Population + climate pressure food innovation 4. What the List Highlights AI foundation cluster (AI software, semiconductors, cloud) is a useful set of related research questions about demand, infrastructure, and value capture. China\u0026rsquo;s policy priorities identify areas for further investigation; the list does not establish their investment outcomes. Electrification and hard tech (EVs, batteries, robotics, space, future air mobility) require separate market definitions before their estimates can be compared. Market sizes and CAGRs vary significantly across sources due to differing definitions; always cross-check the latest primary reports. 5. Background Sources (Not Row-Level Citations) McKinsey Global Institute, Growth industries and the next big arenas of competition (2024): future-arena framework and collective scenario. NDRC, economic press conference (March 2026, Chinese): names of the six emerging pillars and six future industries; aggregate output scenario for the pillars. IEA, Global EV Outlook 2025: EV adoption context; not a citation for the EV market-size row. OECD, The Space Economy in Figures (2023): definitions and measurement context; not a citation for the space market-size row. NIST SP 800-207, Zero Trust Architecture (2020): provided primary context for the cybersecurity category; not a citation for row 8\u0026rsquo;s market-size or CAGR. FDA, Frances Oldham Kelsey and thalidomide: provided primary context for pharmaceutical regulation; not a citation for row 21\u0026rsquo;s growth range. Eight-industry audit updated October 3, 2026. Publication dates, data years and forecast windows are recorded separately below. Unaudited rows retain their original provisional status.\n6. Claim-Level Evidence Appendix All monetary figures below are in US dollars. “This figure is an estimate.” applies to attributed publisher estimates and modeled forecasts; source-checking is not independent certification. A reported historical result, a retrospective estimate, a forecast and a year-on-year change are different measures. Publication date is not data year. For separately cited source entries, CAGRs are publisher-stated rates unless explicitly labeled as an audit calculation; the displayed endpoints may not reproduce their stated precision. No inspected source establishes the provenance or methodology of the seven retained original ranges; alternatives below are not retroactive citations for them. Public summaries may omit detailed methodology. Figures from overlapping industries must not be added as independent value added.\n1. AI Software \u0026amp; Services Original claim: approximately $230–390B around 2025; CAGR 25–35%; $1.5–2.4T in 2030–32. Unverified estimate; source and methodology not confirmed. The original numerical Gartner note is also unverified in this row. The audit found broad-AI figures, not a consistently defined software-and-services-only series, so the row has not been silently replaced.\nIDC, Worldwide GenAI 2025 Predictions, p. 4: publication vintage 2024 (exact day not printed); worldwide AI-solutions spending; $307B for 2025 and $632B for 2028, both forecasts; 29.0% forecast CAGR, 2024–2028. This figure is an estimate. It covers AI solutions overall and is not a software/services subtotal. IDC\u0026rsquo;s August 16, 2024 scope discussion explicitly includes hardware; its approximately 57% software, 24% hardware and 24% services add to 105%, so those inconsistent component percentages must not be used to derive a subtotal. Fortune Business Insights, AI-market release, published May 29, 2025: global AI hardware, software and services; 2024 estimate $233.46B, 2025 forecast $294.16B, 2032 forecast $1,771.62B; 29.2% forecast CAGR, 2025–2032. This figure is an estimate. The approximately $233B figure is for 2024, and the market includes hardware. Neither it nor the 2025 forecast validates the original software/services range. Grand View Research, AI-market report: published June 2026, updated August 2026; global hardware, software and services revenue; 2025 retrospective estimate $390.9B, 2026 estimate $539.5B, 2033 forecast $3,497.3B; 30.6% forecast CAGR, 2026–2033. This figure is an estimate. This later edition cannot be treated as a 2025-vintage source or combined with another publisher\u0026rsquo;s lower endpoint. Disposition: retain the original claim with the unverified label. Broad-AI alternatives require their own broader market labels; they do not supply a source-matched software/services correction.\n2. Semiconductors Original claim: approximately $630–775B around 2025; CAGR 10–18%; $1.5–3.2T in 2030, including the BofA TAM note. Unverified estimate; source and methodology not confirmed. No evidence establishes a common definition, horizon or endpoint-selection method behind that range.\nWSTS, finalized 2025 results, published March 6, 2026: worldwide semiconductor product sales, $795.6B in 2025, a finalized reported full-year result; 26.2% 2025 YoY growth, not CAGR. No multiyear total-market CAGR is supplied. This excludes a combined total for equipment, materials, EDA and other value-chain services. It is a separate sales baseline, above the old range, rather than proof of the original range\u0026rsquo;s provenance. WSTS, Autumn 2025 forecast, p. 2, published December 2, 2025: worldwide semiconductor product sales; 2024 $630.549B, 2025 forecast $772.243B, 2026 forecast $975.460B. The forecast figures are estimates. 22.5% and 26.3% are respective YoY forecasts; no multiyear CAGR. The approximately $630B entry is explicitly 2024, while the later finalized 2025 result supersedes that vintage\u0026rsquo;s forecast. McKinsey, expanded semiconductor-value analysis, published January 15, 2026: worldwide expanded semiconductor value, including imputed captive/OEM chip value and adjusted fabless package/software margins; 2024 estimate $775B, 2030 base-case forecast $1.6T; 13% forecast CAGR, 2024–2030. This figure is an estimate. This is 2024, and the definition differs from WSTS sales; its leading-edge wafer-volume growth is not a total-market revenue CAGR. Disposition: retain the original range as unverified and the WSTS baseline separately. Numerical resemblance does not establish that DEX originally used any of these sources; no direct year correction to an unidentified endpoint is asserted.\n3. Cloud Services Original claim: approximately $220B around 2025; CAGR 12–17%; $1.6–3.4T in 2040. Unverified estimate; source and methodology not confirmed. Public cloud, IaaS-only and infrastructure-cloud series measure different markets.\nGartner, public-cloud spending forecast, published November 19, 2024: worldwide public-cloud end-user spending across SaaS, PaaS, DaaS and IaaS; 2025 total forecast $723.421B, including IaaS-only forecast $211.856B. This figure is an estimate. Growth is 21.5% YoY for the total and 24.8% YoY for IaaS, comparing 2025 with 2024; no multiyear CAGR is supplied. The IaaS subtotal cannot be called the entire cloud-services market. Synergy Research Group, full-year cloud-infrastructure results, published February 5, 2026: worldwide enterprise cloud-infrastructure services, covering IaaS, PaaS and hosted private cloud; 2025 full-year estimate $419B. This figure is an estimate of a completed year, not a forecast. No corresponding multiyear CAGR is supplied; the release\u0026rsquo;s constant-currency growth concerns Q4 year-on-year, not annual CAGR. MarketsandMarkets, Cloud Computing, report TC 1228, published June 2025: global cloud computing, including IaaS/PaaS/SaaS and public/private/hybrid deployments; 2025 forecast $1,294.9B, 2030 forecast $2,281.1B; 12.0% forecast CAGR, 2025–2030. This figure is an estimate. Its CAGR belongs with this larger, differently defined series, not a $220B baseline. Disposition: retain the unverified row. The alternatives do not establish a same-scope correction or validate its 2040 projection.\n4. Electric Vehicles Original claim: approximately $900–990B around 2025; CAGR 10–15%; $1.1–3.7T in 2030–35. Unverified estimate; source and methodology not confirmed. One matching point does not establish the full range or a common forecast period.\nPrecedence Research, publisher-issued dated release, published July 1, 2025: global EV market revenue, covering BEV, PHEV and FCEV and explicit vehicle segments including scooters, motorcycles, three-wheelers, passenger cars, buses and trucks; 2024 base $890.72B, 2025 forecast-period estimate $988.70B, 2034 forecast $2,529.10B; 11% forecast CAGR, 2025–2034. This figure is an estimate. The approximately $990B point is supported under this definition, but not the full DEX interval or 15% endpoint. The current report page, updated September 8, 2026, instead reports a 2026 estimate of $1,097.46B, 2035 forecast of $2,763.17B and a publisher-stated 10.82% CAGR for 2026–2035. Those displayed endpoints imply approximately 10.80% over nine years (audit calculation), so the stated precision is attributed, not independently reproduced. Do not mix the two forecast vintages. IEA, Global EV Outlook 2026: electric-car trends, published May 20, 2026: global spending on electric cars, BEV/PHEV, approximately $860B in 2025, a retrospective estimate. This figure is an estimate. No future revenue CAGR accompanies this claim. The IEA electric-car spending measure is narrower than Precedence\u0026rsquo;s EV-market definition and cannot replace it without changing the measure. Disposition: retain the original range as unverified. The Precedence single-source series and narrower IEA spending measure are separately attributed alternatives, not a reconstructed original range.\n5. Batteries \u0026amp; Energy Storage: Corrected Year and Numerical Scope Original claim: approximately $98B under an approximately 2025 heading, $810B–1.1T in 2040 and CAGR 12–14%. Correction: $98B is a 2022 estimate, and 12–14% is a 2022–2040 modeled-scenario CAGR. The numbers measure defined battery-cell revenues, rather than the complete energy-storage industry. This figure is an estimate. The article row, explorer and CSV now state these boundaries explicitly.\nMcKinsey Global Institute, The next big arenas of competition, published October 23, 2024; full report, PDF p. 140 / printed p. 138, scope on printed pp. 137–140 and pricing basis in endnote 209, printed p. 195: global battery-cell revenues, primarily lithium-ion and sodium-ion for EVs, stationary BESS and consumer electronics; excludes lead-acid and traditional household batteries. 2022 historical estimate approximately $98B; 2040 modeled scenarios $810B–1.1T; 12–14% scenario CAGR, 2022–2040. Values use manufacturer battery-cell prices. These are not an audited total, a 2025 estimate, complete BESS-system revenue or combined mining-to-recycling revenue. The same report treats BESS separately, reinforcing that distinction. IEA, Global battery markets are growing strongly, published February 13, 2026: global lithium-ion battery market, more than $150B in 2025, a retrospective estimate; more than 20% YoY versus 2024, no corresponding future CAGR supplied. This figure is an estimate. Its public opening claim does not fully specify cell-versus-pack valuation, so it is a separate alternative and does not inherit McKinsey\u0026rsquo;s 12–14% CAGR. Reason for revision: the full size/projection/CAGR combination is traceable to a 2022 base and a restricted cell-revenue definition. Correcting those labels is supported; interpolating a new 2025 figure is not. The broad industry heading remains editorial, while the measured market is explicit.\n7. Space Economy Original claim: approximately $550–626B around 2025; CAGR 7–10%; $1–1.8T in 2035–40. Unverified estimate; source and methodology not confirmed. The related sources do not supply one coherent same-year range with that growth band.\nNovaspace, Space Economy Report, 12th-edition release, published January 29, 2026: worldwide space economy, including upstream/downstream activity and space-enabled services; 2025 estimate $626.4B, 2034 forecast $1.01T; 5.5% forecast CAGR, 2025–2034. This figure is an estimate. The separately identified direct space market is $236B in 2025; the broader total is not simply rocket/satellite sales. The source supports the approximately $626B point, but its paired CAGR is 5.5%, not 7–10%. Space Foundation, The Space Report 2025 Q2 release, published July 22, 2025: global commercial space economy plus government budgets, 2024 retrospective estimate $613B; 7.8% YoY in 2024, no matching multiyear CAGR supplied. This figure is an estimate. Its July 25, 2023 release reports $546B for 2022, approximately $550B when rounded, and 8% YoY; that is not 2025 data, and the 2022 estimate was subsequently revised. World Economic Forum / McKinsey, Space: The $1.8 Trillion Opportunity, published April 8, 2024, with definition and estimates and WEF growth summary: global space backbone plus reach activities in other industries; 2023 estimate $630B, 2035 forecast approximately $1.8T; 9% annual forecast growth, 2023–2035, including inflation. This figure is an estimate. This is a different definition, year and forecast horizon from Novaspace, not the other endpoint of one series. Disposition: retain the original as unverified; show Novaspace\u0026rsquo;s paired 2025 estimate and 2025–2034 forecast separately. Do not attach a historical or one-year growth rate to a different publisher\u0026rsquo;s forecast.\n8. Cybersecurity Original claim: approximately $160–240B around 2025; CAGR 8–20%; $590B–1.2T in 2040. Unverified estimate; source and methodology not confirmed. NIST zero-trust material provides category context, not market sizing.\nGartner, information-security spending forecast, published July 29, 2025: worldwide end-user information-security spending, including network security, security services and security software; 2024 historical estimate $193.408B, 2025 forecast $213.025B, 2026 forecast $239.759B. This figure is an estimate. 12.5% growth for 2026 is YoY, not CAGR; no multiyear total-market CAGR is supplied. The approximately $240B forecast is 2026. The site\u0026rsquo;s separate 2024 baseline remains correctly labeled. Gartner, earlier forecast vintage, published August 28, 2024: the same named worldwide information-security spending categories, but a different forecast vintage; 2023 $162.115B, 2024 estimate $183.872B, 2025 forecast $211.552B, 15.1% 2025 YoY, no multiyear CAGR supplied. This figure is an estimate. The approximately $162B value is 2023; it must not be spliced with a later edition\u0026rsquo;s 2026 forecast to produce a 2025 range. MarketsandMarkets, report TC 3485, published June 2025, corroborated by its July 2, 2025 publisher-issued release: global cybersecurity solutions plus professional/managed services; 2025 forecast $227.59B, 2030 forecast $351.92B; 9.1% forecast CAGR, 2025–2030. This figure is an estimate. These are a paired alternative series; the original range and 2040 projection are not established. Disposition: retain the unverified range and the separately sourced 2024 Gartner baseline. The year correspondences are source-specific facts, not proof that those sources generated the old endpoints.\n9. Robotics (Industrial + Service) Original claim: approximately $50–90B around 2025; CAGR 14–20%; $110–205B in 2030. Unverified estimate; source and methodology not confirmed. Hardware, total robot-system revenue, installations and supplier-sample unit sales are not interchangeable.\nABI Research, global robotics outlook: displayed publication July 31, 2025; underlying report PT-3774 dated July 22, 2025. Global robotics hardware across industrial, collaborative, mobile, humanoid and exoskeleton categories; 2025 estimate approximately $50B, 2024 $45B, 2030 forecast $110.7B. This figure is an estimate. The page states 13.8%, rounded to 14%, with a 2024–2030 table label, but $45B to $110.7B over six years implies approximately 16.19% (audit calculation). The stated rate is internally inconsistent and is not accepted as verified. The current body also refers to CES 2026, so the displayed date is not an authenticated July 2025 snapshot. The public catalogue confirms the underlying report\u0026rsquo;s identity and coverage; its paid presentation was not accessed and does not independently resolve this numerical inconsistency here. Mordor Intelligence, robotics-market report: updated July 23, 2026 (original publication date unverified); global industrial/service robot-platform revenue, including hardware, software, integration and RaaS, excluding separately purchased sensors, generic software licenses and unrelated machinery; 2025 estimate $73.64B, 2026 estimate $88.27B, 2031 forecast $218.56B; publisher-stated 19.86% forecast CAGR, 2026–2031. This figure is an estimate. The displayed 2026 and 2031 endpoints imply approximately 19.88% over five years (audit calculation); the source\u0026rsquo;s 19.86% precision is not independently reproduced. The near-$90B figure is 2026, and this broader scope differs from ABI hardware. It does not prove the original $50–90B range or 2030 projection. IFR, World Robotics 2025 industrial release, published September 25, 2025: worldwide industrial-robot installations, 542,000 units in 2024; 2025 forecast 575,000, approximately 6% YoY, not a market-value CAGR. Its World Robotics 2025 service-robot release, published October 7, 2025, reports 2024 supplier-sample unit sales based on 294 service-robot suppliers and explicitly does not extrapolate the sample to the entire industry. Sample composition changes each year, and IFR discourages comparisons across report editions. These are adoption measures; neither establishes a combined industrial/service revenue total or revenue CAGR. Disposition: retain the unverified original figures. Attribute the broader Mordor estimate separately; seek underlying data or publisher clarification before using ABI\u0026rsquo;s precise growth rate.\n","date":"2026-09-19T00:00:00Z","image":"/post/50-high-potential-industries/cover.jpg","permalink":"/post/50-high-potential-industries/","title":"50 High-Potential Industries (2025–2040): Technologies, Markets, and Growth"},{"content":"A source-scoped look at U.S. technology job-cut announcements, AI-attributed reasons, and the limits of causal interpretation. Part 1: Story U.S. technology employers announced substantial job cuts in 2026 even as many companies invested in AI infrastructure. In Challenger, Gray \u0026amp; Christmas\u0026rsquo;s August report, technology accounted for 155,126 announced U.S. cuts from January through August, while employers across all industries cited AI in 116,175 announced cuts over that period. The two figures have different denominators and must not be added or treated as confirmed completed layoffs.\nAI was the most-cited monthly reason for five months beginning in March, but restructuring led in August; the cited reason is what an employer reports, not an independent finding that AI automated every affected role.\nAt the same time, employers announced hiring plans, including in technology. The public narrative split into two camps: one said AI was replacing white-collar work at scale; the other emphasized restructuring and shifts in investment. Neither claim can be established solely from the stated reasons in layoff announcements.\nBoth stories contain partial truths. The useful question is not whether AI “causes” layoffs in a single slogan, but how AI risk narratives, industry restructuring, and labor markets are interacting in practice.\nPart 2: What “AI Risk” Means in This Context 1. Capability Risk vs. Labor Risk Public debate often collapses several different risks into one phrase:\nCapability / safety risk — models that act beyond intended control, amplify errors, or enable large-scale misuse. Economic / labor risk — displacement of tasks and roles, slower hiring for entry-level knowledge work, wage pressure in AI-exposed occupations. Business / capital risk — firms betting so heavily on AI infrastructure that they must cut elsewhere to fund the bet. The relative contribution of these mechanisms cannot be read directly from public layoff totals. Safety debates continue in parallel; labor attribution requires company- and task-level evidence.\n2. The Productivity Promise and the Evidence Gap Some executives anticipate productivity gains from AI, but output must be measured against outcomes rather than only activity, such as code changes or messages handled. The earlier specific account of Meta\u0026rsquo;s internal metrics and proposed team cuts could not be confirmed in a primary publication and has been removed.\nThat pattern matters: announcing AI-driven efficiency is easier than proving it. Until measurement improves, “AI” can function both as a genuine operating shift and as a narrative cover for cost control.\n3. Who Feels the Pressure First Stanford Digital Economy Lab\u0026rsquo;s research points to uneven impact, including weaker employment and hiring for younger workers in AI-exposed occupations; this is an observed association under its methods, not proof that AI alone caused every difference:\nEntry-level and young workers in AI-exposed white-collar roles show weaker employment growth than peers in less-exposed jobs. Pay and job quality may change even when employment does not; the direction and size depend on the occupation and study design. Experienced workers in the same fields often look more resilient so far—suggesting augmentation and selective hiring rather than blanket replacement. Goldman Sachs Research has estimated that on the order of 6%–7% of U.S. workers could be “displaced” over a decade in the sense of needing new employment because of automation—material, but not an overnight wipeout of half the white-collar workforce.\nPart 3: Industry Structure of the Layoff Wave 1. Scale of the Cuts Through August 2026, Challenger counted 155,126 U.S. technology-sector announced cuts, up 52% from the corresponding first eight months of 2025. Across all U.S. industries it counted 116,175 announced cuts citing AI, about 22% of total announced cuts. The technology-sector and AI-reason series overlap but are not interchangeable; these are announcements, not verified individual separations.\nThe previously listed company-specific counts mixed reporting periods, headcount changes and layoff announcements without links to comparable primary disclosures. They are omitted until each example can be checked against the relevant company\u0026rsquo;s dated statement or filing.\n2. Three Forces Running at Once Restructuring and changing demand. Hiring and revenue trajectories differ by firm; a particular reduction does not require an AI explanation.\nCapital reallocation to AI infrastructure. Data centers, chips and power compete for investment budgets, but a direct budget-for-jobs substitution should be documented at the company level.\nTask-level automation. Coding assistants, customer-support bots and internal tools can change staffing needs; displacement and complementary hiring vary across tasks and firms.\n3. Hiring and Firing in the Same Industry Challenger recorded 119,825 announced U.S. hiring plans across all sectors through August 2026, including 19,751 in technology. These are planned hires, not verified jobs filled, and they cannot be subtracted mechanically from announced cuts. The broader economy\u0026rsquo;s net employment effect from AI remains contested.\nPart 4: Challenges and Open Questions 1. Attribution Problem When a company cites AI in a layoff memo, outsiders cannot easily separate:\nroles genuinely automated, roles cut to fund GPUs and data centers, roles that would have been cut in any efficiency drive. Over-attributing everything to AI inflates fear; under-attributing it ignores a real shift in bargaining power and skill demand.\n2. Entry-Level Pipeline Risk If firms use AI to skip junior headcount, they also thin the pipeline that produces senior talent. That is a structural risk for the industry itself: fewer apprenticeships in code, analysis, and operations today mean a thinner expert layer in five to ten years.\n3. Public Expectation vs. Measured Outcome Expectations about AI\u0026rsquo;s job impact and measured employment are different types of evidence. Stanford\u0026rsquo;s research reports no widespread economy-wide displacement in its sample while identifying more specific early-career risks. The gap between expectation and measured outcome warrants continued investigation.\n4. What to Watch Next Whether AI-attributed cuts stay elevated or fade as the “easy” post-pandemic layers are already gone. Whether productivity metrics (revenue per employee, feature velocity, support resolution) catch up with headcount rhetoric. Whether entry-level hiring in AI-exposed fields stabilizes or keeps lagging. How regulation and public opinion respond if displacement concentrates in visible white-collar cohorts. Part 5: Key Takeaways U.S. technology-sector job-cut announcements increased through August 2026 year-on-year in Challenger\u0026rsquo;s series; AI is frequently named across industries, but its exact causal share cannot be derived from announcement reasons. AI risk in the labor market is currently more about task reallocation, slower junior hiring, and narrative cover for restructuring than about a single switch that deletes half of white-collar work overnight. Hiring and cutting can coexist: planned technology hires appear alongside announced cuts, with no guarantee that plans became actual jobs. The hardest problems ahead are measurement (what did AI actually replace?), the entry-level pipeline, and the political gap between public fear and still-mixed macroeconomic data. Sources and scope Challenger, Gray \u0026amp; Christmas: August 2026 job-cuts report — primary source for announced U.S. job cuts and hiring plans through August, including industry and stated-reason breakdowns. It does not establish the causal impact of AI or actual completed layoffs. Stanford Digital Economy Lab: Canaries in the Coal Mine? — observational research using ADP administrative payroll records through June 2026, in the August 12, 2026 revision. It examines AI-exposed occupations and early-career workers and explicitly says it finds no widespread economy-wide displacement. Its findings are descriptive, not causal; this is not a survey of workers. Goldman Sachs Research: AI and the U.S. labor market — the 6%–7% over roughly a decade figure is a scenario for workers potentially needing new employment, not a count of current layoffs. Citation scope reviewed October 3, 2026. Announcement series, administrative-payroll research, and forecast scenarios answer different questions and should not be combined into a single count. Each source retains its stated reporting period.\n","date":"2026-09-19T00:00:00Z","permalink":"/post/ai-risks-layoffs-industry-turbulence/","title":"AI Risks, Industry Turbulence, and the Layoff Wave: What the Numbers Actually Show"},{"content":"A deep dive into the 150-year evolution of the pharmaceutical industry, core market structures, and future challenges. Part 1: Story In 1960, Frances Oldham Kelsey joined the U.S. Food and Drug Administration (FDA) as a medical reviewer.\nOne of her first assignments was an application to market thalidomide in the United States. The sedative was already used in other countries, including by some pregnant patients.\nKelsey and colleagues found the evidence of safety inadequate and asked the applicant for more information, including about reported nerve damage. The FDA did not approve the application.\nThe drug was subsequently linked to severe birth defects in thousands of children internationally. The United States did not approve its commercial sale, although some U.S. patients had received it through investigational distribution.\nThe episode helped build support for the 1962 Kefauver–Harris Drug Amendments. U.S. law had already required evidence of safety for new drugs; the amendments added a requirement to demonstrate effectiveness and strengthened oversight of clinical investigations.\nThe law is a milestone in U.S. drug regulation, but it did not alone determine global R\u0026amp;D costs or create the industry\u0026rsquo;s market structure. Today\u0026rsquo;s long development cycles reflect scientific uncertainty, testing, manufacturing, regulation, and commercialization together.\nPart 2: Industry History The modern pharmaceutical industry developed across more than a century of chemical synthesis, biological research, large-scale manufacturing, clinical testing, and regulation. The timeline below is a simplified guide, not a claim that innovation or manufacturing was confined to the United States and Europe.\n1. Late 19th Century - 1930s: From Dye Workshops to Chemical Synthesis (Disorderly Emergence) Plant Extraction and Accidental Discovery.\nBayer is one example of the connection between synthetic-dye manufacturing and pharmaceutical development. Its company history records a pharmaceutical department in 1888 and the launch of Aspirin in 1899. Bayer joined I.G. Farben in 1925. This company account illustrates one path into pharmaceuticals; it does not establish a single origin for the whole industry or the absence of clinical investigation in that period.\n2. 1940s - 1960s: Mass Production of Antibiotics and the Iron Curtain of Regulation (Foundational Period) Bacterial Culture and Large-Scale Random Screening.\nDuring World War II, Pfizer helped scale penicillin production using deep-tank fermentation; in 1962, the U.S. passed the Kefauver–Harris Drug Amendments amid the thalidomide crisis.\nLarge-scale antibiotic production helped develop fermentation and purification capabilities. The 1962 amendments reinforced the evidence required for new-drug approval in the United States; they did not by themselves eliminate smaller manufacturers or establish an oligopoly.\n3. 1970s - 1990s: The Era of Molecular Biology and Blockbuster Drugs (The Golden Age of Profits) Target-Based Drug Discovery.\nThe lipid-lowering drug Lipitor and the antidepressant Prozac were launched.\nWith a better understanding of receptors and enzymes, target-based discovery became more important alongside screening and empirical methods. Medicines for common chronic diseases supported a blockbuster model, although sales and margins differ markedly by product and company.\n4. 2000s - 2010s: The Biomolecular Revolution and Restructuring of Specialization Recombinant DNA technology and monoclonal antibodies (mAbs).\nProducts such as Humira illustrated the commercial potential of biologics. Researchers also debated the long-run productivity of drug R\u0026amp;D, sometimes calling the observed trend \u0026ldquo;Eroom\u0026rsquo;s Law.\u0026rdquo;\nBiotechnology firms such as Genentech helped bring biologics into mainstream development. Many large firms now combine in-house R\u0026amp;D with licensing and acquisitions; smaller firms and large manufacturers can each participate at multiple stages of development.\n5. 2020s to Present: Multimodal, Precision Medicine, and Computational Drug Development Programmable drugs (mRNA, ADC, CGT) and AI-driven computing (AIDD).\nmRNA vaccines were rapidly developed and launched during the COVID-19 crisis; AI structural biology tools such as AlphaFold have achieved widespread penetration; ADCs (antibody-drug conjugates) have become the mainstay of precision oncology.\nComputational methods increasingly complement laboratory and clinical work. In some disease areas, biomarker-guided treatments supplement broad-market products; computational predictions do not replace experimental validation.\nSimilarly, the pharmaceutical industry, like its history, is extremely complex. Let\u0026rsquo;s break it down simply:\nPart 3: Industry Structure INDUSTRY MAP\nThe pharmaceutical value chain Follow the path from research and production support to medicines, distribution and payment.\nFit to view Expand all Collapse Full screen Read the full text outline PharmaceuticalsUpstream · R\u0026amp;D supportInstruments \u0026amp; reagentsThermo Fisher Scientific Danaher Illumina APIs \u0026amp; intermediatesActive ingredients Chemical intermediates Research \u0026amp; manufacturing servicesCROs CDMOs / CMOs Technology platformsDrug-discovery software Gene-editing tools Midstream · MedicinesBig PharmaPfizer Eli Lilly AstraZeneca BiotechnologyBioNTech Moderna Generic medicinesTeva Sandoz Sun Pharma Specialty medicinesSanten Jazz Pharmaceuticals Downstream · AccessDistributionMcKesson Cencora Cardinal Health Care \u0026amp; retail channelsHospitals Retail and specialty pharmacies Payers \u0026amp; reimbursementPublic health systems Insurers and PBMs DEX editorial map based on the accompanying report. Examples are illustrative, not exhaustive or ranked. Companies can operate across several stages; connections show categories, not verified supplier contracts.\nSources: Biopharmaceutical industry breakdown and source notes. Reviewed 2026-09-29.\nThe pharmaceutical industry is a complex and unique industry characterized by high technological barriers, high compliance thresholds, long cycles, high profit margins, and high risks. The entire industry chain can be clearly divided into three core segments: upstream (R\u0026amp;D support), midstream (pharmaceutical companies and product portfolios), and downstream (distribution, channels, and payers).\nI. Upstream: R\u0026amp;D and Production Support (\u0026ldquo;Water Sellers\u0026rdquo; and Infrastructure) The upstream provides pharmaceutical companies with comprehensive services from target discovery and experimental consumables to contract manufacturing. It is a relatively risk-dispersed and cash-flow-stable \u0026ldquo;water-selling\u0026rdquo; segment. It is divided into four complex parts:\nScientific Reagents and High-End Instruments\nCore Functions and Roles: Providing gene sequencers, high-resolution mass spectrometers, culture media, biochips, and laboratory consumables; it is the \u0026ldquo;arsenal\u0026rdquo; of pharmaceutical R\u0026amp;D. Representative companies/institutions: Thermo Fisher Scientific, Danaher, Merck KGaA, Illumina Active Pharmaceutical Ingredients (APIs) \u0026amp; Intermediates\nCore Functions and Roles: Providing active pharmaceutical ingredients (APIs) and key chemical intermediates. Divided into bulk APIs (such as vitamins and antibiotics) and specialty/high-difficulty APIs. Representative companies/institutions: Zhejiang Medicine, Huahai Pharmaceutical, Lonza, Teva (API division) CXO (Contract Research and Development Organization)\nCore Functions and Roles: CRO (Contract Research Organization) assists in preclinical and clinical trials; CDMO/CMO (Contract Development and Manufacturing Organization) assists in process development and commercial mass production. Representative companies/institutions: WuXi AppTec, Pharmaron, Tigermed, Lonza, Catalent AI Computing Power and Cutting-Edge Technology Platform\nCore Functions and Roles: Provides gene editing tools, protein structure prediction, and AI drug molecular screening platform (AIDD). Representative companies/institutions: Schrödinger, Recursion Pharma, Insilico Medicine II. Midstream: Pharmaceutical Companies and Product Matrix (Core Value Creators and Risk Bearers) The midstream is the main body of the pharmaceutical industry, bearing the most significant financial risks in drug development and being the primary beneficiaries of patent monopoly premiums. It can be broken down into four camps based on business models:\nBig Pharma (Multinational Traditional Giants)\nBusiness Model: In-house research + external mergers and acquisitions (M\u0026amp;A) / licensing introduction. Core Barriers: Extensive global clinical trial compliance capabilities, a global commercial sales network, and abundant cash flow. Representative companies: Pfizer, Eli Lilly, Novartis, Merck (MSD), AstraZeneca. Biotech (Innovative Biotechnology Companies)\nBusiness Model: Focuses on specific cutting-edge targets or new technology platforms (such as ADC, mRNA, CAR-T). They typically lack a mature sales force, and after reaching Phase II clinical trials, they often choose to sell their rights to a Big Pharma or be directly acquired. Core Barriers: The R\u0026amp;D efficiency of top scientists and patent protection. Representative companies: BioNTech, Moderna, BeiGene, Sarepta. Generic Pharma (Generic Drug Companies)\nBusiness Model: Involves rapidly following up on original drug patents after they expire (Paragraph IV challenge), achieving large-scale substitution at extremely low cost. Core Barriers: Extremely high production cost control, ability to tackle complex formulations, and rapid supply chain response capabilities. Representative companies: Teva, Sandoz, Viatris, Sun Pharma. Specialty Pharma\nBusiness Model: Avoid the fierce competition of Big Pharma in areas like oncology and cardiovascular diseases, and instead focus on niche markets such as ophthalmology, dermatology, central nervous system (CNS), or rare diseases. Representative companies: Santen (ophthalmology), Jazz Pharmaceuticals (rare diseases/sleep disorders). III. Downstream: Distribution, Channels, and Payers (Commercial Monetization and Value Loop) After drugs are produced, they must be delivered to end users through an efficient supply chain and monetized through a specific payment system.\nPharmaceutical Distributors\nRole: Warehousing, distribution, financing, and channel management between pharmaceutical companies and end users. Business characteristics: Often relies on scale and efficient working-capital turnover; margins differ by contract and reporting basis. Representative companies: Large U.S. distributors include McKesson, Cencora (formerly AmerisourceBergen), and Cardinal Health. Chinese giants: Sinopharm, Shanghai Pharmaceuticals, China Resources Pharmaceutical, and Jointown Pharmaceutical Group. End Channels\nHospital Market (HCOs): Public/private hospitals. The core market for the vast majority of prescription drugs, injectables, and critical care medications. Retail \u0026amp; DTP Pharmacies: Chain pharmacies, DTP (Direct to Patient) high-value prescription pharmacies, and online pharmacy platforms (such as JD Health and Meituan Pharmacy). Payers \u0026amp; Access – The Industry\u0026rsquo;s \u0026ldquo;Life and Death Power\u0026rdquo;\nPublic financing and procurement (varies by country): In China, the National Healthcare Security Administration (NHSA) plays a central role in reimbursement and procurement; the National Medical Products Administration (NMPA) regulates medicines. European payment systems differ substantially by country. Multi-payer model (US market): Commercial insurance companies (UnitedHealth, Anthem, etc.) and PBMs (Pharmacy Benefit Managers, such as CVS Caremark, Express Scripts) hold the power of life and death over whether a drug can be included in the reimbursement list. PBMs, by controlling a massive number of patient payments, force pharmaceutical companies to offer substantial rebates. IV. The Underlying Operating Mechanism of the Industry Chain: \u0026ldquo;Patent Cliff\u0026rdquo; and Value Cycle The flow of funds and value throughout the pharmaceutical industry chain is driven by the lever of \u0026ldquo;patent protection.\u0026rdquo;\nNew drug R\u0026amp;D investment -\u0026gt; regulatory approval and a period of market exclusivity where applicable -\u0026gt; patent or exclusivity expiry -\u0026gt; potential generic or biosimilar competition -\u0026gt; reinvestment in development. Prices and the duration of effective exclusivity vary by product and jurisdiction.\nBiotech companies undertake high-risk breakthroughs in the early stages; CXOs earn fixed service fees; Large firms may use capital and commercial networks to launch medicines across markets during their remaining effective exclusivity; After patent expiration, generic drug companies and distributors quickly take over the market, significantly reducing healthcare costs, while Big Pharma uses its profits to seek the next biotech target, completing the cycle. Of course, high returns often come with high risks. This complex competitive landscape limits drug development efficiency and triggers various commercial struggles and public health crises.\nWhere medicines are sold The regional sales mix below describes 2025 retail and hospital prescription medicine sales at ex-manufacturer prices, using IQVIA MIDAS figures reproduced in EFPIA’s The Pharmaceutical Industry in Figures — Key Data 2026, page 14. These are geographic sales destinations, not manufacturers’ headquarters or pharmaceutical-company market shares. The DEX CSV is a derivative export of this chart, not independent primary verification.\nMARKET SHARE · Full year 2025\nPrescription medicine sales by region Global retail and hospital prescription medicine sales at ex-manufacturer prices. Regional sales destinations, not manufacturers\u0026#39; headquarters or company market shares.\nWorldwide, grouped by sales market · Share of prescription medicine sales value (%)\nFull screen View the data table Prescription medicine sales by region · Full year 2025 · Share of prescription medicine sales value (%) Sales regionShare North America (US \u0026amp; Canada)54.7%Europe23.7%China6.5%Japan3.5%Latin America4.2%Africa, other Asia \u0026amp; Australia7.4% Download data (CSV) Europe includes Belarus, Turkey, Russia and Ukraine. Other Asia excludes China and Japan. Values retain the publisher\u0026#39;s one-decimal rounding. This is a geographic sales split, not a vendor ranking or total healthcare spending.\nSource: EFPIA / IQVIA MIDAS — The Pharmaceutical Industry in Figures — Key Data 2026 (2026). Reviewed 2026-09-29.\nPart 4: Industry Issues and Challenges 1. The \u0026ldquo;Eroom\u0026rsquo;s Law\u0026rdquo; of R\u0026amp;D Efficiency R\u0026amp;D productivity is difficult to measure: The term \u0026ldquo;Eroom\u0026rsquo;s Law\u0026rdquo; describes a historical observation that research spending rose faster than some measures of output; its rate depends on the period and method used.\nScientific and clinical uncertainty: Some targets are difficult to validate, trials are expensive, and projects can fail at any stage. A single \u0026ldquo;time to market\u0026rdquo; or cost figure does not represent all medicines. 2. Commercial Incentives and Public Health Needs: Severe Mismatch in Public Health Needs Commercial incentives do not always match public-health needs: Expected revenue influences private investment, but clinical feasibility, public funding, and regulation also shape which therapies advance:\nNew Antibiotic Crisis: Given the small dosage, short duration of use, and ease of developing drug resistance, pharmaceutical companies cannot recoup their R\u0026amp;D costs, leading to the bankruptcy of many biotech companies developing antibiotics. The world is facing the threat of untreatable superbugs. Rare and Tropical Diseases Forgotten: Diseases with extremely small patient populations or very low affordability are unlikely to attract commercial R\u0026amp;D funding. 3. Distorted Intermediary Rent-Seeking and High Drug Prices Pricing and intermediary incentives: U.S. pharmacy benefit managers (PBMs) negotiate formularies and rebates for payers. Whether rebate arrangements raise list prices or out-of-pocket costs depends on plan design and contract terms; it cannot be assumed that intermediaries capture the majority of every price increase.\n4. Geopolitics and Global Supply Chain Vulnerability API and intermediate supply: Some important supply chains are geographically concentrated. Their resilience depends on product-specific manufacturing capacity, sourcing and inventories; a blanket country share or impact estimate requires a defined dataset.\nSource notes and primary materials FDA: Frances Oldham Kelsey and the thalidomide application — supports the opening history; it does not establish an exact modern industry-wide R\u0026amp;D cost. Kefauver–Harris Drug Amendments, Public Law 87-781 (1962) — approved October 10, 1962, 76 Stat. 780–796. The original U.S. statute supports the safety/effectiveness and clinical-investigation history; it is not a global law. The 17-page scan ends on a shared page that also begins unrelated Public Law 87-782; that next law is not evidence for drug regulation. Pfizer: company history — self-authored company account of 1941 penicillin-production efforts and 1944 mass production using deep-tank fermentation; not an independent industry-wide history. Bayer: The History of Bayer — current timeline — the current company history was inspected and supports the scoped dye-business, pharmaceutical-department, Aspirin, and 1925 merger example. It is a separate verified destination; the original blocked historical URL remains labelled below. Cencora: AmerisourceBergen becomes Cencora — official August 30, 2023 release confirming the completed name and ticker change; supports the naming claim only. Reference review: 2026-10-03. The accompanying Pharmaceutical Manufacturing working note is a reading list, not an independently verified market dataset. The claim-level sources above are distinguished from research portals, commercial databases, and publication homepages below. Unverified figures have been removed rather than attributed to the note; an accessible homepage does not verify a particular drug, trial, article, or statistic.\nView or download the supplied original Kefauver–Harris Drug Amendments (1962)\nOriginal PDF · U.S. Government Publishing Office, 17 pages\nOpen PDF in a new tab Download original PDFPublisher's copy If the preview is unavailable in your browser, use “Open PDF in a new tab” above.\nThe final scanned page contains the end of Public Law 87-781 and the beginning of unrelated Public Law 87-782. Only the Drug Amendments portion supports this article’s regulatory-history discussion.\nLinks contained in the Pharmaceutical Manufacturing research note These are the note\u0026rsquo;s reading leads, not independent verification of every claim or an endorsement of paid databases. Destinations and scope were reviewed on 2026-10-03. Two links that pointed through Google searches are shown as direct destinations; repeated references share the same underlying source. An unresolved access or content check does not establish that a link is dead.\nRegulation and trials:\nDrugs@FDA — official search portal and direct destination of the note\u0026rsquo;s search link; no particular drug result or approval was verified by checking its homepage. China\u0026rsquo;s CDE — official NMPA Center for Drug Evaluation research portal; use a specific notice or review record for a claim-level citation. European Medicines Agency — redirects to the official English homepage; an institutional research lead, not a specific medicine or regulatory decision. ClinicalTrials.gov — official NLM trial-search and registration portal. Registration is not proof of efficacy or regulatory approval, and the U.S. government does not review or approve the safety and science of every listed study. Chinese Clinical Trial Registry — trial-search and registration lead; no individual trial record or result was verified in this review. EU Clinical Trials Register — legacy EudraCT records — retains EudraCT trials/results and specified third-country records. Ongoing EU/EEA trials are now displayed through CTIS: search for clinical trials. CTIS supplements the legacy register; it does not replace access to its historical records. Neither portal\u0026rsquo;s inclusion verifies a particular trial\u0026rsquo;s results. Scientific and commercial research:\nDrugBank — commercial drug-data and biopharma-intelligence lead; no licensed dataset, individual molecule claim, or market figure was verified. IUPHAR/BPS Guide to Pharmacology — target, ligand, and pharmacology research lead. Its homepage stated that registration is required to use the website; unrestricted access should not be assumed, and no registration was attempted. PubChem — NCBI chemical-information search portal; no specific compound record or scientific claim was verified by inspecting the homepage. DXY Insight — commercial research-database landing page with trial, registration, marketed-drug, and company-data modules; underlying licensed data and report figures were not inspected. FiercePharma — news publication homepage, not a specific article or primary historical, scientific, or market source. FierceBiotech — news publication homepage; a specific article and its underlying evidence are needed for a claim-level citation. Endpoints News — original reference — Content not confirmed in the 2026-10-03 review. The exact homepage returned HTTP 403; related event pages did not verify its content or provide an equivalent replacement. This access block does not establish deletion. BioWorld — direct publication destination of the note\u0026rsquo;s search link; no particular paid article or dataset was independently inspected. Historical reading:\nNLM biography of Frances Oldham Kelsey — supports the historical thalidomide-application and regulatory-career account. It does not establish that thalidomide has never subsequently been approved for any indication. Bayer\u0026rsquo;s historical article — original reference — Content not confirmed in the 2026-10-03 review. Bayer\u0026rsquo;s bot-access block prevented inspection of this path; it is not established to be deleted. Current Bayer history timeline — separately inspected company history used for the scoped historical example above; this is a verified current destination, distinct from the original reference. Bayer homepage — redirects to the English corporate homepage; company context only, not evidence for a particular historical event. Pfizer company history — the same company-authored account cited above, not a second independent historical source. ","date":"2026-08-21T00:00:00Z","image":"/post/pharmaceutical-industry/cover.jpg","permalink":"/post/pharmaceutical-industry/","title":"The Pharmaceutical Industry: History, Structure, and Challenges"},{"content":"What DEX publishes, how the research is produced, and how to suggest corrections or collaborations. About DEX DEX is an independent, student-led business research and publishing project. I study business and accounting and use this site to turn long-form research into practical, readable industry intelligence.\nThe goal is not to predict which company or asset will win. It is to build a durable body of work that explains how industries developed, how their value chains and business models work, where credible growth signals exist, and which risks could change the outcome.\nWhat This Site Publishes Industry reports covering history, market structure, value chains, economics, competition, regulation, and risk. Structured datasets and tools that make assumptions visible and let readers inspect or test the numbers. Video explainers that translate the core findings into a faster format. Each format supports the same research system: one research project can become an article, a dataset, a visual explanation, and a video without changing the underlying evidence.\nResearch Methodology 1. Define the question and scope Every report begins with a specific question, a geographic and time scope, and a set of definitions. When different sources use different market definitions, the report treats them as different estimates rather than combining them into false precision.\n2. Prioritize stronger sources Sources are generally considered in this order:\nGovernment agencies, regulators, statistical offices, and company filings. International institutions, peer-reviewed research, and established industry bodies. Reputable research firms, major financial institutions, and well-sourced journalism. Secondary summaries used only as discovery aids or context. Published research should link material claims to specific sources, including the source\u0026rsquo;s date, geography, market definition, and unit. A source list alone does not establish which source supports a particular estimate. The 50-industry screening dataset retains its original market sizes, growth figures and projections as unverified research leads. Separate historical baselines have been source-checked for semiconductors and cybersecurity; broader prescription-medicine context accompanies the biopharmaceutical row. Each added record includes its own year, unit, geography, market definition, publication date and review date. A partial or context status applies only to that separate record: it does not verify the original screening estimate or forecast. Original-source fields remain empty until the corresponding original claim is verified. A contextual reading link is also not evidence for the row\u0026rsquo;s numbers.\n3. Separate evidence from interpretation Reported figures, editorial classifications, scenarios, and DEX interpretations are not the same thing. Tools on this site label calculated outputs as scenarios. Categories and maturity labels are editorial judgments intended to make comparison easier; they are not official classifications.\n4. State uncertainty and limitations Forecasts are sensitive to definitions, base years, exchange rates, regulation, and adoption assumptions. Missing data is shown as missing rather than silently replaced. A dataset update date records when the file was edited, not when each underlying source was published or checked.\n5. Review, update, and correct As reports are revised, key numbers and links should be checked against cited material; where that review has not happened, the limitation is stated alongside the data. Substantive corrections or new data may be reflected in the report\u0026rsquo;s last-updated date. If you find an error, please send the exact page, claim, and supporting source to dex222444@gmail.com.\nUse of AI AI tools may assist with discovery, outlining, translation, formatting, code, and production. They are not treated as evidence. Factual claims should be checked against the cited public sources, and final editorial responsibility remains with DEX.\nIndependence and Disclaimer Unless a page clearly states otherwise, the content is independently produced and no company paid for inclusion. Future sponsored or commissioned work will be labeled. Nothing on this site is investment, legal, accounting, or medical advice.\nContact and Collaboration Corrections, source suggestions, dataset feedback, sponsorship inquiries, and research collaborations are welcome.\nEmail: dex222444@gmail.com Proton Mail: qizhangdong325@proton.me YouTube: @ZeRuiDong GitHub: BAOZ121 Website: thedexs.com Support the Research If this work is useful, you can support it by sharing a report, subscribing on YouTube, starring the website repository, or using GitHub Sponsors. These actions help fund more research and better datasets.\n","date":"2026-08-12T00:00:00Z","permalink":"/about/","title":"About \u0026 Methodology"},{"content":"A deep dive into the 60-year evolution of cybersecurity, upstream-midstream-downstream value chains, the four major market camps, and core industry bottlenecks. Part 1: Story An often-repeated account describes attackers using a connected aquarium sensor as an entry point to a casino network. It illustrates how an overlooked device can expand a network\u0026rsquo;s attack surface.\nThe public retellings listed below trace to an account by then-Darktrace CEO Nicole Eagan; they are not independent corroboration of one another. They do not supply enough independently verifiable information to confirm the casino, incident date, defenses, or amount of data taken. Treat it as a vendor-origin illustrative anecdote, not a documented case study or a quantitative measure of cyber risk.\nThis story brings us to an important subject—\nWelcome, I\u0026rsquo;m Dex. Welcome to my industry report. Before we dive in, let\u0026rsquo;s take a look at a brief history of the industry.\nPart 2: Industry History 1970s: ARPANET and Creeper Early networked experiments such as Creeper and Reaper are part of the history of self-propagating programs and countermeasures. Assigning a single unqualified \u0026ldquo;first worm\u0026rdquo; or \u0026ldquo;first antivirus\u0026rdquo; to either program obscures differences in definitions and surviving records.\n1980s: Birth of Commercial Antivirus Software Commercial antivirus products emerged in the 1980s, and their precise chronology depends on how a \u0026ldquo;first\u0026rdquo; product is defined. The transition from standalone personal computers to connected business networks expanded the range of threats and defenses.\nKey Turning Point: Mid-1990s During the 1990s, more connected personal computers and business networks created additional opportunities for malicious code, denial-of-service attacks, and intrusions. This is a directional overview rather than a complete chronology of named incidents.\n2000s (2000–2009): Commercial and Organized Cybercrime The 2000s marked a transitional period for cybersecurity threats, shifting from mere pranks to serious, organized, and commercially driven criminal activity. Driven by core threat data and landmark incidents, people began to realize the vulnerabilities inherent in the early digital age during this explosion of cybersecurity incidents:\nFast-spreading worms (2000–2004): Incidents such as ILOVEYOU and SQL Slammer showed how email and software vulnerabilities could cause rapid, widespread disruption. Exact global infection and loss estimates vary by source and method.\nRise of Commercial Cybercrime (Mid-to-Late 2000s): Hacker motivations shifted from technical boasting to economic gain.\nBotnets and data breaches: Compromised computers were increasingly used for spam and fraud, while payment-card incidents highlighted the costs of weak data protection. Incident totals and exposed-record counts require case-specific primary reports. Distributed denial of service: High-profile incidents exposed the operational costs of making online services unavailable; dollar-loss estimates are not directly comparable between incidents.\nEspionage and advanced intrusions: Public disclosures such as Operation Aurora increased attention to persistent, targeted threats; cyber espionage itself predated the incident.\n2010 to Present: Cloud-Native \u0026amp; AI-Driven Era Between 2010 and 2019, the global cybersecurity landscape evolved from simple virus defense to geopolitical cyber warfare, massive data breaches, and ransomware ecosystems (e.g., Stuxnet, Sony Pictures hack, WannaCry).\nSince 2020, the industry has undergone profound transformation characterized by supply chain attacks, open-source vulnerabilities, critical infrastructure ransomware, and AI-driven threats. The industry spans nearly six decades of history.\nPart 3: Industry Value Chain INDUSTRY MAP\nThe cybersecurity value chain Explore the foundations, products and services that connect security suppliers to customers.\nFit to view Expand all Collapse Full screen Read the full text outline CybersecurityUpstream · FoundationsCloud infrastructureAWS Microsoft Azure Alibaba Cloud Core componentsCryptographic libraries Specialized chips Threat intelligenceIndicators and telemetry Threat-data feeds Midstream · ProductsEndpoint \u0026amp; workloadsCrowdStrike Network securityPalo Alto Networks Fortinet Identity \u0026amp; accessOkta CyberArk Security analyticsSplunk / Cisco Downstream · DeliveryIntegration \u0026amp; resaleSystem integrators Value-added resellers Managed securityMSSPs Customer security teams Incident responseMandiant Consulting and response teams DEX editorial map based on the accompanying report. Examples are illustrative, not exhaustive or ranked. Companies can operate across several stages; connections show categories, not verified supplier contracts.\nSources: Cybersecurity industry breakdown and source notes. Reviewed 2026-09-29.\nLet\u0026rsquo;s briefly summarize the structure of the cybersecurity industry.\nUpstream: Foundational Infrastructure \u0026amp; Threat Intelligence The upstream sector serves as the cornerstone of the entire security industry, supplying midstream vendors with computing power, fundamental components, and critical threat intelligence:\nCloud Infrastructure \u0026amp; Computing Power: AWS, Microsoft Azure, and Alibaba Cloud are examples of infrastructure on which security services may run. Foundational Core Components: Deep-tech companies mastering cryptographic algorithm libraries and high-precision processing chips (FPGAs, ASICs). Threat Intelligence Providers: Acting as the \u0026ldquo;radar\u0026rdquo; of the industry. They gather Indicators of Compromise (IOCs) globally and package data feeds to power midstream security engines. Midstream: Core Products \u0026amp; Solutions Midstream vendors directly face hacker attacks and provide defensive tools to clients. Based on modern enterprise IT architecture, midstream is categorized into four major segments:\nEndpoint \u0026amp; Workload Security: Antivirus, endpoint detection and response (EDR), and workload protection address different risks; CrowdStrike is one example of an EDR vendor. Network \u0026amp; Perimeter Security: Firewalls, secure access service edge (SASE), and segmentation coexist; Palo Alto Networks and Fortinet are examples of suppliers. Identity \u0026amp; Access Management (IAM): Identity is an important control alongside devices and networks, not the sole perimeter; Okta and CyberArk are examples of suppliers. Security Operations \u0026amp; Data Analytics: Systems such as Splunk (acquired by Cisco) aggregate and investigate security events. Monitoring and analytics products differ in scope and are not interchangeable. Downstream: Channels \u0026amp; Security Services Because security products are complex, a massive downstream service market has emerged:\nSystem Integrators \u0026amp; VARs: Service providers assisting enterprises with procurement, installation, and basic hardware configuration. Managed Security Service Providers (MSSP): Addressing the global shortage of security engineers by directly managing enterprise security operations 24/7 on a subscription basis. High-End Consulting \u0026amp; Incident Response: Teams like the Big Four or Mandiant providing penetration testing and emergency rescue during ransomware attacks. Summary: Simply put, upstream provides materials and infrastructure; midstream builds weapons and trains troops; downstream handles tactical deployment and command.\nPart 4: Industry Market Landscape MARKET SHARE · Full year 2024\nModern endpoint security revenue share IDC modern endpoint security segment; not the whole cybersecurity market.\nWorldwide · Share of modern endpoint security revenue (%)\nFull screen View the data table Modern endpoint security revenue share · Full year 2024 · Share of modern endpoint security revenue (%) CompanyShare Microsoft28.6%Other71.4% Download data (CSV) IDC estimates reproduced on a vendor\u0026#39;s official website. Other is the residual share of all remaining suppliers. Revenue share does not measure customer counts or product effectiveness. Historical 2024 snapshot.\nSource: IDC, reproduced by Microsoft — Microsoft ranked number one in modern endpoint security market share third year in a row (2025-08-27). Reviewed 2026-09-29.\nThis is a historical 2024 worldwide modern-endpoint-security revenue snapshot, using IDC estimates reproduced by Microsoft on August 27, 2025. The underlying IDC report was not directly retrieved in this review. The chart and its DEX CSV export do not measure the whole cybersecurity market, customer counts, or product effectiveness; the CSV is a derivative of the same chart data, not independent evidence.\nThere is no single comparable \u0026ldquo;cybersecurity market\u0026rdquo; figure without specifying geography, year, whether services and cloud infrastructure are included, and the research method. The supplied 2022 Menlo Ventures map identifies product categories and companies; it does not substantiate this article\u0026rsquo;s earlier $250B–$300B size, $500B forecast, CAGR, or vendor-share estimates. Those numbers have been removed pending a traceable dataset.\nThe global market is divided into four major camps:\n1. Cross-Domain Tech Giants Key Players: Microsoft (Defender / Sentinel), Google (Mandiant) Competitive Moat: Leveraging software ecosystems and distribution to integrate security products. 2. Pure-Play Security \u0026ldquo;Big Three\u0026rdquo; Key Players: Palo Alto Networks, CrowdStrike, Fortinet Product focus: Palo Alto Networks sells network and cloud security; CrowdStrike emphasizes endpoint and cloud protection; Fortinet sells network-security appliances and software. These are illustrative positions, not audited share rankings. 3. Traditional IT \u0026amp; Hardware Giants Key Players: Cisco, IBM, Trend Micro Product focus: Enterprise networking and IT software, with acquisitions used to expand security portfolios. 4. Niche Specialists Key Players: Zscaler (Zero Trust / SASE), Cloudflare (Edge Protection), Okta (Identity) Competitive Moat: Dominating specific technical niches to attract top-tier enterprise clients. Two Trends Shifting Market Dynamics Vendor Consolidation: Some buyers prefer fewer integrations and vendors; the outcome depends on their existing architecture and procurement needs. Cloud and AI: Cloud-delivered tools and automated detection are growing areas of investment, while hardware controls still serve important use cases. Part 5: Industry Challenges \u0026amp; Bottlenecks Despite intense competition, the industry faces fundamental challenges:\n1. Asymmetric Warfare Defenders must protect every single endpoint and password, whereas attackers need only find one weak link using AI tools. Defenders remain in a reactive cycle while AI drastically lowers attack costs and sky-rockets defense expenses.\n2. Compliance-Driven \u0026ldquo;Shelfware\u0026rdquo; Many non-critical enterprises buy security tools primarily to pass audits rather than stop hackers, creating a market flooded with \u0026ldquo;shelfware\u0026rdquo; installed for inspection and then ignored.\n3. Tool Fragmentation \u0026amp; Alert Fatigue Large organizations can struggle with overlapping tools and alert volumes. A universal average number of tools or false-positive rate would need a defined sample and measurement method.\nValue Chain Bottlenecks Upstream: Shared software components can create widespread exposure, as CISA\u0026rsquo;s Log4j advisories illustrate. Midstream: Ongoing research, complex integrations and operational resistance can slow adoption of zero-trust approaches. Downstream: Labor-intensive services face staffing and incident-response challenges; margins vary across businesses. This competition appears to be a death spiral with no end in sight; as for how the cybersecurity industry will evolve—whether a super-giant akin to Google will emerge, or if the advent of AI will trigger a commercial tsunami—only time will tell.\nThat concludes my industry report. If you found it interesting, please like the video and subscribe to my channel. I’m Dex—see you next time.\nSource notes and primary materials NIST SP 800-207: Zero Trust Architecture (2020) — August 2020 architectural guidance supplied with the working materials. This Special Publication defines a zero-trust approach; it is not a product certification, market-size dataset, or company-share ranking. Menlo Ventures: Cybersecurity Market Map (2022) — Menlo Ventures, 2022, 2 pages. A dated category/vendor map, not a revenue-share dataset, current ranking, or endorsement of the named vendors. CISA: Apache Log4j vulnerability advisory AA21-356A — archived primary advisory, revised December 23, 2021, supporting the historical software-library vulnerability example. Direct retrieval returned HTTP 403; its official-domain indexed text was inspected. The 2021 mitigation instructions should not be treated as current operational advice. This advisory is separate from the unresolved CISA guidance-page link in the reading list. Reference review: 2026-10-03. The accompanying Network Security document is a research reading list, not primary verification for the anonymous casino account or the removed market figures. The incident specifics remain unverified in public primary records. Access checks and topic matches do not independently verify every statement in a source.\nView or download the supplied original NIST SP 800-207: Zero Trust Architecture (2020)\nOriginal PDF · National Institute of Standards and Technology, 59 pages\nOpen PDF in a new tab Download original PDFPublisher's copy If the preview is unavailable in your browser, use “Open PDF in a new tab” above.\nThe Menlo Ventures Cybersecurity Market Map PDF is a separate 2-page, 2022 category/vendor map, available directly from Menlo Ventures. It is not the 59-page NIST publication previewed above and does not report revenue shares. It is not hosted here because permission to redistribute that copyrighted PDF has not been established.\nLinks contained in the Network Security research note These are the supplied note\u0026rsquo;s research and video links, reviewed for destination and scope on 2026-10-03. They have not all been independently verified and should not be read as endorsements or claim-level primary evidence. An unresolved access or content check does not establish that a link is dead. Original references are retained so readers can distinguish them from any separately checked destination.\nIncident and industry background:\nThe Hacker News: aquarium thermometer incident — April 16, 2018 retelling of then-Darktrace CEO Nicole Eagan\u0026rsquo;s anonymous casino account; not independent incident verification. Entrepreneur: casino thermometer account — April 14, 2021 retelling citing a 2018 account of the same Darktrace story; not a second independent case or corroboration. Privacy International: aquarium thermometer account — April 15, 2018 summary of the same Darktrace conference account; does not independently identify the casino or confirm incident details. Cyber Magazine: history of cybersecurity — October 4, 2021 secondary overview. Its historical forecasts and broad “first” claims are not verified current market data or primary evidence of priority. History of Information: first computer virus — Creeper history entry drawing on an earlier Wikipedia account; a secondary reading lead, not primary evidence for contested “first virus” terminology. Wikipedia: Creeper and Reaper — encyclopedia synthesis for orientation and underlying references; not primary historical verification. KMC Controls: Creeper and Reaper — July 1, 2024 vendor background article, itself citing a vendor explainer. Its “BBM” spelling is not evidence for the organization\u0026rsquo;s name; do not use it as a primary historical authority. Atari Magazine: Computer Viruses And The ST — archive of George Woodside\u0026rsquo;s May 1990 START article about ST viruses and VKILLER. Historical descriptions and software advice retain their 1990 context. Atari Mania: ST Virus Killer — legacy URL redirects to a catalogue entry attributing the program to 1991; does not establish the earliest antivirus product. Carifred: UVK — Ultra Virus Killer for Windows — modern product whose publisher dates its start to 2010. It is different from the historical Atari Ultimate Virus Killer and cannot substantiate an Atari-era antivirus claim. Wikipedia: ESET NOD32 — encyclopedia product-history lead; inclusion does not verify a specific chronology or company metric. Internet Archive: Malware Museum — original reference — Content not confirmed in the 2026-10-03 review. The collection could not be retrieved or inspected; this does not establish deletion. Wikipedia: G Data CyberDefense — encyclopedia company-history lead, not primary evidence for commercial-antivirus “firsts” or current company metrics. Wikipedia: security-hacking incidents — chronological reading list; specific incident claims require their underlying records. Purdue TAP: hackers of the 2000s — August 27, 2024 historical overview; institutional hosting does not make a retrospective a primary incident record. Cofense: history of phishing — June 6, 2023 vendor-authored historical background, not original incident evidence. Wikipedia: computer virus and worm timeline — orientation and reference-finding only; the inspected page also carried a cleanup warning about entry noteworthiness. CISA: Log4j guidance — original reference — Content not confirmed in the 2026-10-03 review. The exact guidance URL returned HTTP 403, and its content or current destination was not established. The separately cited AA21-356A advisory does not verify this specific page. Wikipedia: Sony Pictures hack — encyclopedia background; specific incident and attribution claims require underlying official evidence. Wikipedia: WannaCry attack — encyclopedia background, not a primary incident report or verified loss estimate. Market and technical references:\nMordor Intelligence: cybersecurity market — commercial report landing page with a 2026–2031 outlook at review. Its changing proprietary estimates do not restore the removed market figures; the paid report was not independently inspected. Menlo Ventures: market map PDF — 2-page 2022 category/vendor map, not revenue shares or a current company ranking. Cloudflare: next-generation firewalls — vendor-authored technical explanation of NGFW features; does not establish market share or product effectiveness. Cybersecurity Ventures / Cybercrime Magazine — publisher homepage and research-discovery lead, not a particular report or traceable dataset for a market number. U.S. Securities and Exchange Commission — official research portal for filings and other materials; a specific filing is needed to substantiate an issuer\u0026rsquo;s financial or cybersecurity metric. IBM: a decade of global cyberattacks — Mike Elgan\u0026rsquo;s retrospective covering 2013–2023; background reading rather than original evidence for all incident figures it recounts. CSO: Target breach timeline search — original reference — Content not confirmed in the 2026-10-03 review. This is a search URL, not a verified direct article; neither the search page nor an underlying timeline was inspected. NIST SP 800-207 PDF — August 2020, 59 pages, architectural guidance; no market size, company-share ranking, or product certification. Video references from the note (third-party material, not licensed for reuse here):\nVideo 1 — original reference — Content not confirmed in the 2026-10-03 review. Title, channel, and topic remain unconfirmed after retrieval attempts; the video is not established to be deleted. Video 2 — original reference — Content not confirmed in the 2026-10-03 review. Title, channel, and topic remain unconfirmed after retrieval attempts; the video is not established to be deleted. Video 3: IBM Technology — Zero Trust Explained in 4 mins — canonical same-ID page identifies IBM Technology, September 10, 2021, and a 3:42 runtime. Title, description, and chapter labels were inspected; the full audiovisual content and transcript were not independently reviewed. Educational reading lead only. Original short-link reference retained for provenance. Video 4 — original reference — Content not confirmed in the 2026-10-03 review. Title, channel, and topic remain unconfirmed after retrieval attempts; the video is not established to be deleted. Video 5 — original reference — Content not confirmed in the 2026-10-03 review. A title-only search result was insufficient to verify the source; a Google unusual-traffic CAPTCHA then blocked inspection. Channel and video content remain unconfirmed, and deletion has not been established. ","date":"2026-08-11T00:00:00Z","image":"/post/cybersecurity-industry-report/cover.jpg","permalink":"/post/cybersecurity-industry-report/","title":"Cybersecurity: An Industry Map From Network Defenses to Zero Trust"},{"content":"A simplified VR/XR industry map, historical context, and what company filings can and cannot establish. What does it mean for a company to spend heavily on an emerging computing platform? Meta\u0026rsquo;s Reality Labs segment reported an operating loss of $19.193 billion for full-year 2025 in its unaudited results release; that is a segment accounting result, not the net loss of Meta Platforms as a whole. Meta acquired Oculus in 2014 and continues to invest in virtual reality, augmented reality, and related devices.\nVR, as the name suggests, is Virtual Reality.\nHistorical context In 1935, more than 90 years ago, Stanley G. Weinbaum\u0026rsquo;s story Pygmalion\u0026rsquo;s Spectacles imagined an immersive experience through glasses; that literary vision preceded the hardware.\nIn the 1960s, Morton Heilig\u0026rsquo;s Telesphere Mask was an early head-mounted stereoscopic display concept. Its historical significance does not make it mechanically equivalent to a modern tracked headset.\nIn his 1965 paper The Ultimate Display, computer graphics pioneer Ivan Sutherland described an ambitious vision for interactive computer displays. A later tracked head-mounted display was presented in 1968; the 1965 paper is historical context, not evidence for the later device\u0026rsquo;s mechanical design.\nJaron Lanier and VPL Research helped popularize the term \u0026ldquo;virtual reality\u0026rdquo; and develop early commercial systems and data gloves. Claims about a single inventor or first company are more contested than this simplified history can establish.\nOver several decades, VR has experienced cycles of enthusiasm and retrenchment. Meta and Apple have invested in different visions of spatial computing, but whether headsets become a mass-market computing platform remains an open question.\nNow that the story is finished, let\u0026rsquo;s look at today in 2026. What does the development of the VR industry look like? Let\u0026rsquo;s simply dissect this industry.\nIndustry chain INDUSTRY MAP\nThe VR / XR value chain Explore the components, headsets and applications behind immersive computing.\nFit to view Expand all Collapse Full screen Read the full text outline VR / XRUpstream · ComponentsComputeQualcomm XR platforms Apple silicon Displays \u0026amp; opticsMicro-OLED displays Pancake optics Tracking \u0026amp; powerCameras and motion sensors Eye / hand tracking Batteries Midstream · HeadsetsDevice brandsMeta Quest Apple Vision Pro Sony PlayStation VR Pico HTC Manufacturing \u0026amp; supplyGoertek Luxshare Precision Downstream · ExperiencesPlatforms \u0026amp; storesSteamVR Meta Quest Store App Store Consumer usesGaming Social experiences Video and entertainment Enterprise usesSimulation and training Design and digital twins Medical education DEX editorial map based on the accompanying report. Examples are illustrative, not exhaustive or ranked. Companies can operate across several stages; connections show categories, not verified supplier contracts.\nSources: Report sources and original materials. Reviewed 2026-09-29.\nFirst is the upstream of this industry:\nChips: Qualcomm (XR2 series), Apple (M-series + R-series chips).\nDisplay/Optics: Sony (Micro-OLED), BOE, Sunny Optical (Pancake lenses).\nSensors \u0026amp; Batteries: Suppliers of various sensors tracking eye movements and gestures.\nThen let\u0026rsquo;s look at the midstream of this industry (hardware terminal manufacturing): Who makes the headsets we buy?\nBrands: Meta (Quest series, affordable mass adoption route), Apple (flagship Vision Pro, high-end spatial computing route), Sony (PSVR products, focusing on console gaming ecosystem), Pico (under ByteDance), HTC.\nContract manufacturers and component suppliers include Goertek and Luxshare Precision. Their consolidated financial reports disclose company-wide results, but do not by themselves reveal or predict the unit sales of an individual headset brand.\nFinally, the downstream of this industry (content and application scenarios): What are its application scenarios?\nC-end ecosystem: SteamVR, Meta Quest Store, App Store. Core applications are gaming, social (VRChat), and video/film.\nB-end applications: Originally used for Air Force flight training and space navigation, later expanded to other fields, such as medical simulation training, industrial digital twins, automobile design, etc.\nCompetition and adoption MARKET SHARE · Full year 2024\nGlobal AR/VR headset shipment share AR/VR headset market tracked by IDC; consumer and commercial shipments. This is broader than VR alone.\nWorldwide · Share of headset unit shipments (%)\nFull screen View the data table Global AR/VR headset shipment share · Full year 2024 · Share of headset unit shipments (%) CompanyShare Meta74.6%Apple5.2%Sony4.3%ByteDance (Pico)4.1%XREAL3.3%Other8.5% Download data (CSV) Other is calculated as 100% minus the five published shares. Shipments are neither installed base nor retail sell-through or revenue. This historical 2024 snapshot is not a 2026 estimate. Do not mix it with Counterpoint\u0026#39;s narrower VR-only estimate.\nSource: IDC — Growth Expected to Pause for AR/VR Headsets, according to IDC (2025-03-25). Reviewed 2026-09-29.\nXR (Extended Reality) is often used as an umbrella term for VR, AR and MR. Categories overlap; market-share claims require carefully defined devices and time periods.\nMeta sells the Quest series and Apple sells Vision Pro, but a 2026 worldwide market-share comparison requires a specified device category, period, geography and shipment source. Meta\u0026rsquo;s segment reports include multiple products and do not split headset and smart-glasses sales sufficiently to establish the relative revenue of those categories here.\nApple positions Vision Pro as a premium spatial-computing device. Its market share and competitive effects should not be inferred from a list price alone.\nNew platforms and lighter glasses may compete for attention, but future product adoption cannot be inferred from announcements alone.\nComfort, fashion, battery life, and the application ecosystem may affect adoption. An earlier attributed quotation about these factors has been removed because its original publication could not be verified.\nTechnical and commercial constraints In the current industry context, the VR/XR industry is plagued by four major challenges:\nFirst, image clarity and display cost. Visible pixel structure depends on more than a single screen-PPI threshold, and Micro-OLED is one of several display approaches. High-resolution optics and displays can add cost, but without a bill of materials it is not possible to attribute a specific portion of a headset\u0026rsquo;s price to one supplier.\nSecond, vergence-accommodation conflict (VAC). In natural viewing, the eyes converge on an object and focus at its distance; a conventional stereoscopic headset can present a different simulated depth while the display\u0026rsquo;s optical focal plane remains fixed.\nThis mismatch can contribute to visual discomfort, although motion, latency, individual sensitivity and other factors also matter. There is no universal severe-nausea rate for all users after a fixed wearing time; device and study conditions matter.\nThird, the current standalone VR headset is a heavily integrated electronic product. Chips, cooling fans, complex Pancake lens sets, and even batteries are all stacked around the user\u0026rsquo;s eye sockets.\nWeight distribution, thermal design and fit can limit comfortable use. Rendering, heat, battery capacity and form factor require tradeoffs, but wear time and battery life vary by headset and workload. There is no single physical \u0026ldquo;impossible trinity\u0026rdquo; or universal one-hour comfort limit.\nFourth, upstream supply chain and scarce materials.\nSupply chains for chips, displays and optics can be exposed to bottlenecks, but the supplied company reports do not establish a direct causal chain from particular minerals to XR display costs. Meta\u0026rsquo;s $19.193 billion 2025 Reality Labs operating loss is reported in its unaudited results release; it alone cannot prove why individual suppliers or competitors changed product plans in 2026.\nWhere the VR/XR industry will ultimately go is a question that only time can answer.\nThat\u0026rsquo;s the simplified map for this video. It is a starting point for checking historical sources, company filings and device-specific data—not a forecast of a winner. I\u0026rsquo;m Dex, see you next time.\nSource notes and primary materials Meta: fourth-quarter and full-year 2025 results — source for Reality Labs\u0026rsquo; 2025 segment operating loss; the release\u0026rsquo;s segment table is unaudited. It is not a headset unit count or company-wide net loss. Ivan Sutherland, The Ultimate Display (1965) — a University of Utah-hosted reproduction of the historical paper, which cites Proceedings of IFIP Congress, pp. 506–508 (1965). It supports an interactive-display vision, not a 2026 product or market forecast, and is not an original publisher-hosted file. Goertek investor relations: 2024 annual report — company-wide filing supplied for manufacturing context; no brand-specific headset shipment claim is inferred. Luxshare Precision investor relations: 2025 annual report and Q1 2026 report — the current official financial-report directory lists the annual report on April 15, 2026, and the Q1 report on April 29, 2026. These are company-wide filings supplied for manufacturing context; the Q1 financial statements are unaudited, and neither report independently measures the entire VR market. The supplied China Mobile Research Institute VR/AR Product Development Status and Trend report dates from November 2022. An original publisher-hosted URL was not verified, so it is recorded here bibliographically rather than linked to an unlicensed copy or used to justify a 2026 market-share claim. The Goertek and Luxshare reports likewise cannot substitute for a defined current shipment survey.\nSupplied documents and reproductions The following links open company-hosted filings, filings on the company\u0026rsquo;s disclosure platform, or the university-hosted reproduction identified below. They are not copied to this website while redistribution rights remain unverified:\nSutherland, The Ultimate Display (1965): University of Utah-hosted reproduction (PDF). Goertek, 2024 Annual Report: publisher-hosted PDF. Luxshare Precision, 2025 Annual Report (English): PDF on the company\u0026rsquo;s disclosure platform. Luxshare Precision, 2026 Q1 Report (Chinese): PDF on the company\u0026rsquo;s disclosure platform. The supplied 2022 China Mobile Research Institute VR/AR report is not offered here as a download because its publisher-hosted original and redistribution terms have not been confirmed.\n","date":"2026-08-11T00:00:00Z","image":"/post/vr-industry-report-2026/cover.jpg","permalink":"/post/vr-industry-report-2026/","title":"Meta Reality Labs' $19.2B 2025 Loss: A VR/XR Industry Map"},{"content":" ","date":"2026-08-10T00:00:00Z","permalink":"/archives/","title":"Archives"},{"content":"Browse the sources, original documents and data files cited across DEX Research articles. Search by article, source or citation context. Sources, documents and data, organized by article. Follow a citation back to its original context, or open the material directly.\n","date":"0001-01-01T00:00:00Z","permalink":"/evidence-library/","title":"Evidence Library"}]