Revision note (v1.2): This version makes two kinds of changes. First, it carries forward the v1.1 correction and rebuild: the original version cited water reclamation figures (10,000 cubic meters per day scaling to 36,000 cubic meters per day by 2026) that describe TSMC’s Taiwan facilities, not its Arizona operations, and Section 4 and the conclusion were substantially rebuilt around a claim-by-claim audit. Every public water-recycling claim made by an operator in this buildout — TSMC, Samsung’s water partner EPCOR, Intel, the city of Sherman, GlobalWafers, Texas Instruments, and the water-engineering firm Gradiant — was checked individually against primary sourcing. One of these, TSMC’s, held up as an honest, precedented, voluntarily adopted standard against which the others can fairly be measured; the rest describe something narrower, softer, less precedented, or less verifiable than their original framing suggested. Section 4 is rebuilt around that finding, and the paper’s conclusion is revised to match.

Second, this version logs five corrections that prior revision notes did not. (1) Section 2 previously identified MP Materials’ magnet manufacturing facility as “Northlake, Texas”; the operating facility is Independence, in Fort Worth, with the 10X expansion facility sited at Northlake. (2) Section 2’s account of MP Materials’ China relationship is made precise: sales to China ceased in July 2025 as a binding condition of the DoD agreement, and the Shenghe Resources offtake was not renewed at its January 2026 expiration, rather than “terminated by 2026.” (3) Section 3’s completion window for Intel’s Fab 62 is corrected from 2026–2027 to approximately 2028, per current reporting. (4) Section 1’s data-center opposition figures are updated from the cumulative $64 billion / 16 projects reported through 2024 to the $130 billion / 75-plus projects blocked in the first quarter of 2026 alone. (5) Section 4 restores the quantitative demand figures an interim draft had removed — TSMC Fab 1’s daily draw, the current recycling rate, the reclamation plant’s staged targets, and the campus-wide contracted-versus-required water figures — and corrects the fab-versus-data-center comparison to state accurately where the two ranges overlap. Substantive conclusions are unchanged.

EXECUTIVE SUMMARY

The Fortress America framework established in Papers I and II predicted that hemispheric consolidation would produce a domestic infrastructure buildout traceable in real time through capital flows, regulatory actions, and physical deployment. Paper III maps where that buildout is physically landing — a California-Texas-Arizona southern spine emerging as the industrial core of the American economy for the next generation. The rare earth supply chain from mine to magnet to motor is now domestically traceable for the first time. More than a dozen semiconductor fabs are at some stage of production or active construction across this corridor. The binding constraint on the entire buildout is water in an arid region, and public claims about the engineering response to that constraint vary widely in how much scrutiny they can withstand — from one operator’s precedented, honestly-stated target to several others whose claims describe less than their framing implies.

SECTION 1 THE SOUTHERN SPINE

Arizona is the most concentrated single-state semiconductor investment in American history. Since 2020, the Arizona Commerce Authority has documented more than 60 semiconductor expansions representing more than $205 billion in investment — a figure Governor Katie Hobbs cited directly at SEMICON West 2025, the semiconductor industry’s own annual trade conference, held in Phoenix for the first time.[1] By early 2025, Industrial Info Resources tracked nearly $50 billion in active industrial construction across the state, with semiconductor fabs, data centers, power plants, and mining operations leading the pipeline.[2]

Texas is building a different but complementary concentration — multiple companies across a contiguous corridor rather than one dominant anchor. Samsung’s Taylor fab on 1,200 acres is the largest foreign direct investment in Texas on record, targeting operational status at the end of 2026 with 1,500 permanent employees and production of 2-nanometer leading-edge chips.[3] In July 2025, Samsung filed a formal regulatory disclosure on the Seoul Stock Exchange confirming a $16.5 billion semiconductor supply agreement running through December 31, 2033; Tesla CEO Elon Musk confirmed the same day that Samsung’s Taylor facility would produce Tesla’s next-generation AI6 chip, used across Tesla’s AI products including autonomous vehicles and the Optimus humanoid robot.[4] The robotics manufacturer and the semiconductor fab are already contractually linked in the same state through the end of the decade.

Both states are absorbing pressure from different directions, and the pattern reveals which part of the buildout is durable. Semiconductor fabs faced real delays in 2023-2024 — Taiwan Semiconductor Manufacturing Company’s (TSMC) first Arizona fab slipped roughly a year due to skilled-labor shortages, and its second fab moved from a 2026 target to 2027-2028 — but both are now active, with TSMC reporting pilot production ahead of its most recent revised schedule as of April 2026.[5] Data centers in the same state have faced a different kind of setback entirely: data center opposition nationwide blocked or delayed at least 75 projects worth roughly $130 billion in the first quarter of 2026 alone — a single-quarter total matching all of 2025 combined, and more than double the $64 billion blocked cumulatively between 2023 and early 2025. Arizona has been part of this wave throughout, where Chandler’s city council unanimously rejected a proposed AI data center in December 2025 and a $14 billion West Valley project was withdrawn after the city declined to approve rezoning.[6] The fabs slipped on labor and equipment timelines and recovered. The data centers are being stopped by the communities they are built in.

The southern spine is the operational core of the domestic semiconductor buildout today, with Arizona in high-volume production and Texas ramping to production in 2026-2027. New York and Ohio represent a second wave of the same institutional pattern — committed capital, steel in the ground, state and federal funding locked in — coming online in the 2030-2032 timeframe.[7] The fab-by-fab production status confirming this geography as the industrial core of Fortress America is detailed in Section 3.

SECTION 2 MINE TO MAGNET TO MOTOR: THE DOMESTIC RARE EARTH SUPPLY CHAIN

Semiconductor capacity is the precondition for advanced manufacturing; without the chip, a robot has no intelligence to actuate in the first place. But intelligence without actuation is just as inert — a robot that can think but cannot move is a research demonstration, not a defense asset. The rare earth supply chain is the second non-negotiable bottleneck, and for the first time, it is now traceable end to end on American soil.

Rare earth elements are not a single-application input. They are foundational across the entire advanced manufacturing stack: permanent magnets containing neodymium, praseodymium, dysprosium, terbium, and samarium power the precision actuators inside semiconductor fabrication equipment itself, optical crystals doped with yttrium and lanthanum enable the lasers used in advanced lithography, and the same magnet chemistry that moves a wafer with micron-level precision inside a fab is what moves a robotic joint with the torque and control a humanoid platform requires.[8] The supply chain risk is not confined to one product category. China accounts for over 90% of global rare earth production broadly, and approximately 85 to 90% of neodymium-iron-boron (NdFeB) magnet production specifically — concentration that touches the fabs documented in Section 1 as much as the robots documented in Paper II.[9]

The chain that resolves this dependency, for the first time, exists entirely within U.S. borders. Mountain Pass, California, operated by MP Materials, is the only operational rare earth mining and processing facility of scale in the United States — the extraction point. Independence, MP Materials’ magnet manufacturing facility in Fort Worth, Texas, is already producing at an initial 1,000 metric ton annual capacity and expanding — the conversion point, where processed rare earth oxide becomes finished magnet.[10] A robot built in Texas with American rare earth magnets processed from California ore is a fundamentally different strategic asset than one assembled with components Beijing can restrict on short notice.

The federal government’s commitment to this chain is structured with more depth than a typical subsidy. In July 2025, the U.S. Department of Defense (DoD) entered a multibillion-dollar agreement with MP Materials that made DoD the company’s largest shareholder, acquiring approximately 15% of outstanding shares through a $400 million purchase of newly created preferred stock, convertible at $30.03 per share.[11] The package extends well beyond equity: a $150 million loan to expand heavy rare earth separation capacity at Mountain Pass; a 10-year price floor of $110 per kilogram for MP’s neodymium-praseodymium (NdPr) products, insulating the company from Chinese price manipulation; and a 10-year offtake agreement under which DoD guarantees that 100% of magnet output from a new “10X” production facility — targeting 10,000 metric tons of annual capacity by 2028 — will be purchased by defense and commercial customers.[12] MP secured an additional $1 billion in commercial financing from JPMorgan Chase and Goldman Sachs on the strength of the federal commitment. As a binding condition of the agreement, MP Materials ceased all sales of rare earth products to China in July 2025 — a cessation already reflected in the company’s own reported revenue — and committed not to renew its existing offtake agreement with China’s Shenghe Resources at its January 2026 expiration, ending the one remaining contractual link between America’s primary rare earth producer and a Chinese state-affiliated buyer.[13]

This is not an isolated transaction. It is the first and most fully developed example of a repeating federal instrument: the government taking direct equity positions in companies that control the extraction and early-stage processing of strategic minerals, rather than in the factories that turn those minerals into finished goods. The same Department of Defense office that structured the MP Materials deal has since taken a 10% stake, with warrants for an additional 7.5%, in Trilogy Metals to advance copper and cobalt development in Alaska, and a separate equity position in Lithium Americas.[14] What this pattern signals about how the government is choosing to secure the upstream end of the supply chain, rather than only the downstream manufacturers, will be explored in full in the final paper of this series.

The one legislative risk to this chain is the Section 45X Advanced Manufacturing Production Credit, which subsidizes domestic production of critical minerals and components.[15] For most 45X-eligible products, the credit phases down starting in 2030 and expires entirely in 2033. Critical minerals, however, sit on a separate track within the same credit: the phaseout for critical minerals does not begin until 2031, with full expiration in 2033 — and unlike solar and wind components, critical mineral production was carved out from the general sunset schedule precisely because of its strategic designation.[16] Even within that more favorable schedule, non-renewal in 2033 would still represent a real long-term risk to the economics of expanding domestic magnet production at the pace the DoD offtake agreement assumes.

But the DoD offtake agreement itself is what makes that risk politically difficult to realize. The Pentagon is now MP Materials’ largest shareholder, has committed to a 10-year price floor, and has guaranteed purchase of 100% of a facility’s output through approximately 2035 — two years past the 45X expiration date. A Congress that allowed the critical minerals credit to lapse without renewal would be undermining an investment the Department of Defense itself structured, sized, and continues to hold equity in. Attacking the credit’s renewal becomes, in practice, attacking the defense department’s own balance sheet and its own stated rare earth independence strategy — a substantially higher political bar than allowing a generic manufacturing subsidy to expire on schedule.

The mine-to-magnet-to-motor chain described here is not a projection. Mountain Pass is producing today. Independence is producing today. The DoD’s equity, loan, and offtake commitments are signed, public, and filed with the U.S. Securities and Exchange Commission (SEC).[17] What remains is scale — the 10X facility in Northlake reaching its 10,000-metric-ton target by 2028, and the broader expansion of domestic magnet capacity beyond MP Materials alone, an effort already extending into Wyoming and Alaska on separate tracks. The chain that supplies the humanoid robotics manufacturing imperative identified in Paper II now exists, end to end, without crossing Chinese territory at any point.

SECTION 3 THE SEMICONDUCTOR SPINE

The strategic vulnerability this buildout addresses is stark and well documented. The United States declined from producing roughly 37 to 40% of global semiconductors in 1990 to approximately 10 to 12% today, and as of the passage of the CHIPS and Science Act in 2022, none of the world’s most advanced chips were manufactured domestically.[18] The correction underway since is not a slow policy response — it is one of the fastest industrial buildouts in American history, and its pace is best understood by naming what is actually under construction, what is already producing, and what remains delayed.

TSMC’s Arizona campus is the clearest evidence of acceleration. Fab 21’s first facility entered high-volume 4-nanometer production in Q4 2024 and is now manufacturing chips for Apple and Nvidia — the first time TSMC has produced cutting-edge AI silicon outside Taiwan.[19] The second fab’s construction was completed in 2025, with equipment installation beginning in the third quarter of 2026 and 3-nanometer production targeted for 2027 — a full year ahead of the original schedule, according to TSMC’s own CEO.[20] A third fab broke ground in April 2025, targeting 2-nanometer and A16 process technology by the end of the decade.[21] Total committed investment has grown from an initial $12 billion in 2020 to $165 billion across six planned fabs, with TSMC describing long-term framework expansion reaching approximately $465 billion — the largest single foreign direct investment in American history.[22]

Texas Instruments offers the clearest example of speed. Construction began on its Sherman, Texas site in mid-2022; the first fab, SM1, began production on December 17, 2025 — three and a half years from groundbreaking to output.[23] The facility produces foundational analog and embedded processing chips, the components used in nearly every electronic device, including industrial robotics and automotive systems.[24] SM2’s exterior shell is already complete, with cleanroom installation underway in 2026. The Sherman campus is part of a $60 billion investment across seven planned fabs in Texas and Utah.[25]

Samsung’s Taylor, Texas fab presents a more complicated picture, and an honest accounting requires saying so. Originally targeted for 2024 production, the facility was delayed twice — once to upgrade its process technology from 4-nanometer to the more advanced 2-nanometer node, and again reportedly due to insufficient near-term customer demand.[26] As of May 2026, Samsung’s own foundry leadership confirmed that customer production, including for Tesla, is scheduled to begin in 2027, with the facility’s third-generation 2-nanometer process now in installation.[27] The delay is a demand and technology-upgrade story, not a capital withdrawal — Samsung’s total investment in the site grew from $17 billion to $44 billion over the same period, and the company received $4.75 billion in direct CHIPS Act funding to proceed.[28]

Intel’s Chandler, Arizona campus adds two more fabs to the spine. Fab 52 is in high-volume production on Intel’s 18A process — the first U.S. facility to cross the 2-nanometer threshold, with Intel’s own chief technology officer describing it as capable of more than 10,000 18A wafer starts per week — while Fab 62 is under construction and expected to be ready around 2028.[29] Intel’s commitment to Arizona has held even as the company cancelled comparable projects in Germany and Poland, consolidating its advanced manufacturing investment domestically rather than abroad.[30]

Taken together, the southern spine’s fab buildout shows a pattern consistent with the rest of this paper’s findings: delays have been common, but they have been delays of one to two years driven by labor availability, technology upgrades, and customer demand — not cancellations, and not capital withdrawal. Every fab named above is either producing today or under active construction with a committed completion date. That distinguishes this buildout sharply from the data center cancellation wave described in Section 1, where the setbacks have come from local political rejection rather than engineering or market timing.

SECTION 4 THE WATER CONSTRAINT AND THE ENGINEERING RESPONSE

The binding constraint on the southern spine is water, in a region already under acute and well-documented stress. Arizona’s water supply depends heavily on the Colorado River, which has experienced sustained shortage conditions for over two decades, and on groundwater aquifers facing long-term depletion as the state’s population and industrial base both grow.[31] Texas faces a parallel and independently documented problem: the state’s own water planning authorities project a long-term supply deficit of nearly 6.9 million acre-feet, with officials warning that without expanded infrastructure, demand could outstrip supply during the next prolonged drought.[32] Into both constrained systems, the fab buildout documented in Section 3 introduces a new and significant industrial water demand layered on top of an existing shortage.

The scale of that demand is best understood by direct comparison. Estimates of a single fab’s daily water draw vary by facility size and process node — independent technical sources put the range anywhere from roughly 3 million to 10 million gallons per day, with the largest individual facilities and multi-fab campuses running higher still.[33] The World Economic Forum’s commonly cited industry figure, 10 million gallons per day, sits at the top of that range rather than describing a typical fab. A single large data center, by contrast, consumes roughly 1.5 to 5 million gallons per day.[34] Across most of both ranges, the fab is the larger consumer — by a factor of two to four at the figures most often cited in public discussion — though the two ranges overlap at their edges: the smallest fabs draw less than the largest data centers.

The demand this section examines is directly quantified at the buildout’s most advanced site. TSMC’s first Arizona fab draws 4.75 million gallons of water per day, recycled at a 65% rate through the campus’s existing in-house water resource center, which converts industrial wastewater for use in the systems supporting operations.[35] The company’s dedicated Industrial Reclamation Water Plant (IRWP), which broke ground in August 2025, is designed to open at an initial recycling rate of 85% and reach a 90% goal, cutting Fab 1’s net daily draw to under 1.2 million gallons once operational in 2028.[35] At full buildout, TSMC’s three planned Arizona fabs will require 17.2 million gallons per day against a development agreement with the City of Phoenix for 11.4 million gallons per day.[37] And the industry-wide backdrop that makes recycling architecture decisive: an average of 76% of the water a fab withdraws is used directly in the manufacturing process, while most current recycling captures water for secondary systems — cooling towers, scrubbers — rather than returning it to the ultrapure loop that touches the wafer.[45]

One operator’s public commitments stand apart from the rest of this section, and it is worth establishing why before turning to the others. TSMC states plainly, in its own public materials, that its Arizona fabs “aim to achieve” a 90% water recycling rate, pursued through the purpose-built IRWP with a stated “design goal” of near-zero liquid discharge.[35] The IRWP broke ground in August 2025 and will not be operational until 2028. TSMC’s own language has always reflected that timeline honestly — an aim, a goal, a plant under construction. Their aims and designs have been clearly stated and appear to model the existing capabilities already demonstrated in their Taiwanese operations, where the company maintains recycling rates exceeding 85%.[36] That standard is also the harder and more expensive of the two paths available to a fab operator: recycling the core ultrapure water that touches the wafer, rather than the comparatively simpler work of recycling water used in cooling towers and air scrubbers. Nothing required TSMC to choose the harder path. The City of Phoenix’s own public position is that it has sufficient water capacity for TSMC regardless of whether the 90% target is ever reached, and the figure does not appear among the enforceable milestones attached to TSMC’s $6.6 billion CHIPS Act award.[37] TSMC set this target voluntarily, stated it in public, and did so in a community where water has been, by local reporting, “the flash point in many neighborhood meetings and planning sessions” — in front of the audience most likely to notice if the company fell short.[38] Whether TSMC’s Arizona plant reaches 90% recycling on the 2028 timeline the company has set for it is a separate, later question, and this paper takes no position on it in advance. What can be said now is that TSMC’s public description of its own progress has not, at any point, claimed more than the company has actually built.

Measured against that standard — public claims that describe no more than what has actually been achieved or actually been committed to, and that point to demonstrated precedent where precedent is claimed — the other claims made across this buildout hold up less consistently.

Samsung’s water partner in Taylor, EPCOR, announced a target in a July 2023 press release: reclaiming or reusing 75% of the Blue Sky Water Reclamation Facility’s process water.[39] Unlike TSMC’s figure, this one has not been paired with an operational timeline, an engineering description, or any subsequent public update. A local outlet that sought detail on how the target would be met, roughly six weeks after the announcement, received no response beyond the same press-release language, and none has surfaced publicly in the nearly three years since. The claim may be accurate. It has not been made possible to check.

Intel describes its Ronler Acres, Oregon operations as having reached “net positive” water status as of 2022 — returning more treated water to the local watershed than the company withdraws.[40] This claim is true, and it is a genuine achievement. It is also, by industry reporting on the underlying mechanism, an “end-of-pipe” result: water is treated and returned to secondary systems and to the watershed, rather than recycled back into the core ultrapure loop that touches the wafer — the same category of water TSMC’s existing Arizona system currently handles at a 65% rate, achieved more completely. “Net positive” is accurate. It answers a different question than the one this section is asking.

In Sherman, Texas, three separate things get discussed together and are worth pulling apart. The city’s Post Oak Wastewater Treatment Plant is real, built, and operating — but its 16 million gallon per day permitted capacity is citywide infrastructure, serving “domestic, commercial and industrial wastewaters” for the whole community, not a figure specific to either fab.[41] Its stated purpose is to treat wastewater to a standard “suitable and safe for reintroduction back into the natural stream environment” — it returns water to a creek, not to TI’s or GlobalWafers’ own production lines. Separately, GlobalWafers carries a water commitment NIST itself lists as a condition of its CHIPS Act award: recycling “at least 50% of the process water used onsite” within a year of completing its project, industry convention suggesting “process water” refers to the core production loop rather than secondary systems, though the government’s own award language does not spell out that distinction explicitly.[42] That commitment is qualified by “commercially reasonable efforts,” a standard that does not require the 50% figure actually be reached. TI has separately stated, in its own press materials rather than in NIST’s listed award conditions, that it is “endeavoring to achieve a 70% water reuse capability” across Sherman and its Lehi, Utah site — a company statement with no equivalent government-tracked commitment behind it.[43] TI’s own most recent disclosed company-wide water reuse figures — 27% in 2020, 29% in 2023 — show little movement over that period and sit far below the 70% target, with no demonstrated site achieving anything close to that rate publicly disclosed as of this writing.[44] No public reporting yet confirms whether either company’s Sherman-specific water target has been met.

Independent water-engineering firms serving the semiconductor sector, including Gradiant, advertise recycling capabilities of up to 99% for individual clients, at investments running into the hundreds of millions of dollars.[45] The company has not disclosed the client or site behind its highest-profile version of this claim, describing it only as “one of the world’s largest semiconductor manufacturers.” No public source connects this specific claim to any facility in the southern spine.

None of this means the companies making these claims are being dishonest. It means that, measured against the plainest and most verifiable standard available in this section — a company that stated a hard goal, described it honestly, pointed to demonstrated precedent, and made that statement in public to the community most likely to hold it accountable — most of the other claims in this buildout describe something narrower, softer, less precedented, or less verifiable than their framing suggests: a true result answering a different question, a target that has gone unconfirmed for years, a commitment qualified by its own contract language and unsupported by any comparable track record, or a capability whose relevance to this region has not been established. The water constraint documented at the start of this section has not been resolved by any of it. TSMC’s Arizona fabs have been drawing water and producing chips since late 2024. The plant designed to substantially close that gap will not exist until 2028 — and of everyone whose public claims this section has examined, TSMC is the only one who has said so plainly, in advance, without being asked.

CONCLUSION

Across this paper and the series it belongs to, the forces shaping the buildout of Fortress America are moving with a speed that is itself a finding. Section 1 documented more than $205 billion in announced semiconductor investment concentrated in Arizona alone, with Texas building a complementary corridor anchored by a $16.5 billion contractual link between Samsung’s Taylor fab and Tesla’s robotics and vehicle silicon. Section 2 traced the rare earth supply chain from Mountain Pass to Independence to a $400 million Department of Defense equity stake — the federal government’s largest and most structurally complete bet on any single company in this series, complete with a 10-year price floor and an offtake agreement guaranteeing the purchase of an entire facility’s output. Section 3 showed the semiconductor spine itself: fabs producing today in Arizona, fabs ramping in Texas, delays measured in months and quarters rather than cancellations. Section 4 turned to the constraint that ties all of it together — water — and found something more specific than a shortfall: a set of public claims that, examined individually, describe the state of that constraint less completely than their framing suggests.

What several billion-dollar projects already producing, and several more under active construction, demonstrate is that these are not speculative bets being made by capital hoping a policy environment holds steady. They are commitments structured to survive a change in administration, because the surrounding institutions — state legislatures funding water infrastructure before the fabs are finished, the Department of Defense taking equity rather than simply writing a grant, municipal governments rebuilding their utilities ahead of demand — are treating the underlying strategic problem as a fixed feature of the next decade, not a four-year policy preference.

The clearest evidence of that institutional permanence is the rare earth chain itself, and the timeline is worth stating plainly because it spans three administrations without reversing direction once. The Department of Defense’s first rare earth investment in MP Materials came in 2020, under the first Trump administration, a $10 million Defense Production Act award to a company struggling against Chinese retaliatory tariffs.[46] The Biden administration not only continued the relationship but expanded it — $45 million for Mountain Pass processing, more than $288 million to a second rare earth company, and a formal five-year mine-to-magnet investment strategy announced in 2024.[47] The second Trump administration then escalated that foundation into the $400 million equity stake, the price floor, and the offtake guarantee detailed in Section 2 — by far the largest single commitment in the chain’s six-year history. Three administrations, two parties, one direction, each handoff larger than the last.

This is the new mechanism this series has identified — equity rather than grants, ownership rather than subsidy — and it will be explored in full in the final paper of this series, which examines how the federal government is securing the raw material end of the supply chain across rare earths, copper, cobalt, and lithium, and what that ownership costs in dollars and control.

The water constraint is the one piece of this picture without a finish line, and it deserves a more precise closing than “underway but incomplete.” One company in this section set the standard by which the rest can fairly be judged. TSMC’s Arizona target — 90% recycling of the water that actually touches the wafer, not just the water that cools the building around it — was pointed explicitly at a demonstrated capability the company already runs in Taiwan, and was stated in public, in a community watching closely, with no regulator or contract requiring the company to say anything at all. Measured against that same bar, the rest of the buildout falls short in different and specific ways. Samsung’s reclamation partner announced a target three years ago and has answered no question about it since. Intel’s celebrated “net positive” result is real, and it is not the result this section is asking about. Sherman’s wastewater plant is a genuine, completed exception — but it treats water for the whole city, not for the two fabs it gets credited with serving, and it returns that water to a creek, not to either company’s production line. The one Sherman commitment actually aimed at a fab’s own process water, GlobalWafers’ federally tracked 50% target, is real and government-listed, and is also softened by language that does not require the number be reached. Texas Instruments’ matching 70% figure carries no such federal tracking at all, and the company’s own most recent disclosed results — 27% company-wide in 2020, 29% in 2023 — sit nowhere near it, with no comparable site TI has pointed to as precedent. And the water-engineering industry’s own flagship claim of 99% recycling, cited widely as evidence the private sector has this handled, cannot be confirmed to describe any facility in the region this paper is about.

None of this means the companies making these claims are lying. It means that TSMC’s Arizona fabs have been drawing water and producing chips since late 2024, that the plant designed to substantially close that gap will not exist until 2028, and that of everyone whose public statements this paper has examined on the hardest and most expensive part of the water problem, exactly one company has been consistently precise about what it has and has not yet done. Closing the mine-to-magnet-to-motor chain in Section 2 is a solved problem as of this writing. Closing the water gap in Section 4 is not — and the public record on how close anyone actually is to closing it is considerably thinner than it first appears.

REFERENCES

[1] Arizona Commerce Authority; Office of Governor Katie Hobbs, remarks at SEMICON West 2025, Phoenix, October 2025.

[2] Industrial Info Resources, Arizona Industrial Construction Pipeline, 2025.

[3] Office of Governor Greg Abbott, Texas Semiconductor Innovation Fund Grant Announcement, September 17, 2025.

[4] Samsung Electronics, Seoul Stock Exchange Regulatory Filing, July 28, 2025; Bloomberg, Reuters, CNN, July 28, 2025.

[5] Tom’s Hardware, “TSMC accelerates production timeline for new Arizona factory,” December 2025; Data Center Dynamics, “TSMC saysArizona fab is now ahead of schedule,” 2026.

[6] Data Center Watch (10a Labs), Q1 2026 Report; NBC News, “Study shows state and local opposition to new data centers is gaining steam,”2026; Tom’s Hardware, “More than 75 data center build-outs worth $130 billion have been successfully blocked in the first three months of 2026,” 2026; Fox Business, “Chandler, Arizona, city council unanimously votes against AI data center,” December 2025.

[7] UltraFacility, “Semiconductor in numbers: Global fab construction timelines,” April 2026; Intel Newsroom, Ohio project status statements,2025-2026.

[8] Rare Earth Exchanges, “How Rare Earth Elements Enable Modern Semiconductor Manufacturing Equipment,” January 2026.

[9] World Population Review / USGS rare earth production data, 2025; Optimusk, “Tesla Optimus Supply Chain,” 2026; 36kr English, April 2026.

[10] MP Materials Corp., Form 8-K, January 22, 2025, U.S. Securities and Exchange Commission; Fort Worth Report, “Fort Worth manufacturerbegins producing rare earth magnets,” January 2025.

[11] MP Materials Corp., Form 8-K, July 2025, U.S. Securities and Exchange Commission; CNBC, “Pentagon to become largest shareholder inrare earth miner MP Materials,” July 2025.

[12] MP Materials Corp., Form 8-K, July 2025, U.S. Securities and Exchange Commission; The Defense Post, “Pentagon Takes Stake in USRare Earth Company,” July 2025.

[13] MP Materials Corp., Form 8-K, July 2025, U.S. Securities and Exchange Commission; Select Committee on the Chinese Communist Party,U.S. House of Representatives, witness testimony, November 2025; C&EN, “US invests in rare earth firm MP Materials,” July 2025.

[14] Investing News Network, “Trilogy Metals Shares Rocket as US Government Takes Stake in Alaska Project,” October 2025; Axios, “US totake 10% stake in Trilogy Metals,” October 2025; Mayer Brown, “US Government Equity and Equity-Linked Investments in Critical Minerals,” April 2026.

[15] Internal Revenue Code Section 45X, Advanced Manufacturing Production Credit, as amended.

[16] U.S. Department of the Treasury, guidance on Section 45X critical minerals phasedown schedule, 2025-2026.

[17] MP Materials Corp., Form 8-K filings, U.S. Securities and Exchange Commission, 2025.

[18] Council on Foreign Relations, “The CHIPS Act: How U.S. Microchip Factories Could Reshape the Economy,” 2024; Semiconductor IndustryAssociation, Chip Incentives & Investments data.

[19] TSMC Arizona, company facility status page, 2026; Tech Insider, “TSMC’s $165B Arizona GigaFab: Reshaping US Chips,” 2026.

[20] Tom’s Hardware, “TSMC brings its most advanced chipmaking node to the US yet,” December 2025.

[21] TSMC Arizona, company facility status page, 2026.

[22] BlackRidge Research, “TSMC Arizona Fab: USD 165 Billion Semiconductor Project,” 2026.

[23] Texas Instruments, “Texas Instruments begins production at its newest 300mm semiconductor manufacturing facility in Sherman, Texas,”December 17, 2025.

[24] EE Times, “Inside Texas Instruments’ New 300mm Fab in Sherman, Texas,” December 2025.

[25] TI.com, “Sherman, Texas: 300mm wafer fabs,” company site, 2026.

[26] Tom’s Hardware, “Samsung delays $44 billion Texas chip fab,” July 2025; Electronics360, “Report: Full production of Samsung’s Texas fabpossibly delayed to 2027,” March 2026.

[27] TechTimes, “Samsung Taylor Fab Production Confirmed for 2027,” May 2026.

[28] MLQ.ai, “Samsung Delays Completion of $44 Billion Texas Chip Plant,” July 2025.

[29] CNBC, “Intel aims to find clients and catch TSMC with new chip fab in Arizona,” December 19, 2025; Tom’s Hardware, “Intel’s fab roadmapexamined — Arizona, Ohio, Ireland, and the two deadlines deciding 14A process node,” June 2026.

[30] Ibid.

[31] U.S. Bureau of Reclamation, Colorado River shortage condition declarations, 2022-2026.

[32] Texas 2036, “Foundation for Economic Growth: Assessing Texas’ Water Infrastructure Needs,” 2024.

[33] World Economic Forum, “Semiconductor manufacturing and big tech’s water challenge,” 2024; CWR, “8 Things You Should Know About

Water & Semiconductors”; IDE Tech, “Water Sustainability in the Semiconductor Industry”; Semiconductor Engineering, “How Semiconductor Fabs Use Water,” August 2025; SAMCO Technologies, “Industry Focus: Semiconductor industry trends and the importance of water resource management,” April 2025.

[34] EESI, “Data Centers and Water Consumption”; MOST Policy Initiative, “Data Center Water Use,” April 2026.

[35] TSMC, “TSMC Arizona and U.S. Department of Commerce Announce up to US$6.6 Billion in Proposed CHIPS Act Direct Funding,”pr.tsmc.com, April 2024; TSMC Arizona, “Sustainability” statement, tsmc.com, 2026; Data Center Dynamics, “TSMC breaks ground on water reclamation project in Phoenix, Arizona,” September 2025; Arizona Technology Council, “TSMC Breaks Ground on ‘Near-Zero’ Discharge Water Plant to Back Fabs,” September 2025; AZBEX, “TSMC Building Water Reclamation Plant,” September 2025; National Institute of Standards and Technology, Draft Environmental Assessment for TSMC Arizona, NIST-CPO/EA-002, May 2024.

[36] TSMC, ESG “Water Management: Green Manufacturing” disclosures, esg.tsmc.com (average production-process water recycling rate of86.7%); Ahwatukee.com / Times Media Group, “TSMC Arizona’s Water Reclamation Initiative Forwards Sustainability in Semiconductor Manufacturing,” September 2025, citing TSMC’s 2024 Sustainability Report.

[37] City of Phoenix, statement of Mayor Kate Gallego, August 2025; City of Phoenix development agreement figures as reported; SemiWikiforum discussion, January 2025; Senator Mark Kelly, “Kelly and Arizona Leaders Celebrate Finalized $6.6 Billion CHIPS and Science Act Award to TSMC,” press release, November 2024; National Institute of Standards and Technology, “TSMC Arizona,” nist.gov/chips.

[38] Hoodline, “North Phoenix Scores Big As TSMC’s Second Chip Plant Hits Finish Line,” May 2026; Fortune, “Water-guzzling chipmakerTSMC and drought-plagued Arizona are an unlikely pair, but Phoenix says it has enough water,” April 2024.

[39] EPCOR USA, “EPCOR Tapped as Water Partner in Central Texas,” July 13, 2023; DNA Systems, “EPCOR - Sandow Water Project,”project completion record, October 2024; East Wilco Insider, “Looking for Enough Water,” September 1, 2023.

[40] ScaleBan Equipments, “How Semiconductor Industry is Tackling Wastewater Challenges”; industry reporting on Intel Ronler Acresend-of-pipe treatment mechanism and Intel’s 2022 net positive water milestone.

[41] City of Sherman, Texas, “Wastewater,” official city website; KXII, “Sherman unveils new multi million dollar wastewater treatment plant,”October 15, 2025.

[42] National Institute of Standards and Technology, “GlobalWafers (Texas),” nist.gov/chips, CHIPS award terms and environmentalcommitments.

[43] Texas Instruments, “Texas Instruments announces award agreement for CHIPS and Science Act funding,” ti.com, December 20, 2024;National Institute of Standards and Technology, “Texas Instruments (Utah),” nist.gov/chips.

[44] Texan By Nature, “Texas Instruments,” txn20.org, 2020 water reuse disclosure; Texas Instruments, CDP Water Security Questionnaire2023, ti.com (25.5% of water recycled in 2022); Texas Instruments, 2024 Corporate Citizenship Report (29% of water reclaimed and reused in other applications, 2023).

[45] The Register, “Mystery German chip fab sips on Gradiant’s ultrapure water,” January 22, 2024; Manufacturing Dive / ESG Dive,“Semiconductor industry faces water, sustainability challenges,” August 2025 (industry-average 76% direct process use of withdrawn water).

[46] Heatmap News, “The Pentagon’s Rare Earths Deal Is Making Former Biden Officials Jealous,” July 2025.

[47] The White House (Biden-Harris Administration Archives), “Fact Sheet: Biden-Harris Administration Takes Further Action to Strengthen andSecure Critical Mineral Supply Chains,” September 2024; Global Policy Watch, “Made in America: The Outlook for Critical Minerals,” October 2025; Vulcan Elements, “Money Finally Flowing to US Rare Earths Can’t Come Fast Enough,” August 2025.

FORTRESS AMERICA SERIES

Paper I: Fortress America — Hemispheric Consolidation, Infrastructure Cascades, and the Regional Transformation of the American Economy

Paper II: If It Can Work, It Can Fight — The Case for Domestic Humanoid Robotics Manufacturing

Paper III: Where Fortress America Lands — The Southern Spine and the Race to Build It

Paper IV: The Government Stake — Equity, Speed, and the Limits of Disclosure

Paper V: The Atom and the Chip

This paper represents independent analytical and systems research and is the third in the Blue Collar Analytics Fortress America series. Nothing contained herein constitutes financial advice, investment recommendations, or legal counsel. Readers should conduct their own due diligence and consult qualified professionals before making investment decisions.

Revision note (v2.0): This version makes two kinds of changes. First, it adds three new sections — The Demand (Section 2), The Fuel Bottleneck (Section 6), and The Military Buildout (Section 5) — built from three weeks of continued research after version 1.0’s publication. None of that material corrects the original paper; it confirms it. The named hyperscaler contracts, the enrichment awards, and the three parallel military reactor programs documented below are the direct evidence version 1.0’s central bet could only infer, surfacing at a rate that is itself proof of the original argument: capital and government moving faster than public visibility can track. Second, this version corrects an error version 1.0 introduced on its own: Section 5 of the original paper restated Paper III’s water findings using figures from Paper III’s pre-v1.2 text — the single 10-million-gallon fab figure rather than the audited 3-to-10-million range, a TSMC reclamation figure that describes the company’s Taiwan operations rather than Arizona, an Intel “zero-liquid-discharge” characterization that Paper III’s claim-by-claim audit replaced with the accurate and narrower “net positive,” and a description of Sherman’s wastewater plant as fab-specific when it is citywide infrastructure. The water section (now Section 7) is rebuilt to match Paper III v1.2’s audited findings. The Mechanism section is expanded from two federal postures to three, reflecting the contracting model the military programs introduce. Substantive conclusions are unchanged; the corrections align this paper with the standard the rest of the series has already been held to.

Second, on TerraPower’s Natrium plant at Kemmerer, this version adds material rather than correcting it: version 1.0 noted only that Natrium uses conventional steam-cycle generation, and said nothing about its water siting. Section 7 now documents that the plant is built beside the Naughton generating station — frequently described in public materials as a “retiring coal plant,” a characterization accurate as to coal but incomplete, since Naughton is converting to natural gas and continuing to operate. The point is not a correction to the earlier version but an illustration of a pattern this paper flags throughout: a claim can be literally true and still leave a materially incomplete impression, and such claims are treated here as soft claims and stated in full.

EXECUTIVE SUMMARY

The federal government and private capital are betting, at scale and at speed, that the AI and semiconductor buildout this series has already documented will need far more electricity than the grid currently supplies — and nuclear power is the answer they have chosen to fund. Version 1.0 of this paper made that claim by inference: hundreds of billions of dollars moving toward generation capacity sized to match an industrial demand curve, with no single reactor traceable to any single fab. The inference is no longer necessary. The demand side of the bet now has names, dollar figures, and contract terms.

Because this paper tracks a buildout that produces new announcements almost daily, it rates every commitment by the enforceability of the instrument behind it, not the size of its headline number — and in this buildout, those two run in opposite directions. From strongest to weakest: a binding contract or executed transaction — a signed power-purchase agreement, a deed — where money and obligation are both real; a conditional commitment, real money obligated but gated on a future decision; a memorandum of understanding, a framework to cooperate with no purchase obligation; a letter of intent, a stated intent to transact that binds no one; and a pipeline figure, a self-reported demand number with no instrument behind it at all. Throughout this paper, every deal is tagged against this ladder. The reason it is necessary shows at the extremes: the largest firmly contracted commitment in the entire buildout is Amazon’s 1.92 gigawatts, a signed agreement — while the largest gigawatt figures anywhere belong to the bottom rung, self-reported pipelines attached to reactors that do not yet exist. The numbers grow as the commitments soften.

The four largest AI infrastructure operators have each signed binding, long-term power-purchase agreements tying named nuclear plants to their own demand, running seventeen to twenty years against restarted, existing, and not-yet-built capacity. One brings a federal utility inside the deal as the contracted buyer; another sits alongside a federal loan to the plant it draws from. These are the top of the ladder: signed, enforceable, decades long. No company signs a twenty-year power contract against a demand curve it believes might flatten.

The fuel that demand requires has its own federal response, bipartisan in origin: a ban on Russian enriched uranium that unlocked billions in domestic enrichment funding, followed by major federal awards to three enrichment companies. And the military is running three distinct reactor programs under a contracting model that makes the government a guaranteed customer rather than an owner — a third federal posture alongside the deregulation and equity-adjacent lending version 1.0 documented.

The original paper’s findings stand unchanged beneath these additions. Four reactors reached criticality under new, deregulated federal testing pathways, all privately financed, meeting and exceeding an executive order’s target of three by July 4, 2026. The Department of Energy issued a $17.5 billion conditional loan commitment to Westinghouse — real and funded, but gated on a final investment decision, and structured through vehicles and utility partners not yet named. It sits high on the ladder as a commitment and remains, as of this writing, invisible in its particulars.

That combination — enormous commitment, minimal disclosure — is the paper’s throughline, and the new material widens the gap rather than closing it. The five sites that will receive the Westinghouse billions are not public. What the hyperscalers are paying, in most cases, is not public. The military’s final vendor and site selections are pending. And the water requirement of the new-build reactors in this paper is, with a single partial exception, unaddressed in any document this paper has reviewed — one project has visibly changed course to confront the constraint; the rest have not. That gap is the sharpest, because the hyperscaler deals overwhelmingly attach to existing or restarting plants whose water footprints were settled decades ago. The water question concentrates almost entirely on the new-build track — the same track where disclosure is thinnest.

None of the material added in this version contradicts the paper it revises. All of it confirms the original thesis at a scale and speed the first publication could not capture. Everything documented here is a floor, not a ceiling — and the rate at which new, dated, independently sourced material surfaced in three weeks is itself evidence for this paper’s core argument: pace outrunning visibility.

SECTION 1 — THE BUILDOUT

The deregulatory and capacity-building groundwork for this buildout is bipartisan, and the clearest evidence of that sits at the beginning of the timeline. The ADVANCE Act passed both chambers of Congress in mid-2024 by margins rarely seen on energy legislation — 88-2 in the Senate, 393-13 in the House — directing the Nuclear Regulatory Commission to cut review fees for advanced reactor applicants and build a faster, technology-neutral licensing framework. President Biden signed it into law on July 9, 2024.[1] Four months later, on November 12, 2024, his administration issued the U.S. Nuclear Energy Deployment Framework, setting a target of roughly 300 gigawatts of nuclear capacity by 2050 — triple the current fleet.[2]

On May 23, 2025, President Trump signed four executive orders that built directly on that foundation. EO 14300, “Ordering the Reform of the Nuclear Regulatory Commission,” set a target of 400 gigawatts by 2050 — four times today’s roughly 100 — alongside the licensing timelines discussed in Section 4.[10] EO 14302, “Reinvigorating the Nuclear Industrial Base,” directed the Department of Energy’s Loan Programs Office — since renamed the Office of Energy Dominance Financing — to prioritize nuclear restarts, uprates, and new construction.[3] EO 14299, “Deploying Advanced Nuclear Reactor Technologies for National Security,” directed the rapid deployment of reactors at DOE sites supporting AI infrastructure and, on the national-security track later implemented through the Army’s Janus program, a reactor operating at a domestic military base by September 30, 2028.[9] EO 14301, “Reforming Nuclear Reactor Testing at the Department of Energy,” created the Reactor Pilot Program and set the target this section traces: at least three advanced reactor designs reaching criticality outside the national laboratories by July 4, 2026.[6]

That target was not merely met but exceeded, and by more programs than one. Four privately developed reactors reached criticality by the July 4 deadline. Antares Nuclear’s Mark-0 was first, achieving zero-power criticality at Idaho National Laboratory on June 4, 2026 — the first privately developed non-light-water reactor to do so in the United States in more than forty years; Mark-0 uses a sodium heat-pipe design.[4] Zero-power criticality — a self-sustaining chain reaction at essentially no measurable output — proves a reactor’s core physics before any attempt to generate power, and it was this milestone, not power generation, that the executive order’s July 4 target required. Valar Atomics’ Ward 250 followed on June 18 at the Utah San Rafael Energy Lab in Emery County, becoming the first reactor built and operated entirely outside a national laboratory; it is helium-cooled, a third approach distinct from Mark-0’s sodium and from the ordinary water used in the AP1000 design discussed later.[5] Deployable Energy’s Unity reached criticality on June 30 — the third to do so, and the first under the Nuclear Energy Launch Pad, the successor pathway the DOE established in March 2026 to extend Reactor Pilot Program authorization to developers beyond the original cohort.[13a] Aalo Atomics’ Critical Test Reactor was the fourth and last, going critical at 12:20 a.m. on July 4 itself.[13b] Three of the four came through the Reactor Pilot Program; Unity came through its successor — four reactors, two programs, against a target of three.

Only one of the four has so far gone beyond cold criticality. In the days after June 18, Valar began power ascension, reaching its rated output of roughly 100 kilowatts thermal, and on July 1 briefly powered an Nvidia AI chip from that output — the first time a U.S. advanced reactor has done so, and an early proof of concept far below any commercial scale. Valar’s own framing captures the distinction this paper relies on throughout: “Cold ≠ Hot: Cold proves the physics. Hot proves the power.”[11] By that standard, the July 4 milestone was a cold one, met four times over; the hot demonstration, so far, belongs to Valar alone.

Ward 250 was also the subject of the first aerial transport of a reactor in U.S. history. In February 2026, U.S. Air Force C-17s carried the disassembled, unfueled unit from March Air Reserve Base, California, to Hill Air Force Base, Utah — two states, not a cross-country journey — in an operation named Windlord, a multi-wing effort drawing on the airlift capability that routinely transports U.S. nuclear weapons.[12] The mission used military aircraft and crews, though the Wall Street Journal reported that Valar paid the cost of the flight. All four criticality reactors were developed with private capital — reported figures include roughly $130 million raised by Valar toward Ward 250 — and required no federal money to reach the milestone, only a faster federal permitting pathway than the Nuclear Regulatory Commission’s standard licensing process.[14]

The federal financing story is a separate and much larger track, running in parallel. On June 23, 2026, the Department of Energy’s Office of Energy Dominance Financing issued a conditional loan commitment of up to $17.5 billion to Westinghouse Electric Company, to finance the fixed-price purchase of long-lead equipment for up to ten AP1000 reactors — a large, roughly 1,100-megawatt pressurized water reactor design, and the only large-scale advanced reactor currently licensed for commercial operation in the United States — across up to five two-reactor project sites.[7] The announced structure routes the loan through a Westinghouse special purpose vehicle expected to administer up to five project funding vehicles, each jointly owned by Westinghouse and a utility partner and each required to commit $500 million in equity — $1 billion per project — before any federal funds are released, with the loan repaid from the sale of the long-lead items. As of this writing, none of the five sites, utility partners, or vehicles has been named; Westinghouse has stated it signed letters of intent with seven candidate utilities, to be narrowed to five.[7] “Conditional” matters here: the commitment depends on conditions not yet satisfied, including a final investment decision that has not been made, and is distinct from money that has actually been disbursed. On the paper’s commitment ladder, this is a conditional commitment — strong, funded, and signed, but gated and, in its particulars, not yet public.

This $17.5 billion loan operationalizes part of a larger $80 billion partnership the Department of Commerce announced in October 2025, under which — after a final investment decision — the federal government is entitled to 20 percent of Westinghouse distributions in excess of $17.5 billion, and may require an initial public offering and receive warrants for an equity stake of up to 20 percent if Westinghouse is valued at $30 billion or more by January 2029.[7a] This is the same equity-adjacent architecture Paper IV traced through MP Materials, Vulcan Elements, and Trilogy Metals, now applied at a larger dollar scale to the company building the reactors. Westinghouse itself is majority owned by Brookfield, a private asset management firm, with Cameco, a uranium mining company, holding the remainder — the same upstream-supply-chain pattern Paper IV documented, in which a critical-materials company’s equity holders benefit directly from federal backing.[22]

Geographically, the buildout is concentrated but not confined to two states, and those states have begun coordinating directly. Wyoming hosts TerraPower’s Natrium reactor near Kemmerer, a sodium-cooled fast reactor that received its NRC construction permit in March 2026 — the first commercial reactor construction permit the agency has issued in nearly a decade, and the first ever for a commercial non-light-water power reactor.[15] Construction on the plant’s non-nuclear components was already under way, and the project is targeted for completion around 2030. Utah hosts the densest concentration of nuclear activity in the country: at least eight companies — TerraPower’s second site, Valar Atomics, Anfield Energy, Nusano, General Matter, Atlas Atomics, Curio, and the revived Blue Castle project near Green River — operating under Governor Spencer Cox’s Operation Gigawatt, an initiative to double the state’s energy production by 2034.[16] Idaho, Wyoming, and Utah have formalized their coordination through a tri-state agreement, signed by the three governors in April 2025, tying nuclear development to shared energy goals, with Idaho National Laboratory — where Antares went critical — serving as the shared technical anchor.[17] South Carolina, Ohio, and Tennessee each host at least one additional project: the abandoned V.C. Summer AP1000 units, which Santee Cooper voted to restart with Brookfield in October 2025; a new Elementl Power plant in development; and Kairos Power’s Hermes reactor under construction at Oak Ridge, respectively.[18,19,20]

None of this geography overlaps with the southern spine this series mapped in Paper III. Arizona, Texas, and California host the semiconductor fabs and water infrastructure documented there; no reactor in this paper sits near any of them. That absence does not weaken this paper’s central claim. The bet being made is about total generation capacity matching total projected demand, not about any single reactor’s transmission lines reaching any single fab. And as the next section documents, the demand side of that bet no longer has to be inferred.

SECTION 2 — THE DEMAND: THE HYPERSCALERS

Version 1.0 of this paper stated its central limitation plainly: whether any specific reactor ever powers any specific fab was not the point, because no contract connected them. That limitation no longer describes the record. Between early 2024 and mid-2026, the four largest AI infrastructure operators in the world each signed long-term agreements tying named nuclear plants to their own demand, in deals whose duration — seventeen to twenty years — are themselves a finding. No company signs a two-decade power contract against a demand curve it believes might flatten.

Because this buildout produces new announcements almost daily, this paper rates every commitment by the enforceability of the instrument behind it, not the size of its headline number — and those two run in opposite directions. From strongest to weakest: a binding contract or executed transaction — a signed power-purchase agreement, a deed — where money and obligation are both real; a conditional commitment, where real money is obligated but gated on a future decision; a memorandum of understanding, a framework to cooperate carrying no purchase obligation; a letter of intent, a stated intent to transact that binds no one; and a pipeline figure, a self-reported demand number with no instrument behind it at all. The ladder is necessary because in this buildout the gigawatt figures grow as the commitments soften. Amazon’s 1.92 gigawatts sit at the top as a signed contract. At the bottom sits a company like Deep Fission, whose reactor-development pitch advertises a customer “pipeline” that grew from 12.5 to 18.5 gigawatts between late 2025 and mid-2026 — a figure larger than any contracted commitment in this paper, attached to a company whose reactor, as of this writing, exists as a design and a non-nuclear prototype. Deep Fission has drilled a single data-acquisition well and is still seeking permits for its full-scale test borehole; the canister it has delivered to its Kansas site is, by the company’s own description, non-nuclear — built to rehearse the underground installation workflow before any fuel is introduced. The largest number on the board belongs to a company that has not yet drilled a reactor hole, let alone filled one. Every deal below is tagged to its rung.

The four hyperscaler agreements are the top of that ladder — binding, enforceable, decades long. Microsoft signed a twenty-year power-purchase agreement for the output of the Crane Clean Energy Center, the 835-megawatt plant formerly known as Three Mile Island Unit 1, whose restart is targeted for 2027 — accelerated from an original 2028 estimate — and backed by a $1 billion loan from the same DOE Energy Dominance Financing Program that issued the Westinghouse commitment in Section 1.[24] The federal lending instrument and the private demand contract meet inside a single project. Meta signed a twenty-year agreement for 1,121 megawatts from Constellation’s Clinton Clean Energy Center in Illinois, beginning June 2027 — an existing plant whose economics the contract effectively underwrites for a generation.[25] Amazon’s commitment came in two parts: in 2024 it acquired the data-center campus adjacent to Talen Energy’s Susquehanna plant in Pennsylvania for $650 million, and in June 2025 it signed a seventeen-year agreement for 1.92 gigawatts of that plant’s output — the largest single hyperscaler-nuclear commitment on record — collapsing the distance between generation and load to the width of a property line.[26] Google’s arrangement is structurally the most novel. It signed a Master Plant Development Agreement with Kairos Power in October 2024 for up to 500 megawatts of new reactor capacity by 2035 — among the first corporate commitments to reactors that do not yet exist — and a follow-on agreement in August 2025 brought in the Tennessee Valley Authority, a federal utility, as the contracted buyer of the power from the first plant.[27] A federal entity now sits inside a private hyperscaler’s nuclear procurement chain as the off-taker, a blend of public and private roles that fits no prior category cleanly.

Two further agreements sit one rung down, and the distinction is the point. Equinix and Prometheus Hyperscale have each signed not power-purchase agreements but letters of intent with Oklo — Equinix for 500 megawatts, accompanied by a $25 million pre-payment, and Prometheus for 100 megawatts.[28] These are real signals of demand, and the Equinix pre-payment puts money behind one of them, but a letter of intent binds no one to buy or sell; it belongs in the same category as the seven utility letters of intent behind the Westinghouse loan. Oklo — a company that will reappear in this paper’s military and fuel sections — sells power it intends to generate itself, which makes these customer commitments rather than plant-output contracts, and softer ones than the four above.

Two features of this record matter more than any individual deal. The first is timing. The Google-Kairos agreement dates to October 2024; the joint Request for Information that Google, Microsoft, and Nucor issued to aggregate demand for advanced clean-firm power — nuclear among several named technologies, alongside next-generation geothermal, clean hydrogen, and long-duration storage — dates to early 2024, both predating the May 2025 executive orders by more than a year.[29] The demand signal was not conjured by the current administration’s deregulation; it preceded it, which is the same continuity finding this series has made in every sector it has examined. The second is coordination. The joint RFI means the hyperscaler bet is not parallel isolated deal-making but a pooled industrial signal — three of the largest capital allocators in the country telling the clean-power industry, in one document, that the demand is real enough to build against.

This is the missing direct evidence for this paper’s central thesis. Version 1.0 argued that capital was betting on nuclear to match AI and chip demand. The bet now has counterparties, contract durations, megawatt figures, and, in Amazon’s case, a deed.

SECTION 3 — THE MECHANISM

Version 1.0 of this paper identified two federal postures operating inside this buildout. The record now requires three, and they should not be mistaken for one another.

The first is deregulation. Executive Order 14301 created the Reactor Pilot Program, a Department of Energy authorization pathway that lets private developers construct and operate full-scale test reactors outside the standard Nuclear Regulatory Commission licensing process, using the Department’s own authority under the Atomic Energy Act rather than a commercial NRC license.[6] Antares Nuclear, Valar Atomics, and Aalo Atomics moved through this pathway, and Deployable Energy through its successor, the Nuclear Energy Launch Pad — all developing their reactors with private capital, with a compressed federal review standard and shared DOE-NRC staff as the program’s actual contribution.[21] Operation Windlord, the February 2026 airlift of Valar’s unfueled reactor aboard Air Force C-17s, is the apparent exception that proves the rule: though the mission used military aircraft and crews, the Wall Street Journal reported that Valar paid the cost of the flight — estimated at under $1 million — so even this most visibly governmental moment was, in the end, privately funded.[12]

The second is direct federal financing through an equity-adjacent instrument — the posture this series documented at length in Paper IV. The Department of Energy’s Office of Energy Dominance Financing issued its $17.5 billion conditional loan commitment to Westinghouse on June 23, 2026, to finance the fixed-price purchase of long-lead reactor equipment.[7] The loan is expected to route through a Westinghouse special purpose vehicle administering up to five project funding vehicles, each jointly owned by Westinghouse and a utility partner committing $500 million in equity per project before any federal funds are released, with repayment drawn from the sale of the equipment — a structure whose sites, partners, and vehicles remain unnamed as of this writing. The equity-adjacent terms sit in the parallel $80 billion partnership the Department of Commerce announced in October 2025: after a final investment decision, the government is entitled to 20 percent of Westinghouse distributions above $17.5 billion, and may require an IPO and take an equity stake of up to 20 percent at a $30 billion valuation by January 2029.[7,7a] This is the same architecture Paper IV traced through MP Materials, Vulcan Elements, and Trilogy Metals, applied at a larger dollar scale to the company building the reactors. Westinghouse itself is majority owned by Brookfield, with Cameco holding the remainder — the same upstream-supply-chain pattern documented in Paper IV, in which a critical-materials company’s equity holders benefit directly from federal backing.[22]

The third is the guaranteed customer, and it is new to this paper. Two of the three military reactor programs documented in the next section — the Army’s Janus program and the Advanced Nuclear Power for Installations (ANPI) program — share a single contracting model: reactors that are commercially owned and operated (COCO), contracted through Other Transaction Authority, paid against milestones, and explicitly modeled on NASA’s Commercial Orbital Transportation Services program, the framework that produced SpaceX’s cargo and crew capability.[38] Under this posture the government neither steps back as deregulator nor buys in as shareholder. It makes milestone payments as each development stage is met, and commits to buy the resulting power — funding the reactor’s progress in stages while guaranteeing itself as the eventual customer. It is a generalization of the same economic function the Department of Defense’s offtake agreement performed for MP Materials in Paper III, now built into a standing contract architecture. The third military program, Project Pele, does not share this model: it is an earlier, government-directed Defense Department demonstration, built by BWX Technologies (BWXT) under a conventional contract, that Janus and ANPI cite as the precedent they were designed to move beyond.

Taken together, the three postures let the federal government occupy every position around this buildout at once: the regulator stepping aside, the lender-shareholder stepping in, and the customer standing at the end of the line. The distinction among them separates what has already happened from what has only been promised. The Reactor Pilot Program’s reactors were developed with private capital and have gone critical; that work is done and cannot be unwound. The Westinghouse program is a conditional commitment, dependent on a final investment decision not yet made, structured around sites and partners not yet named. The military programs sit in between: contracts signed or shortlisted, sites named, deadlines statutory — but the reactors that will fulfill them, Janus and ANPI alike, are not yet built.

SECTION 4 — THE SPEED

What changed in 2025 was the pace at which the ADVANCE Act’s groundwork got implemented. Executive Order 14300 set explicit numeric deadlines: eighteen months for the NRC to issue a final decision on a new reactor license, and roughly twelve months for license renewals.[10] On March 25, 2026, the NRC finalized “Part 53,” the first entirely new commercial reactor licensing framework since 1989, when the agency created Part 52. Part 53 does not replace the existing Part 50 and Part 52 pathways; it stands alongside them as a third, optional framework developers can choose instead.[23] The order behind this reform cites the underlying bottleneck directly: between 1954 and 1978, the NRC authorized 133 reactors that were completed; since 1978, only two have entered commercial operation.[10] The rule was published in the Federal Register five days after the vote and took effect April 29, 2026.[41] The NRC also cut its hourly review fee for advanced reactor applicants from $318 to $148, a reduction of more than half, effective October 2025 — implementing a fee structure the ADVANCE Act had directed it to build the year before.[41]

Energy Secretary Chris Wright has supplied the historical scale that makes this reform meaningful: in the roughly twenty-five years after the first reactor generated civilian electricity at Idaho National Laboratory in the 1950s, the United States permitted and began construction on more than a hundred reactors before the industry, in his account, “ground to a halt” on regulatory grounds.[17] The current target — ten large reactors under the Westinghouse program, plus a handful of pilot microreactors — is a small fraction of that historical pace, even with review timelines compressed by years. This is the civilian speed this paper has referenced throughout: real, congressionally authorized, bipartisan in origin, and now backed by a specific regulatory framework and fee structure, applying to every developer in this buildout regardless of company or state.

Military speed is a different case, and version 1.0 undersold how much of one. Operation Windlord demonstrated what an actor operating outside the civilian licensing structure entirely can accomplish, using military aircraft, crews, and the Department’s own authority under the Atomic Energy Act — even as the private developer, not the government, paid for the flight. Section 5 shows that demonstration was not isolated: three parallel programs, statutory deadlines, and a contracting model built for speed. None of this says anything about whether the civilian licensing system has gotten faster — that case is made on its own terms above. But the thread this paper flagged in version 1.0 as “worth watching” has, in three weeks of subsequent research, turned out to be a program of record three times over.

SECTION 5 — THE MILITARY BUILDOUT

Version 1.0 treated the military as a single data point — one airlift, Operation Windlord, offered as evidence of what an actor outside the civilian licensing system could do. Continued research shows that framing was too small. The military track is not one demonstration; it is three distinct, parallel programs, each with named vendors, named or shortlisted sites, and dated deadlines.

Project Pele is the foundation — a transportable microreactor program run by the Defense Department’s Strategic Capabilities Office, with BWX Technologies as integrator, targeted for operation in 2028. It predates the others in this paper’s timeline and established the premise they build on: that a reactor can be a deliverable unit rather than a construction site.[35]

The Advanced Nuclear Power for Installations program — ANPI — launched in June 2024, under the Biden administration and a year before the May 2025 executive orders, run jointly by the Defense Innovation Unit, the Army, and the Air Force. Eight companies were named eligible in April 2025: Antares, BWX Technologies, General Atomics, Kairos, Oklo, Radiant, Westinghouse, and X-energy. In April 2026 the program announced its first vendor-site pairings: Radiant at Buckley Space Force Base in Colorado, Westinghouse at Malmstrom Air Force Base in Montana, and Antares at Joint Base San Antonio in Texas.[36] Separately — and often conflated with ANPI in public accounts — the Air Force is running an older, standalone microreactor pilot at Eielson Air Force Base in Alaska, dating to a 2019 congressional mandate, with Oklo’s Aurora as its selected reactor; that pilot is not part of ANPI, though it is frequently reported alongside it.[36]

The Janus Program, launched October 14, 2025, is Army-specific and directly fulfills Executive Order 14299’s September 30, 2028 deadline for a reactor operating on a domestic military base. Nine candidate sites have been named: Fort Benning, Fort Bragg, Fort Campbell, Fort Drum, Fort Hood, Fort Wainwright, Holston Army Ammunition Plant, Joint Base Lewis-McChord, and Redstone Arsenal.[37]

Two observations from this record carry weight beyond the program details.

The first is cross-program overlap — the same finding Paper IV made about federal agencies, now visible among vendors. Oklo appears in the hyperscaler letters of intent of Section 2, the fuel chain of Section 6, and the Eielson pilot at once. Antares appears in the Reactor Pilot Program’s criticality race and in an ANPI site pairing. Westinghouse holds both the $17.5 billion civilian loan commitment of Section 1 and an ANPI pairing at Malmstrom. A small set of companies is being threaded through every federal posture simultaneously.

The second is a precision this paper commits to plainly, because the temptation to overstate is obvious and the series’ credibility depends on resisting it. In their current program definitions, these are fixed, on-base power installations. The documented record describes no combat or weapons application, and this paper describes none. It is worth noting only what the Army’s own solicitation language says: the reactors are to power “installations and nonpermanent operations.” The paper takes that phrase no further than the Army does, and draws no inference from it here.

The continuity finding arrives one more time, because the record keeps producing it: ANPI launched in June 2024, under the prior administration, a year before the executive orders usually credited with the military nuclear push. Like the ADVANCE Act, like the uranium import ban, like the hyperscaler Request for Information, the institutional motion was already under way before the administration most associated with it took office. Three administrations’ worth of the same direction is the oldest finding in this series, and the military track now exhibits it too.

SECTION 6 — THE FUEL BOTTLENECK

A reactor without fuel is a pressure vessel. The buildout documented in Sections 1 and 2 runs, at every point, through a supply chain this series’ readers will recognize: a critical material whose enrichment is dominated by a strategic rival, a bipartisan legislative response, and federal money moving at scale to build a domestic replacement.

The legislative foundation, like the ADVANCE Act, predates the current administration. The Prohibiting Russian Uranium Imports Act, signed by President Biden on May 13, 2024, bans imports of Russian low-enriched uranium through 2040 and unlocked $2.72 billion in domestic enrichment funding that had been conditioned on the ban’s passage.[30] The parallel to Paper II’s rare earth story is exact in structure: an adversary’s dominance of a processing tier, addressed not by tariff but by prohibition plus subsidized domestic replacement.

The money followed on January 5, 2026, when the Department of Energy awarded $900 million each to three companies — Centrus Energy, General Matter, and Orano — to build domestic enrichment capacity, with Centrus and General Matter tasked to produce high-assay low-enriched uranium (HALEU) and Orano to produce conventional low-enriched uranium.[31] The allocation is itself a signal worth reading. A fourth award the same day went to Global Laser Enrichment for $28 million — a developer of laser enrichment, an alternative to the centrifuge technology the three larger awards fund. The government placed roughly $2.7 billion behind centrifuge enrichment and $28 million behind the laser alternative, a ratio of nearly a hundred to one, with no public explanation of why the balance fell so heavily on one side. Whatever the reasoning, the capital reveals a decisive preference; the record does not disclose its basis.

One of the three larger recipients should be familiar: General Matter appears in Section 1’s list of companies operating under Utah’s Operation Gigawatt, meaning the same firm now holds a position in a state-level reactor buildout and a federal enrichment award simultaneously — the cross-program overlap documented in Section 5.

Centrus is the most fully documented of the three. Alongside its federal award, the company announced a $560 million expansion of centrifuge manufacturing at Oak Ridge in January 2026, has disclosed a commercial contract backlog exceeding $2 billion, and has signed a non-binding memorandum of understanding with Korea Hydro & Nuclear Power and POSCO International to explore potential investment.[32] The MOU sits on the ladder’s middle rung — a framework to cooperate, not a commitment to fund — but its direction matters: the domestic enrichment buildout is already reaching toward the same allied-nation partnership structure Paper IV documented in critical minerals.

All of this funds future capacity, and that timing exposes the sharpest fact in the fuel chain: the reactors that reached criticality this year are not running on it. Domestic commercial enrichment barely exists — Centrus, the only U.S. producer, has delivered on the order of 900 kilograms of HALEU to date, against a Department of Energy estimate that domestic demand could reach 50 metric tons per year by 2035.[31b] The fuel actually powering the buildout comes instead from a finite bridge: surplus highly-enriched uranium — weapons-grade and research-reactor material — downblended into the HALEU range. DOE is downblending surplus HEU at the Savannah River Site, recovering more from legacy EBR-II fuel at Idaho National Laboratory, and in May 2026 the National Nuclear Security Administration transferred 1.7 metric tons of surplus HALEU from Japan, the largest single international uranium shipment in its history.[31a] Each of these is described in DOE’s own materials as a bridge — a stopgap until commercial enrichment reaches scale. And the size of that bridge cannot be independently assessed, because the total surplus inventory available for downblending is classified. The visible pieces are a few tonnes here, a shipment there; the whole is not public. The buildout is running, for now, on a stockpile of downgraded weapons material whose remaining depth no outside analyst can measure.

There is a proposed way off that bridge that is not new enrichment at all: recycling spent nuclear fuel to recover usable material. Curio — one of the eight companies in Section 1’s Operation Gigawatt roster — is developing such a process. But it is early-stage: no commercial spent-fuel reprocessing facility operates in the United States, and Curio’s effort is a design, not a plant. It belongs in this section as a stated intention, not a supply. It also points at a constraint this paper does not take up: the spent fuel these reactors will eventually produce is both a disposal question and, potentially, a future fuel source — which is precisely why recycling sits at the seam between the fuel bottleneck and a waste problem that does not yet exist at scale, and will.

The demand side and the fuel side formally connected on June 18, 2026 — the same day Ward 250 reached criticality — when Oklo and Centrus signed a letter of intent for Centrus to supply HALEU for up to five of Oklo’s Aurora reactors at its planned 1.2-gigawatt Ohio campus, with deliveries beginning in 2029.[33] The companies described the agreement as addressing “one of the central constraints facing the advanced nuclear sector.” Precision requires noting what the document is: a letter of intent, not a supply contract — a letter-of-intent instrument, the same category as the seven utility letters of intent behind the Westinghouse loan, real as a signal and unenforceable as a commitment. The connection it describes is real; the obligation it creates is not.

The pattern across this section is the one the whole series has traced. An adversary controls a processing tier. The government responds with prohibition and subsidy, bipartisan across administrations. Domestic capacity gets funded at scale, reaching toward allied partners as it grows. And the instruments doing the reaching — an MOU here, a letter of intent there — are, on inspection, softer than the headlines announcing them, while the fuel actually in the reactors comes from a stockpile no one outside government can size. This is why the paper tags each instrument to its rung rather than counting them all as done, and why “supply secured” is not the same as fuel in a core.

SECTION 7 — THE WATER PROBLEM

Every reactor design in this paper solves the same engineering problem differently and arrives at the same downstream requirement regardless. Sodium, helium, or ordinary water can carry heat out of a reactor core. But once that heat is extracted, every design converts it to electricity the same way every thermal plant has since the nineteenth century: by boiling water into steam and driving a turbine with it. The exotic coolant changes how heat moves from the core to the water; it does not change the water-intensive steam cycle that follows. A sodium-cooled fast reactor and a helium-cooled high-temperature reactor both still need to reject the steam’s heat afterward — and rejecting that heat is where the water is spent. The only way to escape it is to reject the heat to air instead, through an air-cooled condenser, which is the exception this section ends on, not the rule it begins with.

The industry’s engineering response, audited claim by claim in Paper III, resolved into one fully creditable standard and a set of softer ones. TSMC’s first Arizona fab draws 4.75 million gallons a day and recycles 65 percent through its existing water resource center. Its dedicated reclamation plant — planned before the fab opened, but broken ground only in August 2025 and not operational until 2028 — is designed to reach a 90 percent recycling goal aimed at the hardest version of the problem: the ultrapure water that touches the wafer, not merely the water that cools the building. The sequence is worth stating exactly, because it is the honest version of the claim. TSMC did not solve its water problem before switching on the fab. The fab has drawn water at 65 percent recycling since late 2024; the plant that reaches 90 percent will not exist until 2028. What TSMC did was commit to the hardest version of the problem, state plainly what it had built and what it had not, name the date, and start building against demonstrated precedent in its own Taiwan operations. Measured against that standard, the other claims held up less well. Intel’s Arizona “net positive” status is real but is an end-of-pipe watershed return, not core-loop recycling. Sherman, Texas’s new wastewater plant is genuine, completed, citywide infrastructure that returns treated water to a creek — not a fab-specific system. GlobalWafers’ federally tracked 50 percent commitment is qualified by “commercially reasonable efforts”; Texas Instruments’ 70 percent aspiration carries no federal tracking and sits far above the company’s own disclosed 27-to-29-percent record.[40]

That corrected restatement matters here because the comparison holds — it just holds more precisely. The semiconductor industry saw its water constraint coming, and its most advanced operator committed capital to the hardest version of the problem while its fab was already running, and said plainly what it had and had not yet built. The nuclear buildout, with limited exceptions, has not done the same — and the exceptions are instructive precisely because of how few and how partial they are.

Start with what the criticality milestones did and did not prove. The reactors that reached criticality this year demonstrated core physics at zero or near-zero power; none ran the steam cycle that consumes water, because that cycle only operates at power, which none of them has yet sustained. The water-consuming systems of these designs are, as of this writing, undemonstrated — not because anyone has hidden them, but because the tests conducted so far do not reach them. The part of the machine that determines its water footprint has not yet been switched on.

Two projects have engaged the water constraint directly, by opposite routes, and both are worth stating exactly. The first is Blue Castle, near Green River, Utah — the one project that has visibly changed course because of water. Its original design, two large AP1000 reactors, depended on water rights the company secured around 2012 and then lost amid a mix of financing trouble, regulatory delay, and unresolved litigation. Its 2026 revival did not restore that position; it moved toward Holtec SMR-300 units, whose available air-cooled condenser configuration is marketed for exactly this kind of arid siting. The redesign identifies air cooling as its intended answer to the water problem — though, per the most recent local reporting, the project has not confirmed that the air-cooled configuration is finalized. It is the clearest case in this paper of the water constraint forcing a change in reactor technology, and even it is not yet settled.

The second is TerraPower’s Natrium plant at Kemmerer, Wyoming, and it engages the constraint the other way — by inheriting existing infrastructure rather than redesigning around it. Natrium is sited beside the Naughton generating station, and its developers cite the existing plant’s cooling-water intakes among the reasons for the location. Public materials frequently describe Naughton as a “retiring coal plant,” which is accurate as to coal and incomplete as to the plant: Naughton’s coal units ceased operation at the end of 2025, but the station is converting to natural gas and continuing to run. Natrium is therefore not stepping into a departing plant’s freed-up water; it is adding a new reactor’s steam-cycle demand alongside a fossil plant that keeps drawing. Its stated water plan reaches no further than the shared existing intakes — which is thinner than it sounds, because those intakes still serve an operating gas plant.

Everything else in the new-build track is quieter still. The hyperscaler agreements of Section 2 overwhelmingly attach to existing or restarting plants — Susquehanna, Clinton, Crane — whose water footprints were permitted decades ago and are already sunk into their watersheds. The water question therefore concentrates almost entirely on the genuinely new build: the ten Westinghouse AP1000s at five undisclosed sites, the Kairos fleet Google’s agreement contemplates, and the military microreactors of Section 5, whose cooling architectures are the least publicly specified systems in this entire paper. The track where the water question is unanswered and the track where disclosure is thinnest are the same track. And several of these plants use the same conventional steam-cycle generation in some of the same water-stressed basins Paper III already documented absorbing demand from semiconductor fabrication — the Colorado River, on which Utah’s projects draw, has been in measured, sustained decline for over two decades, independent of anything in this buildout.[29-W]

The semiconductor industry’s best operator did not wait for its water constraint to become a crisis. It named the gap while its fab was already running, committed capital to closing it, and said publicly how far it had and had not gotten. This buildout, with one redesign and one inherited intake, has not yet shown that it has done the same — and the plants where the question is hardest are the ones about which the least has been said.

CONCLUSION

This paper set out, in its first version, to determine whether the speed and structure of the federal government’s nuclear buildout matched the pattern this series identified in critical minerals. It does. Three weeks of continued research have not altered that finding; they have supplied the evidence the original publication date could not yet capture, at a rate that is itself part of the finding.

The demand side of the bet now has names. Four hyperscalers have signed binding, long-term agreements against named plants — restarted, existing, and not yet built — with a federal utility sitting inside one deal as the offtaker and a federal loan inside another. Below them, softer instruments carry larger numbers, which is why this paper rated every commitment by the enforceability of the instrument behind it rather than the size of its headline figure: the ladder runs from a signed contract down to a self-reported pipeline, and in this buildout the gigawatts grow as the commitments soften.

The fuel side has a bipartisan prohibition, billions in unlocked and awarded funding, and a lopsided bet — roughly a hundred to one — on proven centrifuge enrichment over the laser alternative, placed without public explanation. And beneath the new commercial capacity being built sits a quieter fact: the reactors that reached criticality this year are running not on that capacity but on a finite bridge of downblended surplus weapons and research uranium, whose remaining depth is classified and cannot be measured from outside.

The military side is not one airlift but three parallel programs, with statutory deadlines and a contracting model — the government as guaranteed customer — distinct enough from deregulation and equity-adjacent lending that this version names it a third federal posture in its own right. Taken together, the three postures let the federal government stand in every position around this buildout at once: the regulator stepping aside, the lender-shareholder stepping in, and the customer waiting at the end of the line.

The institutional record still runs continuously across the handoff between administrations, and the new material deepens that continuity rather than complicating it: the ADVANCE Act, the uranium import ban, the hyperscaler Request for Information, and the ANPI military program all predate the executive orders usually credited with this buildout. Three administrations, one direction — the oldest constant in this series, now confirmed in a fourth sector.

And the disclosure gap has widened in proportion to the activity. Which five sites will receive the $17.5 billion is not public. What the hyperscalers are paying, in most cases, is not public. The military’s final vendor and site selections are pending. The depth of the fuel stockpile now powering the reactors is classified. And the water requirement of every genuinely new-build reactor in this paper — beyond one project that redesigned toward air cooling and one that inherited a fossil plant’s intakes — goes unaddressed in any document this paper has reviewed. That last gap is the sharpest, because the hyperscaler deals overwhelmingly attach to existing or restarting plants whose water footprints were settled decades ago. The water question concentrates almost entirely on the new-build track — the same track where disclosure is thinnest.

None of the material added in this version contradicts the paper it revises. All of it confirms the original thesis at a scale and speed the first publication could not capture. Everything documented here is a floor, not a ceiling — and the rate at which new, dated, independently sourced material surfaced in three weeks is itself evidence for this paper’s core argument: pace outrunning visibility.

REFERENCES

Sources are primary wherever available — SEC filings, the Federal Register, agency press releases and program records, and company disclosures. Where a claim could only be reached through secondary reporting, that is noted in the entry.

[1] ADVANCE Act of 2024, Public Law 118-67, signed July 9, 2024; U.S. Senate Committee on Environment and Public Works, “Signed: Bipartisan ADVANCE Act to Boost Nuclear Energy Now Law,” July 9, 2024. (Senate 88-2; House 393-13.)

[2] U.S. Department of Energy, “U.S. Sets Targets to Triple Nuclear Energy Capacity by 2050” and the U.S. Nuclear Energy Deployment Framework, energy.gov, November 12, 2024; DOE, “Pathways to Commercial Liftoff: Advanced Nuclear” (~100 GW in 2024 to ~300 GW by 2050).

[3] Executive Order 14302, “Reinvigorating the Nuclear Industrial Base,” May 23, 2025, Federal Register.

[4] U.S. Department of Energy, “Department of Energy Celebrates First Advanced Reactor Achieving Criticality” (Antares Nuclear, Mark-0), energy.gov, June 4, 2026.

[5] U.S. Department of Energy, “Department of Energy Celebrates Second Advanced Reactor Achieving Criticality” (Valar Atomics, Ward 250), energy.gov, June 18, 2026.

[6] Executive Order 14301, “Reforming Nuclear Reactor Testing at the Department of Energy,” May 23, 2025, 90 Fed. Reg. 22591.

[7] Westinghouse Electric Company; Brookfield Asset Management; and U.S. Department of Energy, Office of Energy Dominance Financing, conditional loan commitment announcement, June 23, 2026; Cameco Corporation disclosure of the American Nuclear Supply Chain Loans structure (Westinghouse SPV; up to five project funding vehicles; $500 million equity per project), June 2026.

[7a] Cameco Corporation, Form 40-F (FY2025), U.S. Securities and Exchange Commission; Brookfield Asset Management, Form 8-K, Exhibit 99.1, October 28, 2025; “United States Government, Brookfield and Cameco Announce Transformational Partnership,” joint press release, October 28, 2025 (the $80 billion partnership; 20% of distributions above $17.5 billion; IPO/equity terms).

[9] Executive Order 14299, “Deploying Advanced Nuclear Reactor Technologies for National Security,” May 23, 2025, 90 Fed. Reg. 22581.

[10] Executive Order 14300, “Ordering the Reform of the Nuclear Regulatory Commission,” May 23, 2025, 90 Fed. Reg. 22587 (400 GW by 2050; 18-month/12-month licensing deadlines; the 133/two-reactor figure appears in the order’s preamble).

[11] Valar Atomics, technical materials and public statements, valaratomics.com, 2026 (“Cold ≠ Hot” framing).

[12] U.S. Air Force / DVIDS, Operation Windlord release, February 15, 2026 (multi-wing airlift, March ARB to Hill AFB); The Wall Street Journal, reporting that Valar paid the cost of the flight (estimated under $1 million), as cited in Partnership for Global Security and The National Interest, 2026. Marked WSJ-reported-via-secondary.

[13a] U.S. Department of Energy, statement on third advanced reactor criticality (Deployable Energy, Unity, Nuclear Energy Launch Pad), energy.gov, July 1, 2026; Deployable Energy, criticality announcement, July 1, 2026.

[13b] U.S. Department of Energy and Aalo Atomics, criticality announcement (Critical Test Reactor, July 4, 2026), energy.gov, July 2026.

[14] Valar Atomics funding disclosures, 2025-2026 (~$130 million raised toward Ward 250, per company statements and contemporaneous reporting).

[15] U.S. Nuclear Regulatory Commission, press release No. 26-028, “NRC Issues First Commercial Reactor Construction Permit in Nearly a Decade” (TerraPower Natrium), March 2026; U.S. Department of Energy, “NRC Issues Construction Permit for TerraPower’s Natrium Advanced Reactor.”

[16] State of Utah, Office of Governor Spencer Cox, “Operation Gigawatt” initiative materials, October 8, 2024 and 2025-2026.

[17] State of Utah, Office of Governor Cox, tri-state (Utah-Idaho-Wyoming) nuclear coordination agreement, April 2025; Cowboy State Daily, coverage of Energy Secretary Chris Wright remarks, December 2025.

[18] Santee Cooper, V.C. Summer site status statements, October 2025 (board approval to restart with Brookfield); American Nuclear Society reporting, 2025-2026.

[19] Elementl Power, southeast Ohio project announcement, June 2026; American Nuclear Society reporting.

[20] Kairos Power, Hermes reactor construction status, Oak Ridge, Tennessee; U.S. Department of Energy Advanced Reactor Demonstration Program records.

[21] U.S. Department of Energy, Reactor Pilot Program materials, 2025-2026; Nuclear Innovation Alliance analysis of the DOE 45-day safety-document review standard and embedded NRC staff.

[22] Brookfield Corporation and Cameco Corporation, Westinghouse Electric Company ownership disclosures, public filings.

[23] U.S. Nuclear Regulatory Commission, press release No. 26-035, “NRC Approves Final Rule for Advanced Reactor Licensing” (Part 53), March 25, 2026; Part 53 final rule, 91 Fed. Reg. 15696, published March 30, 2026 (effective April 29, 2026).

[24] Constellation Energy, Crane Clean Energy Center (Three Mile Island Unit 1) 20-year PPA with Microsoft, press release, September 2024, and Constellation Form 10-K, U.S. Securities and Exchange Commission (835 MW; 2028 in-service estimate); U.S. Department of Energy, Energy Dominance Financing $1 billion loan (closed November 18, 2025), energy.gov (restart accelerated to 2027 per subsequent reporting).

[25] Constellation Energy, Clinton Clean Energy Center 20-year PPA with Meta, press release, June 3, 2025 (1,121 MW; begins June 2027).

[26] Talen Energy, Form 8-K, U.S. Securities and Exchange Commission, June 2025 (1,920 MW Susquehanna PPA, ~17 years to 2042); Amazon/Talen $650 million data-center campus acquisition, 2024.

[27] Kairos Power, Google Master Plant Development Agreement, October 14, 2024 (up to 500 MW by 2035); Kairos Power / Tennessee Valley Authority / Google follow-on agreement, August 2025 (TVA as contracted buyer, Hermes 2).

[28] Oklo Inc., Form 8-K and shareholder materials, U.S. Securities and Exchange Commission, 2025-2026 (non-binding letters of intent: Equinix 500 MW with $25 million pre-payment; Prometheus Hyperscale 100 MW).

[29] Nucor Corporation, joint Request for Information with Google and Microsoft for advanced clean-firm power (nuclear, next-generation geothermal, clean hydrogen, long-duration storage), press release, March 19, 2024.

[29-W] U.S. Bureau of Reclamation, Colorado River operating-condition news releases and 24-Month Studies, usbr.gov, 2025-2026 (two decades of drought; Lake Powell ~25% full; Level 1 Shortage; Post-2026 Operational Guidelines Draft EIS reframing the crisis as a fundamental hydrologic shift).

[30] Prohibiting Russian Uranium Imports Act, signed May 13, 2024 (ban through 2040); associated $2.72 billion domestic enrichment appropriation, U.S. Department of Energy; U.S. Department of State.

[31] U.S. Department of Energy, enrichment capacity awards of $900 million each to Centrus Energy, General Matter, and Orano, January 5, 2026, energy.gov; Centrus Energy Form 8-K, U.S. Securities and Exchange Commission (Global Laser Enrichment received a separate $28 million award the same day).

[31a] U.S. Department of Energy, HALEU Availability Program materials and Savannah River Site / EBR-II downblending descriptions, energy.gov; U.S. Department of Energy, National Nuclear Security Administration, “U.S. Secures Largest-Ever HALEU Shipment” (1.7 metric tons transferred from Japan), May 2026.

[31b] U.S. Department of Energy, Office of Nuclear Energy, “What Is High-Assay Low-Enriched Uranium (HALEU)?” and “HALEU Frequently Asked Questions,” energy.gov (domestic HALEU demand could reach 50 metric tons per year by 2035).

[32] Centrus Energy, press releases: $560 million Oak Ridge centrifuge manufacturing expansion (January 23, 2026); commercial contract backlog disclosures (exceeding $2 billion); non-binding memorandum of understanding with Korea Hydro & Nuclear Power and POSCO International, August 25, 2025.

[33] Centrus Energy, Form 8-K, U.S. Securities and Exchange Commission, June 18, 2026; Oklo Inc., joint announcement (letter of intent, HALEU for up to five Aurora reactors, 1.2-gigawatt Ohio campus, deliveries beginning 2029).

[35] U.S. Department of Defense, Strategic Capabilities Office, Project Pele program materials; BWX Technologies (BWXT) integrator role; TRISO fuel delivery and 2028 operational target.

[36] U.S. Department of the Air Force, Advanced Nuclear Power for Installations (ANPI) program: eight-company eligibility (April 2025) and April 2026 vendor-site pairings (Radiant/Buckley SFB; Westinghouse/Malmstrom AFB; Antares/Joint Base San Antonio); World Nuclear News and Partnership for Global Security reporting. The Eielson AFB microreactor pilot (Oklo Aurora), dating to a 2019 NDAA mandate, is a separate standalone program, not part of ANPI.

[37] U.S. Army, Janus Program announcement, army.mil, October 14, 2025; nine candidate sites (Fort Benning, Fort Bragg, Fort Campbell, Fort Drum, Fort Hood, Fort Wainwright, Holston Army Ammunition Plant, Joint Base Lewis-McChord, Redstone Arsenal); Army solicitation language (“installations and nonpermanent operations”).

[38] U.S. Army, Janus Program materials, army.mil (commercially owned and operated (COCO) structure, Other Transaction Authority, milestone-based contracting explicitly modeled on NASA’s Commercial Orbital Transportation Services program); Partnership for Global Security on the ANPI COCO/PPA structure.

[40] Blue Collar Analytics, “Where Fortress America Lands,” Fortress America Series, Paper III, v1.2, Section 4, and its underlying primary sources: TSMC Arizona sustainability disclosures, tsmc.com (4.75 MGD; 65% recycling; IRWP August 2025 groundbreak, 90% goal by 2028); Intel newsroom (Arizona “net positive” water); City of Sherman, Texas, Post Oak Wastewater Treatment Plant (16 MGD citywide); National Institute of Standards and Technology CHIPS award pages (GlobalWafers 50% “commercially reasonable efforts”; Texas Instruments); Texas Instruments corporate disclosures (70% aspiration; 27-29% recorded).

[41] U.S. Nuclear Regulatory Commission, Part 53 final rule, 91 Fed. Reg. 15696 (effective April 29, 2026); NRC advanced-reactor hourly fee reduction from $318 to $148, effective October 2025, implementing ADVANCE Act Section 201.

FORTRESS AMERICA SERIES

Paper I: Fortress America — Hemispheric Consolidation, Infrastructure Cascades, and the Regional Transformation of the American Economy

Paper II: If It Can Work, It Can Fight — The Case for Domestic Humanoid Robotics Manufacturing

Paper III: Where Fortress America Lands — The Southern Spine and the Race to Build It

Paper IV: The Government Stake — Equity, Speed, and the Limits of Disclosure

Paper V: The Atom and the Chip — The Rush to Nuclear, and the Water Left Behind

This paper represents independent analytical and systems research and is the fifth paper in the Blue Collar Analytics Fortress America series. Nothing contained herein constitutes financial advice, investment recommendations, or legal counsel. Readers should conduct their own due diligence and consult qualified professionals before making investment decisions.

WHERE FORTRESS AMERICA LANDS
The Southern Spine and the Race to Build It

Author: Adam Wood | Publication: Blue Collar Analytics | Series: Fortress America, Paper III | Date: June 2026 | Version: 1.1

Revision note (v1.1): This version corrects a citation error in Section 4 regarding TSMC’s Arizona water reclamation system. The original version cited figures (10,000 cubic meters per day scaling to 36,000 cubic meters per day by 2026) that describe TSMC’s Taiwan facilities, not its Arizona operations. This version also substantially rebuilds Section 4 and the paper’s conclusion. Every public water-recycling claim made by an operator in this buildout — TSMC, Samsung’s water partner EPCOR, Intel, the city of Sherman, GlobalWafers, Texas Instruments, and the water-engineering firm Gradiant — was checked individually against primary sourcing. One of these, TSMC’s, held up as an honest precedent, voluntarily-adopted standard against which the others can fairly be measured; the rest describe something narrower, softer, or less verifiable than their original framing suggested. Section 4 is rebuilt around that finding, and the paper’s conclusion is revised to match. Substantive conclusions elsewhere in the paper are unchanged.

EXECUTIVE SUMMARY

The Fortress America framework established in Papers I and II predicted that hemispheric consolidation would produce a domestic infrastructure buildout traceable in real time through capital flows, regulatory actions, and physical deployment. Paper III maps where that buildout is physically landing — a California-Texas-Arizona southern spine emerging as the industrial core of the American economy for the next generation. The rare earth supply chain from mine to magnet to motor is now domestically traceable for the first time. More than a dozen semiconductor fabs are at some stage of production or active construction across this corridor. The binding constraint on the entire buildout is water in an arid region, and public claims about the engineering response to that constraint vary widely in how much scrutiny they can withstand — from one operator’s precedented, honestly-stated target to several others whose claims describe less than their framing implies.

SECTION 1 THE SOUTHERN SPINE

Arizona is the most concentrated single-state semiconductor investment in American history. Since 2020, the Arizona Commerce Authority has documented more than 60 semiconductor expansions representing more than $205 billion in investment — a figure Governor Katie Hobbs cited directly at SEMICON West 2025, the semiconductor industry’s own annual trade conference, held in Phoenix for the first time. [1] By early 2025, Industrial Info Resources tracked nearly $50 billion in active industrial construction across the state, with semiconductor fabs, data centers, power plants, and mining operations leading the pipeline. [2]

Texas is building a different but complementary concentration — multiple companies across a contiguous corridor rather than one dominant anchor. Samsung’s Taylor fab on 1,200 acres is the largest foreign direct investment in Texas on record, targeting operational status at the end of 2026 with 1,500 permanent employees and production of 2-nanometer leading-edge chips. [3] In July 2025, Samsung filed a formal regulatory disclosure on the Seoul Stock Exchange confirming a $16.5 billion semiconductor supply agreement running through December 31, 2033; Tesla CEO Elon Musk confirmed the same day that Samsung’s Taylor facility would produce Tesla’s next-generation AI6 chip, used across Tesla’s AI products including autonomous vehicles and the Optimus humanoid robot. [4] The robotics manufacturer and the semiconductor fab are already contractually linked in the same state through the end of the decade.

Both states are absorbing pressure from different directions, and the pattern reveals which part of the buildout is durable. Semiconductor fabs faced real delays in 2023-2024 — Taiwan Semiconductor Manufacturing Company’s (TSMC) first Arizona fab slipped roughly a year due to skilled-labor shortages, and its second fab moved from a 2026 target to 2027-2028 — but both are now active, with TSMC reporting pilot production ahead of its most recent revised schedule as of April 2026. [5] Data centers in the same state have faced a different kind of setback entirely: data center opposition nationwide blocked or delayed at least 75 projects worth roughly $130 billion in the first quarter of 2026 alone — a single-quarter total matching all of 2025 combined, and more than double the $64 billion blocked cumulatively between 2023 and early 2025. Arizona has been part of this wave throughout, where Chandler’s city council unanimously rejected a proposed AI data center in December 2025 and a $14 billion West Valley project was withdrawn after the city declined to approve rezoning. [6] The fabs slipped on labor and equipment timelines and recovered. The data centers are being stopped by the communities they are built in.

The southern spine is the operational core of the domestic semiconductor buildout today, with Arizona in high-volume production and Texas ramping to production in 2026-2027. New York and Ohio represent a second wave of the same institutional pattern — committed capital, steel in the ground, state and federal funding locked in — coming online in the 2030-2032 timeframe. [7] The fab-by-fab production status confirming this geography as the industrial core of Fortress America is detailed in Section 3.

SECTION 2 MINE TO MAGNET TO MOTOR: THE DOMESTIC RARE EARTH SUPPLY CHAIN

Semiconductor capacity is the precondition for advanced manufacturing; without the chip, a robot has no intelligence to actuate in the first place. But intelligence without actuation is just as inert — a robot that can think but cannot move is a research demonstration, not a defense asset. The rare earth supply chain is the second non-negotiable bottleneck, and for the first time, it is now traceable end to end on American soil.

Rare earth elements are not a single-application input. They are foundational across the entire advanced manufacturing stack: permanent magnets containing neodymium, praseodymium, dysprosium, terbium, and samarium power the precision actuators inside semiconductor fabrication equipment itself, optical crystals doped with yttrium and lanthanum enable the lasers used in advanced lithography, and the same magnet chemistry that moves a wafer with micron-level precision inside a fab is what moves a robotic joint with the torque and control a humanoid platform requires. [8] The supply chain risk is not confined to one product category. China accounts for over 90% of global rare earth production broadly, and approximately 85 to 90% of neodymium-iron-boron (NdFeB) magnet production specifically — concentration that touches the fabs documented in Section 1 as much as the robots documented in Paper II. [9]

The chain that resolves this dependency, for the first time, exists entirely within U.S. borders. Mountain Pass, California, operated by MP Materials, is the only operational rare earth mining and processing facility of scale in the United States — the extraction point. Independence, MP Materials’ magnet manufacturing facility in Fort Worth, Texas, is already producing at an initial 1,000 metric ton annual capacity and expanding — the conversion point, where processed rare earth oxide becomes finished magnet. [10] A robot built in Texas with American rare earth magnets processed from California ore is a fundamentally different strategic asset than one assembled with components Beijing can restrict on short notice.

The federal government’s commitment to this chain is structured with more depth than a typical subsidy. In July 2025, the U.S. Department of Defense (DoD) entered a multibillion-dollar agreement with MP Materials that made DoD the company’s largest shareholder, acquiring approximately 15% of outstanding shares through a $400 million purchase of newly created preferred stock, convertible at $30.03 per share. [11] The package extends well beyond equity: a $150 million loan to expand heavy rare earth separation capacity at Mountain Pass; a 10-year price floor of $110 per kilogram for MP’s neodymium-praseodymium (NdPr) products, insulating the company from Chinese price manipulation; and a 10-year offtake agreement under which DoD guarantees that 100% of magnet output from a new “10X” production facility — targeting 10,000 metric tons of annual capacity by 2028 — will be purchased by defense and commercial customers. [12] MP secured an additional $1 billion in commercial financing from JPMorgan Chase and Goldman Sachs on the strength of the federal commitment. As a binding condition of the agreement, MP Materials ceased all sales of rare earth products to China in July 2025 — a cessation already reflected in the company’s own reported revenue — and committed not to renew its existing offtake agreement with China’s Shenghe Resources at its January 2026 expiration, ending the one remaining contractual link between America’s primary rare earth producer and a Chinese state-affiliated buyer. [13]

This is not an isolated transaction. It is the first and most fully developed example of a repeating federal instrument: the government taking direct equity positions in companies that control the extraction and early-stage processing of strategic minerals, rather than in the factories that turn those minerals into finished goods. The same Department of Defense office that structured the MP Materials deal has since taken a 10% stake, with warrants for an additional 7.5%, in Trilogy Metals to advance copper and cobalt development in Alaska, and a separate equity position in Lithium Americas. [14] What this pattern signals about how the government is choosing to secure the upstream end of the supply chain, rather than only the downstream manufacturers, will be explored in full in the final paper of this series.

The one legislative risk to this chain is the Section 45X Advanced Manufacturing Production Credit, which subsidizes domestic production of critical minerals and components. [15] For most 45X-eligible products, the credit phases down starting in 2030 and expires entirely in 2033. Critical minerals, however, sit on a separate track within the same credit: the phaseout for critical minerals does not begin until 2031, with full expiration in 2033 — and unlike solar and wind components, critical mineral production was carved out from the general sunset schedule precisely because of its strategic designation. [16] Even within that more favorable schedule, non-renewal in 2033 would still represent a real long-term risk to the economics of expanding domestic magnet production at the pace the DoD offtake agreement assumes.

But the DoD offtake agreement itself is what makes that risk politically difficult to realize. The Pentagon is now MP Materials’ largest shareholder, has committed to a 10-year price floor, and has guaranteed purchase of 100% of a facility’s output through approximately 2035 — two years past the 45X expiration date. A Congress that allowed the critical minerals credit to lapse without renewal would be undermining an investment the Department of Defense itself structured, sized, and continues to hold equity in. Attacking the credit’s renewal becomes, in practice, attacking the defense department’s own balance sheet and its own stated rare earth independence strategy — a substantially higher political bar than allowing a generic manufacturing subsidy to expire on schedule.

The mine-to-magnet-to-motor chain described here is not a projection. Mountain Pass is producing today. Independence is producing today. The DoD’s equity, loan, and offtake commitments are signed, public, and filed with the U.S. Securities and Exchange Commission (SEC). [17] What remains is scale — the 10X facility in Northlake reaching its 10,000-metric-ton target by 2028, and the broader expansion of domestic magnet capacity beyond MP Materials alone, an effort already extending into Wyoming and Alaska on separate tracks. The chain that supplies the humanoid robotics manufacturing imperative identified in Paper II now exists, end to end, without crossing Chinese territory at any point.

SECTION 3 THE SEMICONDUCTOR SPINE

The strategic vulnerability this buildout addresses is stark and well documented. The United States declined from producing roughly 37 to 40% of global semiconductors in 1990 to approximately 10 to 12% today, and as of the passage of the CHIPS and Science Act in 2022, none of the world’s most advanced chips were manufactured domestically. [18] The correction underway since is not a slow policy response — it is one of the fastest industrial buildouts in American history, and its pace is best understood by naming what is actually under construction, what is already producing, and what remains delayed.

TSMC’s Arizona campus is the clearest evidence of acceleration. Fab 21’s first facility entered high-volume 4-nanometer production in Q4 2024 and is now manufacturing chips for Apple and Nvidia — the first time TSMC has produced cutting-edge AI silicon outside Taiwan. [19] The second fab’s construction was completed in 2025, with equipment installation beginning in the third quarter of 2026 and 3-nanometer production targeted for 2027 — a full year ahead of the original schedule, according to TSMC’s own CEO. [20] A third fab broke ground in April 2025, targeting 2-nanometer and A16 process technology by the end of the decade. [21] Total committed investment has grown from an initial $12 billion in 2020 to $165 billion across six planned fabs, with TSMC describing long-term framework expansion reaching approximately $465 billion — the largest single foreign direct investment in American history. [22]

Texas Instruments offers the clearest example of speed. Construction began on its Sherman, Texas site in mid-2022; the first fab, SM1, began production on December 17, 2025 — three and a half years from groundbreaking to output. [23] The facility produces foundational analog and embedded processing chips, the components used in nearly every electronic device, including industrial robotics and automotive systems. [24] SM2’s exterior shell is already complete, with cleanroom installation underway in 2026. The Sherman campus is part of a $60 billion investment across seven planned fabs in Texas and Utah. [25]

Samsung’s Taylor, Texas fab presents a more complicated picture, and an honest accounting requires saying so. Originally targeted for 2024 production, the facility was delayed twice — once to upgrade its process technology from 4-nanometer to the more advanced 2-nanometer node, and again reportedly due to insufficient near-term customer demand. [26] As of May 2026, Samsung’s own foundry leadership confirmed that customer production, including for Tesla, is scheduled to begin in 2027, with the facility’s third-generation 2-nanometer process now in installation. [27] The delay is a demand and technology-upgrade story, not a capital withdrawal — Samsung’s total investment in the site grew from $17 billion to $44 billion over the same period, and the company received $4.75 billion in direct CHIPS Act funding to proceed. [28]

Intel’s Chandler, Arizona campus adds two more fabs to the spine. Fab 52 is in high-volume production on Intel’s 18A process — the first U.S. facility to cross the 2-nanometer threshold, with Intel’s own chief technology officer describing it as capable of more than 10,000 18A wafer starts per week — while Fab 62 is under construction and expected to be ready around 2028. [29] Intel’s commitment to Arizona has held even as the company cancelled comparable projects in Germany and Poland, consolidating its advanced manufacturing investment domestically rather than abroad. [30]

Taken together, the southern spine’s fab buildout shows a pattern consistent with the rest of this paper’s findings: delays have been common, but they have been delays of one to two years driven by labor availability, technology upgrades, and customer demand — not cancellations, and not capital withdrawal. Every fab named above is either producing today or under active construction with a committed completion date. That distinguishes this buildout sharply from the data center cancellation wave described in Section 1, where the setbacks have come from local political rejection rather than engineering or market timing.

SECTION 4 THE WATER CONSTRAINT AND THE ENGINEERING RESPONSE

The binding constraint on the southern spine is water, in a region already under acute and well-documented stress. Arizona’s water supply depends heavily on the Colorado River, which has experienced sustained shortage conditions for over two decades, and on groundwater aquifers facing long-term depletion as the state’s population and industrial base both grow. [31] Texas faces a parallel and independently documented problem: the state’s own water planning authorities project a long-term supply deficit of nearly 6.9 million acre-feet, with officials warning that without expanded infrastructure, demand could outstrip supply during the next prolonged drought. [32] Into both constrained systems, the fab buildout documented in Section 3 introduces a new and significant industrial water demand layered on top of an existing shortage.

The scale of that demand is best understood by direct comparison. Estimates of a single fab’s daily water draw vary by facility size and process node — independent technical sources put the range anywhere from roughly 3 million to 10 million gallons per day, with the largest individual facilities and multi-fab campuses running higher still. [33] The World Economic Forum’s commonly cited industry figure, 10 million gallons per day, sits at the top of that range rather than describing a typical fab. A single large data center, by contrast, consumes roughly 1.5 to 5 million gallons per day. [34] Even using the low end of the fab range, a fab remains the larger water consumer — and at the high end, cited most often in public discussion, the gap widens to a factor of two to four.

One operator’s public commitments stand apart from the rest of this section, and it is worth establishing why before turning to the others. TSMC states plainly, in its own public materials, that its Arizona fabs “aim to achieve” a 90% water recycling rate, pursued through a purpose-built Industrial Reclamation Water Plant (IRWP) with a stated “design goal” of near-zero liquid discharge. [35] The IRWP broke ground in August 2025 and will not be operational until 2028. TSMC’s own language has always reflected that timeline honestly — an aim, a goal, a plant under construction. Their aims and designs have been clearly stated and appear to model the existing capabilities already demonstrated in their Taiwanese operations, where the company maintains recycling rates exceeding 85%. [36] That standard is also the harder and more expensive of the two paths available to a fab operator: recycling the core ultrapure water that touches the wafer, rather than the comparatively simpler work of recycling water used in cooling towers and air scrubbers. Nothing required TSMC to choose the harder path. The City of Phoenix’s own public position is that it has sufficient water capacity for TSMC regardless of whether the 90% target is ever reached, and the figure does not appear among the enforceable milestones attached to TSMC’s $6.6 billion CHIPS Act award. [37] TSMC set this target voluntarily, stated it in public, and did so in a community where water has been, by local reporting, “the flash point in many neighborhood meetings and planning sessions” — in front of the audience most likely to notice if the company fell short. [38] Whether TSMC’s Arizona plant reaches 90% recycling on the 2028 timeline the company has set for it is a separate, later question, and this paper takes no position on it in advance. What can be said now is that TSMC’s public description of its own progress has not, at any point, claimed more than the company has actually built.

Measured against that standard — public claims that describe no more than what has actually been achieved or actually been committed to, and that point to demonstrated precedent where precedent is claimed — the other claims made across this buildout hold up less consistently.

Samsung’s water partner in Taylor, EPCOR, announced a target in a July 2023 press release: reclaiming or reusing 75% of the Blue Sky Water Reclamation Facility’s process water. [39] Unlike TSMC’s figure, this one has not been paired with an operational timeline, an engineering description, or any subsequent public update. A local outlet that sought detail on how the target would be met, roughly six weeks after the announcement, received no response beyond the same press-release language, and none has surfaced publicly in the nearly three years since. The claim may be accurate. It has not been made possible to check.

Intel describes its Ronler Acres, Oregon operations as having reached “net positive” water status as of 2022 — returning more treated water to the local watershed than the company withdraws. [40] This claim is true, and it is a genuine achievement. It is also, by industry reporting on the underlying mechanism, an “end-of-pipe” result: water is treated and returned to secondary systems and to the watershed, rather than recycled back into the core ultrapure loop that touches the wafer — the same category of water TSMC’s existing Arizona system currently handles at a 65% rate, achieved more completely. “Net positive” is accurate. It answers a different question than the one this section is asking.

In Sherman, Texas, three separate things get discussed together and are worth pulling apart. The city’s Post Oak Wastewater Treatment Plant is real, built, and operating — but its 16 million gallon per day permitted capacity is citywide infrastructure, serving “domestic, commercial and industrial wastewater” for the whole community, not a figure specific to either fab. [41] Its stated purpose is to treat wastewater to a standard “suitable and safe for reintroduction back into the natural stream environment” — it returns water to a creek, not to TI’s or GlobalWafers’ own production lines. Separately, GlobalWafers carries a water commitment NIST itself lists as a condition of its CHIPS Act award: recycling “at least 50% of the process water used onsite” within a year of completing its project, industry convention suggesting “process water” refers to the core production loop rather than secondary systems, though the government’s own award language does not spell out that distinction explicitly. [42] That commitment is qualified by “commercially reasonable efforts,” a standard that does not require the 50% figure actually be reached. TI has separately stated, in its own press materials rather than in NIST’s listed award conditions, that it is “endeavoring to achieve a 70% water reuse capability” across Sherman and its Lehi, Utah site — a company statement with no equivalent government-tracked commitment behind it. [43] TI’s own most recent disclosed company-wide water reuse figures — 27% in 2020, 29% in 2023 — show little movement over that period and sit far below the 70% target, with no demonstrated site achieving anything close to that rate publicly disclosed as of this writing. [44] No public reporting yet confirms whether either company’s Sherman-specific water target has been met.

Independent water-engineering firms serving the semiconductor sector, including Gradiant, advertise recycling capabilities of up to 99% for individual clients, at investments running into the hundreds of millions of dollars. [45] The company has not disclosed the client or site behind its highest-profile version of this claim, describing it only as “one of the world’s largest semiconductor manufacturers.” No public source connects this specific claim to any facility in the southern spine.

None of this means the companies making these claims are being dishonest. It means that, measured against the plainest and most verifiable standard available in this section — a company that stated a hard goal, described it honestly, pointed to demonstrated precedent, and made that statement in public to the community most likely to hold it accountable — most of the other claims in this buildout describe something narrower, softer, less precedented, or less verifiable than their framing suggests: a true result answering a different question, a target that has gone unconfirmed for years, a commitment qualified by its own contract language and unsupported by any comparable track record, or a capability whose relevance to this region has not been established. The water constraint documented at the start of this section has not been resolved by any of it. TSMC’s Arizona fabs have been drawing water and producing chips since late 2024. The plant designed to substantially close that gap will not exist until 2028 — and of everyone whose public claims this section has examined, TSMC is the only one who has said so plainly, in advance, without being asked.

CONCLUSION

Across this paper and the series it belongs to, the forces shaping the buildout of Fortress America are moving with a speed that is itself a finding. Section 1 documented more than $205 billion in announced semiconductor investment concentrated in Arizona alone, with Texas building a complementary corridor anchored by a $16.5 billion contractual link between Samsung’s Taylor fab and Tesla’s robotics and vehicle silicon. Section 2 traced the rare earth supply chain from Mountain Pass to Independence to a $400 million Department of Defense equity stake — the federal government’s largest and most structurally complete bet on any single company in this series, complete with a 10-year price floor and an offtake agreement guaranteeing the purchase of an entire facility’s output. Section 3 showed the semiconductor spine itself: fabs producing today in Arizona, fabs ramping in Texas, delays measured in months and quarters rather than cancellations. Section 4 turned to the constraint that ties all of it together — water — and found something more specific than a shortfall: a set of public claims that, examined individually, describe the state of that constraint less completely than their framing suggests.

What several billion-dollar projects already producing, and several more under active construction, demonstrate is that these are not speculative bets being made by capital hoping a policy environment holds steady. They are commitments structured to survive a change in administration, because the surrounding institutions — state legislatures funding water infrastructure before the fabs are finished, the Department of Defense taking equity rather than simply writing a grant, municipal governments rebuilding their utilities ahead of demand — are treating the underlying strategic problem as a fixed feature of the next decade, not a four-year policy preference.

The clearest evidence of that institutional permanence is the rare earth chain itself, and the timeline is worth stating plainly because it spans three administrations without reversing direction once. The Department of Defense’s first rare earth investment in MP Materials came in 2020, under the first Trump administration, a $10 million Defense Production Act award to a company struggling against Chinese retaliatory tariffs. [46] The Biden administration not only continued the relationship but expanded it — $45 million for Mountain Pass processing, more than $288 million to a second rare earth company, and a formal five-year mine-to-magnet investment strategy announced in 2024. [47] The second Trump administration then escalated that foundation into the $400 million equity stake, the price floor, and the offtake guarantee detailed in Section 2 — by far the largest single commitment in the chain’s six-year history. Three administrations, two parties, one direction, each handoff larger than the last.

This is the new mechanism this series has identified — equity rather than grants, ownership rather than subsidy — and it will be explored in full in the final paper of this series, which examines how the federal government is securing the raw material end of the supply chain across rare earths, copper, cobalt, and lithium, and what that ownership costs in dollars and control.

The water constraint is the one piece of this picture without a finish line, and it deserves a more precise closing than “underway but incomplete.” One company in this section set the standard by which the rest can fairly be judged. TSMC’s Arizona target — 90% recycling of the water that actually touches the wafer, not just the water that cools the building around it — was pointed explicitly at a demonstrated capability the company already runs in Taiwan, and was stated in public, in a community watching closely, with no regulator or contract requiring the company to say anything at all. Measured against that same bar, the rest of the buildout falls short in different and specific ways. Samsung’s reclamation partner announced a target three years ago and has answered no question about it since. Intel’s celebrated “net positive” result is real, and it is not the result this section is asking about. Sherman’s wastewater plant is a genuine, completed exception — but it treats water for the whole city, not for the two fabs it gets credited with serving, and it returns that water to a creek, not to either company’s production line. The one Sherman commitment actually aimed at a fab’s own process water, GlobalWafers’ federally tracked 50% target, is real and government-listed, and is also softened by language that does not require the number be reached. Texas Instruments’ matching 70% figure carries no such federal tracking at all, and the company’s own most recent disclosed results — 27% company-wide in 2020, 29% in 2023 — sit nowhere near it, with no comparable site TI has pointed to as precedent. And the water-engineering industry’s own flagship claim of 99% recycling, cited widely as evidence the private sector has this handled, cannot be confirmed to describe any facility in the region this paper is about.

None of this means the companies making these claims are lying. It means that TSMC’s Arizona fabs have been drawing water and producing chips since late 2024, that the plant designed to substantially close that gap will not exist until 2028, and that of everyone whose public statements this paper has examined on the hardest and most expensive part of the water problem, exactly one company has been consistently precise about what it has and has not yet done. Closing the mine-to-magnet-to-motor chain in Section 2 is a solved problem as of this writing. Closing the water gap in Section 4 is not — and the public record on how close anyone actually is to closing it is considerably thinner than it first appears.

REFERENCES

[1] Arizona Commerce Authority; Office of Governor Katie Hobbs, remarks at SEMICON West 2025, Phoenix, October 2025.
[2] Industrial Info Resources, Arizona Industrial Construction Pipeline, 2025.
[3] Office of Governor Greg Abbott, Texas Semiconductor Innovation Fund Grant Announcement, September 17, 2025.
[4] Samsung Electronics, Seoul Stock Exchange Regulatory Filing, July 28, 2025; Bloomberg, Reuters, CNN, July 28, 2025.
[5] Tom’s Hardware, “TSMC accelerates production timeline for new Arizona factory,” December 2025; Data Center Dynamics, “TSMC says Arizona fab is now ahead of schedule,” 2026.
[6] Data Center Watch (10a Labs), Q1 2026 Report; NBC News, “Study shows state and local opposition to new data centers is gaining steam,” 2026; Tom’s Hardware, “More than 75 data center build-outs worth $130 billion have been successfully blocked in the first three months of 2026,” 2026; Fox Business, “Chandler, Arizona, city council unanimously votes against AI data center,” December 2025.
[7] UltraFacility, “Semiconductor in numbers: Global fab construction timelines,” April 2026; Intel Newsroom, Ohio project status statements, 2025-2026.
[8] Rare Earth Exchanges, “How Rare Earth Elements Enable Modern Semiconductor Manufacturing Equipment,” January 2026.
[9] World Population Review / USGS rare earth production data, 2025; Optimusk, “Tesla Optimus Supply Chain,” 2026; 36kr English, April 2026.
[10] MP Materials Corp., Form 8-K, January 22, 2025, U.S. Securities and Exchange Commission; Fort Worth Report, “Fort Worth manufacturer begins producing rare earth magnets,” January 2025.
[11] MP Materials Corp., Form 8-K, July 2025, U.S. Securities and Exchange Commission; CNBC, “Pentagon to become largest shareholder in rare earth miner MP Materials,” July 2025.
[12] MP Materials Corp., Form 8-K, July 2025, U.S. Securities and Exchange Commission; The Defense Post, “Pentagon Takes Stake in US Rare Earth Company,” July 2025.
[13] MP Materials Corp., Form 8-K, July 2025, U.S. Securities and Exchange Commission; Select Committee on the Chinese Communist Party, U.S. House of Representatives, witness testimony, November 2025; C&EN, “US invests in rare earth firm MP Materials,” July 2025.
[14] Investing News Network, “Trilogy Metals Shares Rocket as US Government Takes Stake in Alaska Project,” October 2025; Axios, “US to take 10% stake in Trilogy Metals,” October 2025; Mayer Brown, “US Government Equity and Equity-Linked Investments in Critical Minerals,” April 2026.
[15] Internal Revenue Code Section 45X, Advanced Manufacturing Production Credit, as amended.
[16] U.S. Department of the Treasury, guidance on Section 45X critical minerals phasedown schedule, 2025-2026.
[17] MP Materials Corp., Form 8-K filings, U.S. Securities and Exchange Commission, 2025.
[18] Council on Foreign Relations, “The CHIPS Act: How U.S. Microchip Factories Could Reshape the Economy,” 2024; Semiconductor Industry Association, Chip Incentives & Investments data.
[19] TSMC Arizona, company facility status page, 2026; Tech Insider, “TSMC’s $165B Arizona GigaFab: Reshaping US Chips,” 2026.
[20] Tom’s Hardware, “TSMC brings its most advanced chipmaking node to the US yet,” December 2025.
[21] TSMC Arizona, company facility status page, 2026.
[22] BlackRidge Research, “TSMC Arizona Fab: USD 165 Billion Semiconductor Project,” 2026.
[23] Texas Instruments, “Texas Instruments begins production at its newest 300mm semiconductor manufacturing facility in Sherman, Texas,” December 17, 2025.
[24] EE Times, “Inside Texas Instruments’ New 300mm Fab in Sherman, Texas,” December 2025.
[25] TI.com, “Sherman, Texas: 300mm wafer fabs,” company site, 2026.
[26] Tom’s Hardware, “Samsung delays $44 billion Texas chip fab,” July 2025; Electronics360, “Report: Full production of Samsung’s Texas fab possibly delayed to 2027,” March 2026.
[27] TechTimes, “Samsung Taylor Fab Production Confirmed for 2027,” May 2026.
[28] MLQ.ai, “Samsung Delays Completion of $44 Billion Texas Chip Plant,” July 2025.
[29] CNBC, “Intel aims to find clients and catch TSMC with new chip fab in Arizona,” December 19, 2025; Tom’s Hardware, “Intel’s fab roadmap examined — Arizona, Ohio, Ireland, and the two deadlines deciding 14A process node,” June 2026.
[30] Ibid.
[31] U.S. Bureau of Reclamation, Colorado River shortage condition declarations, 2022-2026.
[32] Texas 2036, “Foundation for Economic Growth: Assessing Texas’ Water Infrastructure Needs,” 2024.
[33] World Economic Forum, “Semiconductor manufacturing and big tech’s water challenge,” 2024; CWR, “8 Things You Should Know About Water & Semiconductors”; IDE Tech, “Water Sustainability in the Semiconductor Industry”; Semiconductor Engineering, “How Semiconductor Fabs Use Water,” August 2025; SAMCO Technologies, “Industry Focus: Semiconductor industry trends and the importance of water resource management,” April 2025.
[34] EESI, “Data Centers and Water Consumption”; MOST Policy Initiative, “Data Center Water Use,” April 2026.
[35] TSMC, “TSMC Arizona and U.S. Department of Commerce Announce up to US$6.6 Billion in Proposed CHIPS Act Direct Funding,” pr.tsmc.com, April 2024; TSMC Arizona, “Sustainability” statement, tsmc.com, 2026; Data Center Dynamics, “TSMC breaks ground on water reclamation project in Phoenix, Arizona,” September 2025; Arizona Technology Council, “TSMC Breaks Ground on ‘Near-Zero’ Discharge Water Plant to Back Fabs,” September 2025.
[36] Ahwatukee.com / Times Media Group, “TSMC Arizona’s Water Reclamation Initiative Forwards Sustainability in Semiconductor Manufacturing,” September 2025, citing TSMC’s 2024 Sustainability Report.
[37] City of Phoenix, statement of Mayor Kate Gallego, August 2025; SemiWiki forum discussion, “In the city of Phoenix, TSMC Arizona will represent the world’s most advanced semiconductor technology in the United States,” January 2025; Senator Mark Kelly, “Kelly and Arizona Leaders Celebrate Finalized $6.6 Billion CHIPS and Science Act Award to TSMC,” press release, November 2024.
[38] Hoodline, “North Phoenix Scores Big As TSMC’s Second Chip Plant Hits Finish Line,” May 2026; Fortune, “Water-guzzling chipmaker TSMC and drought-plagued Arizona are an unlikely pair, but Phoenix says it has enough water,” April 2024.
[39] EPCOR USA, “EPCOR Tapped as Water Partner in Central Texas,” July 13, 2023; DNA Systems, “EPCOR – Sandow Water Project,” project completion record, October 2024; East Wilco Insider, “Looking for Enough Water,” September 1, 2023.
[40] ScaleBan Equipments, “How Semiconductor Industry is Tackling Wastewater Challenges”; industry reporting on Intel Ronler Acres end-of-pipe treatment mechanism and Intel’s 2022 net positive water milestone.
[41] City of Sherman, Texas, “Wastewater,” official city website; KXII, “Sherman unveils new multi million dollar wastewater treatment plant,” October 15, 2025.
[42] National Institute of Standards and Technology, “GlobalWafers (Texas),” nist.gov/chips, CHIPS award terms and environmental commitments.
[43] Texas Instruments, “Texas Instruments announces award agreement for CHIPS and Science Act funding,” ti.com, December 20, 2024; National Institute of Standards and Technology, “Texas Instruments (Utah),” nist.gov/chips.
[44] Texan By Nature, “Texas Instruments,” txn20.org, 2020 water reuse disclosure; Texas Instruments, 2024 Corporate Citizenship Report, cited via MarketScreener, June 2025 (2023 water reuse figures).
[45] The Register, “Mystery German chip fab sips on Gradiant’s ultrapure water,” January 22, 2024; Manufacturing Dive / ESG Dive, “Semiconductor industry faces water, sustainability challenges,” August 2025.
[46] Heatmap News, “The Pentagon’s Rare Earths Deal Is Making Former Biden Officials Jealous,” July 2025.
[47] The White House (Biden-Harris Administration Archives), “Fact Sheet: Biden-Harris Administration Takes Further Action to Strengthen and Secure Critical Mineral Supply Chains,” September 2024; Global Policy Watch, “Made in America: The Outlook for Critical Minerals,” October 2025; Vulcan Elements, “Money Finally Flowing to US Rare Earths Can’t Come Fast Enough,” August 2025.

FORTRESS AMERICA SERIES

Paper I: Fortress America — Hemispheric Consolidation, Infrastructure Cascades, and the Regional Transformation of the American Economy
Paper II: If It Can Work, It Can Fight — The Case for Domestic Humanoid Robotics Manufacturing
Paper III: Where Fortress America Lands — The Southern Spine and the Race to Build It
Paper IV: The Government Stake — Equity, Speed, and the Limits of Disclosure
Paper V: The Atom and the Chip
Paper V.5: Forthcoming

This paper represents independent analytical and systems research and is the third in the Blue Collar Analytics Fortress America series. Nothing contained herein constitutes financial advice, investment recommendations, or legal counsel. Readers should conduct their own due diligence and consult qualified professionals before making investment decisions.

Where Fortress America Lands Paper3 V1

34.9KB ∙ PDF file

Download

Download

Fortress America Series, Paper V

Author: Adam Wood | Publication: Blue Collar Analytics | Date: June 2026 | Version: 1.0

EXECUTIVE SUMMARY

The federal government and private capital are betting, at scale and at speed, that the AI and semiconductor buildout this series has already documented will need far more electricity than the grid currently supplies — and nuclear power is the answer they have chosen to fund. Whether any specific reactor in this paper ever powers any specific fab in Paper III is not the point. The point is that hundreds of billions of dollars are moving toward generation capacity sized to match an industrial demand curve that, by every account in this series, is already here. This paper maps where that money is going, how fast it is moving, and how much of it the public can actually see.

Two reactors have already reached criticality under a new, deregulated federal testing pathway, financed entirely with private capital. A third is expected to follow before a July 4, 2026 deadline. Separately, and on a far larger scale, the Department of Energy issued a $17.5 billion conditional loan commitment to Westinghouse Electric Company in June 2026, structured with the same equity-adjacent terms — a permanent profit share now, a convertible equity stake later — that Paper IV documented in the critical minerals buildout. The five sites that loan will fund are not yet public.

Wyoming and Utah anchor the geography, with Utah hosting the densest concentration of nuclear activity in the country. One project there, the long-dormant Blue Castle site, illustrates a real physical constraint this paper traces throughout: water. The project lost its water rights in 2021 amid a mix of financing, regulatory, and water-supply problems, and its 2026 revival arrived alongside a switch to an air-cooled reactor design and a dramatically faster federal licensing environment — two changes that happened together, and this paper does not claim to know which one mattered more. No comparable water plan, and no comparable capital commitment against the water constraint, has surfaced publicly for any other project in this buildout — a contrast this paper develops directly against what Paper III already documented the semiconductor industry doing in the same circumstances.

Finally, this paper distinguishes between civilian licensing speed, which is real, bipartisan in origin, and available to every developer, and a single military demonstration of speed by an actor operating outside that licensing structure entirely. Both are real. They are not the same thing, and this paper treats them as the two distinct cases they are.

SECTION 1 — THE BUILDOUT

The deregulatory and capacity-building groundwork for this buildout is bipartisan, and the clearest evidence of that sits at the beginning of the timeline. In July 2024, Congress passed the ADVANCE Act by a vote of 88-2 in the Senate and 393-13 in the House — among the most lopsided margins of any energy legislation in recent years — directing the Nuclear Regulatory Commission to cut review fees for advanced reactor applicants and build a faster, technology-neutral licensing framework.[1] President Biden signed it into law. Four months later, on November 12, 2024, his administration issued the U.S. Nuclear Energy Deployment Framework, setting a target of tripling the country’s nuclear capacity by 2050.[2]

On May 23, 2025, President Trump signed four executive orders that built directly on that foundation. EO 14300, “Ordering the Reform of the Nuclear Regulatory Commission,” set a target of quadrupling national nuclear capacity, from roughly 100 gigawatts today to 400 gigawatts by 2050, alongside the licensing timelines discussed in Section 3.[10] EO 14302, “Reinvigorating the Nuclear Industrial Base,” directed the Department of Energy’s Loan Programs Office — since renamed the Office of Energy Dominance Financing — to prioritize nuclear restarts, updates, and new construction.[3] EO 14299, “Deploying Advanced Nuclear Reactor Technologies for National Security,” directed the rapid deployment of reactors at DOE sites supporting AI infrastructure and ordered the Army to operate a reactor at a domestic base by September 30, 2028.[9] EO 14301, “Reforming Nuclear Reactor Testing at the Department of Energy,” created the Reactor Pilot Program and set the target this section traces: at least three advanced reactor designs reaching criticality outside the national laboratories by July 4, 2026.[6]

Two reactors have already met that target. Antares Nuclear Mark-0 reached zero-power criticality at Idaho National Laboratory on June 4, 2026, the first reactor cooled by something other than ordinary water to do so in the United States in more than forty years; Mark-0 uses liquid sodium to move heat out of the core.[4] Zero-power criticality means the reactor sustained a controlled chain reaction while generating no meaningful electricity — a deliberate first test of whether the design functions at all before any attempt to generate power. Valar Atomics’ Ward 250 followed on June 18, at the Utah San Rafael Energy Lab in Orangeville, Emery County, becoming the first reactor built entirely outside a national laboratory.[5] Ward 250 is helium-cooled, a third approach distinct from Mark-0’s sodium and from the ordinary water used in the AP1000 design discussed later in this section. Valar’s own technical materials describe the distinction this paper relies on throughout: “Cold ≠ Hot: Cold proves the physics. Hot proves the power.”[11]

Ward 250 was also the subject of the first aerial transport of a reactor in U.S. history. In February, three Air Force C-17s carried the disassembled, unfueled unit from March Air Reserve Base, California, to Hill Air Force Base, Utah — two states, not a cross-country journey — in an operation named Windlord, flown by the 62nd Airlift Wing, the only Air Force unit certified to routinely transport U.S. nuclear weapons.[12] The flight was carried out at federal expense, distinct from the private capital that funded the reactor’s development. A third project, Aalo Atomics, received approval to proceed in late June, with the Energy Secretary expressing confidence it would reach criticality before the deadline.[13] All three reactors were developed with private capital — reported figures include more than $130 million raised by Valar toward Ward 250 specifically, with broader company fundraising cited elsewhere as high as $489 million[14] — and required no federal money to reach the criticality milestone, only a faster federal permitting pathway than the Nuclear Regulatory Commission’s standard licensing process.

The federal financing story is a separate and much larger track, running in parallel. On June 23, 2026, the Department of Energy’s Office of Energy Dominance Financing — the renamed Loan Programs Office described above — issued a conditional loan commitment of up to $17.5 billion to Westinghouse Electric Company, intended to finance ten new AP1000 reactors, a large, roughly 1,100-megawatt pressurized water reactor design and the only large-scale advanced reactor currently licensed for commercial operation in the United States, at five sites nationwide, selected from seven utilities that have already signed letters of intent.[7] The Energy Secretary declined to name the candidate sites when asked directly. This single loan sits inside a far larger $80 billion framework, under which the federal government is entitled to keep 20 percent of profit distributions above the loan repayment threshold indefinitely, with the right to convert that claim into a 20 percent equity stake if Westinghouse goes public above a $30 billion valuation by January 2029.[7] “Conditional” matters here: the commitment depends on conditions not yet satisfied, including a final investment decision that has not yet been made, and is distinct from money that has actually been disbursed.

Geographically, the buildout is concentrated but not confined to two states, and those states have begun coordinating directly. Wyoming hosts TerraPower’s Natrium reactor near Kemmerer, a sodium-cooled fast reactor that received its construction permit in March 2026 — the first commercial reactor construction permit the NRC has issued in nearly a decade, and separately, the first ever issued for a non-light-water reactor, a category the agency had not approved in more than 40 years.[15] The plant broke ground on its nuclear systems in April 2026 and is targeted for completion in 2030-2031. Utah hosts the densest concentration of nuclear activity in the country: at least eight companies — TerraPower’s second site, Valar Atomics, Anfield Energy, Nusano, General Matter, Atlas Atomics, Curio, and the revived Blue Castle project near Green River — operating under Governor Spencer Cox’s Operation Gigawatt initiative, which state officials describe as a hundred-billion-dollar effort spanning reactor manufacturing, uranium mining, fuel enrichment, and waste handling.[16] Idaho, Wyoming, and Utah have formalized this coordination directly through a three-state agreement tying nuclear development to shared energy goals, with Idaho National Laboratory, where Antares went critical, serving as the shared technical anchor.[17] South Carolina, Ohio, and Tennessee each host at least one additional project: the long-idle V.C. Summer AP1000 site, a new Elementl Power plant in development, and Kairos Power’s Hermes reactor under construction at Oak Ridge, respectively.[18][19][20]

None of this geography overlaps with the southern spine this series mapped in Paper III. Arizona, Texas, and California host the semiconductor fabs and water infrastructure documented there; no reactor in this paper sits near any of them. That absence does not weaken this paper’s central claim. The bet being made is about total generation capacity matching total projected demand, not about any single reactor’s transmission lines reaching any single fab.

SECTION 2 — THE MECHANISM

Two different federal postures are operating inside this buildout at once, and they should not be mistaken for the same thing.

The first is deregulation. Executive Order 14301 created the Reactor Pilot Program, a new Department of Energy authorization pathway that lets private developers construct and operate full-scale test reactors outside the standard Nuclear Regulatory Commission licensing process, using the Department’s own authority under the Atomic Energy Act rather than a commercial NRC license.[6] Antares Nuclear, Valar Atomics, and Aalo Atomics moved through this pathway, developing their reactors entirely with private capital — a 45-day review standard and shared staff between DOE and NRC were the program’s actual contribution.[21] Operation Windlord is the one exception inside this category: the Air Force’s airlift of Valar’s unfueled reactor was flown at federal expense, using military aircraft, crews, and the 62nd Airlift Wing’s own logistics planning, a cost the private capital figures above do not include.[12]

The second posture is the one this series has already documented at length: direct federal financing through an equity-adjacent instrument. The Department of Energy’s Office of Energy Dominance Financing — the same agency structure Paper IV identified financing critical mineral companies — issued its $17.5 billion conditional loan commitment to Westinghouse Electric Company on June 23, 2026.[7] Once Westinghouse repays the loan and the projects begin generating profit beyond that threshold, the federal government keeps 20 percent of every dollar of that future profit, indefinitely — a permanent claim structured as a loan term rather than a stock purchase, but functioning the same way, with the right to convert into an outright 20 percent equity stake if Westinghouse pursues an IPO valued above $30 billion by January 2029.[7] The loan does not go directly to Westinghouse or to the participating utilities; it is structured through five newly created special purpose vehicles — separate legal entities, one per project, formed specifically to hold each project’s assets and debt apart from Westinghouse’s own balance sheet — jointly owned by Westinghouse and an as-yet-unnamed utility partner for each site, each required to commit $500 million in equity before any federal funds are released.[7]

This is the same architecture Paper IV traced through MP Materials, Vulcan Elements, and Trilogy Metals, applied at a larger dollar scale to a different sector. Westinghouse itself is majority owned by Brookfield, a private asset management firm, with Cameco, a uranium mining company, holding the remainder[22] — the same upstream-supply-chain pattern documented in Paper IV, where a critical-minerals company’s equity holders benefit directly from federal backing, now applies to the company building the reactors as well.

The distinction between the two postures separates what has already happened from what has only been promised. The Reactor Pilot Program’s reactors were developed with private capital and have already gone critical; that work is done and cannot be unwound. The Westinghouse program is conditional, dependent on a final investment decision not yet made, structured around sites and utility partners not yet named. One track is finished. The other has barely started.

SECTION 3 — THE SPEED

What changed in 2025 was the pace at which the ADVANCE Act’s groundwork got implemented. Executive Order 14300 set explicit numeric deadlines: 18 months for the NRC to issue a final decision on a new reactor license, and roughly 12 months for license renewals.[10] On March 25, 2026, the NRC Commissioners voted to finalize “Part 53,” the first entirely new commercial reactor licensing category since 1989, when the agency created Part 52. Part 53 does not replace the existing Part 50 and Part 52 pathways; it stands alongside them as a third, optional framework developers can choose instead.[23] The order behind this reform cites a long-running bottleneck in those older frameworks directly: between 1954 and 1978, the NRC authorized 133 reactors that were completed; since 1978, only two have entered commercial operation.[10] The rule was published in the Federal Register five days after the vote and took effect April 29, 2026.[31] The NRC also cut its hourly review fee for advanced reactor applicants from $318 to $148, a reduction of more than 50 percent, effective October 2025, implementing a fee structure the ADVANCE Act had already directed it to build the year before.[31]

Energy Secretary Chris Wright has supplied the historical scale that makes this reform meaningful: in the roughly 25 years following the first reactor to generate civilian electricity at Idaho National Laboratory in the 1950s, the United States permitted and began construction on more than 100 reactors before the industry, in his account, “ground to a halt” on regulatory grounds.[17] The current target — ten large reactors under the Westinghouse program, plus a handful of pilot microreactors — is a small fraction of that historical pace, even with review timelines compressed by years. This is the civilian speed this paper has referenced elsewhere: real, congressionally authorized, bipartisan in origin, and now backed by a specific regulatory framework and fee structure, applying to every developer in this buildout regardless of company or state.

Military speed is a different case, and an interesting one in its own right. Operation Windlord demonstrated what an actor operating outside this licensing structure entirely can accomplish, using its own aircraft, its own crews, and its own authority under the Atomic Energy Act. It says nothing about whether the civilian licensing system has gotten faster — that case is made on its own terms above — but it may say something about where the military intends to go next with reactor logistics, a thread worth watching rather than dismissing.

SECTION 4 — THE DISCLOSURE GAP

The pattern this series identified in Paper IV repeats here without modification. Disclosure of where the money and the reactors are actually going depends on what someone in government chooses to say, not on any standing requirement to say it.

The clearest example is the simplest one. Asked directly which five sites would receive the Westinghouse loan’s $17.5 billion, the Energy Secretary declined to answer, while confirming that seven utilities have already signed letters of intent tied to specific, identified sites.[7] The information exists. Seven companies know it. The public does not, and no law currently requires that it be disclosed before the final five are chosen.

The corporate structure compounds the gap rather than closing it. The loan flows through five special purpose vehicles that do not yet exist in public records, each jointly owned by Westinghouse and a utility partner the government has also declined to name.[7] A citizen attempting to trace $17.5 billion in federal commitments would need to identify entities that have not been created, owned in part by companies that have not been disclosed, before any tracing could begin.

This is not a uniquely nuclear problem. It is the same gap Paper IV documented in the critical-minerals buildout: public money moving through private corporate structures with no disclosure requirement attached to the structure itself, only to whichever party happens to be publicly traded and therefore bound by securities law. Westinghouse is privately held. Nothing requires Brookfield, Westinghouse, or any of the five forthcoming special purpose vehicles to disclose terms beyond what the Department of Energy chooses to publish.

The Reactor Pilot Program sits on the other side of this line, and the contrast is worth naming directly. Antares, Valar, and Aalo’s funding came from private investors who chose, on their own, to publicize their raises and milestones — Valar’s own social media announcement of Operation Windlord is one of the primary sources for this paper’s account of that event.[12] That openness was not required by any statute; it happened because the companies involved had a commercial incentive to be visible. The Westinghouse program has the opposite incentive structure: a federal loan large enough to draw scrutiny, structured in a way that delays the moment at which that scrutiny becomes possible.

SECTION 5 — THE WATER PROBLEM

Every reactor design in this paper solves the same basic engineering problem differently and arrives at the same downstream requirement regardless. Sodium, helium, or ordinary water can carry heat out of a reactor core. Once that heat is extracted, every design converts it to electricity the same way every thermal power plant has since the nineteenth century: by boiling water into steam and using that steam to turn a turbine. The technology changes how heat moves from the core to the water. It does not change the water-intensive steam cycle that follows.

Paper III measured this constraint precisely in the semiconductor industry, and the comparison is worth restating in full rather than borrowed as a conclusion: a single fab consumes roughly 10 million gallons of ultrapure water per day, against 1.5 to 5 million for a large data center — fabs accounting for roughly 40 percent of new water demand in an AI-era buildout against data centers’ 4 percent.[25] That finding mattered because it was measured, not estimated, and because the industry did not wait for the constraint to become a crisis before acting on it. TSMC built a reclamation system scaling to 36,000 cubic meters of recycled water per day.[25] Intel achieved zero-liquid-discharge status at its Oregon fab.[25] The city of Sherman, Texas rebuilt its water infrastructure with more than $400 million in capital specifically to support Texas Instruments and GlobalWafers before either company’s fab reached full production.[25]

No comparable response has surfaced publicly anywhere in this paper’s nuclear buildout, with one exception. Reporting on the Idaho National Laboratory milestone has independently noted the same gap this paper identifies: the criticality tests achieved so far prove the reactor core works, but leave the cooling systems — the part of the design most directly tied to water use — explicitly untested.[26] Blue Castle is the one project that has visibly changed course because of this constraint. Its original design, two large AP1000 reactors, required water rights the company secured in 2012 and then lost in 2021 amid a mix of financing trouble, regulatory delay, and an unresolved legal fight over the water itself.[8] Its 2026 revival did not restore that water position. It replaced it: the new design uses Holtec SMR-300 units with air-cooled condenser systems, chosen specifically to operate in arid conditions with far less water than the original plan required.[28] That redesign arrived in the same window as a dramatically faster federal licensing environment, and this paper does not have evidence to say which factor did more to make the revival possible — only that both happened together, and that water was real enough as a constraint to force a change in reactor technology regardless of which factor gets the credit.

No other project in this paper has made a comparable adjustment, or stated a comparable water plan at all. TerraPower’s Natrium design and the ten AP1000 reactors in the Westinghouse program use the same conventional steam-cycle generation that Blue Castle’s original design used before its water rights were lost, in some of the same water-stressed states Paper III already documented absorbing new demand from semiconductor fabrication — Utah’s Great Salt Lake and the Colorado River are both in measured long-term decline, independent of anything in this buildout.[29] The chip industry saw its water constraint coming and moved capital against it before its fabs went live. With one exception, this buildout has not yet shown that it has done the same.

CONCLUSION

This paper set out to determine whether the speed and structure of the federal government’s nuclear buildout matched the pattern this series identified in critical minerals. It does.

The institutional record runs continuously across the handoff between administrations: a Democratic Congress and president built the licensing reform and set the first capacity target; the current administration funded the financing instrument and accelerated the timeline. That continuity is the same constant this series has tracked since its first paper, and it requires no further argument here — Section 1 already made the case in full.

What the two postures inside this buildout share, and what separates them, is the same distinction Paper IV drew for critical minerals. A regulatory system getting faster benefits every developer equally. A single actor operating outside that system, as the military did in February, proves something narrower and more specific about that actor alone.

This paper’s title makes a claim about direction, not geography. The federal government and private capital are moving hundreds of billions of dollars toward generation capacity sized to match the AI and semiconductor demand this series has already documented — visibly, at a pace and through a structure the public can only partly see. Which five sites will receive $17.5 billion in federal loan commitments is not public information as of this writing. The water requirement of every reactor in this buildout beyond Blue Castle’s redesigned project is not addressed in any document this paper has reviewed. Both gaps sit in the same place Paper IV’s central finding sat: not in what has been built, which is real and dated and verifiable, but in what has only been promised, where the public’s ability to verify ends exactly where the government’s disclosure choices begin.

REFERENCES

[1] ADVANCE Act of 2024, Public Law 118-67, signed July 9, 2024; U.S. Senate Committee on Environment and Public Works, “Signed: Bipartisan ADVANCE Act to Boost Nuclear Energy Now Law,” July 9, 2024.

[2] U.S. Department of Energy, “U.S. Nuclear Energy Deployment Framework,” November 12, 2024.

[3] Executive Order 14302, “Reinvigorating the Nuclear Industrial Base,” May 23, 2025.

[4] U.S. Department of Energy, “Department of Energy Celebrates First Advanced Reactor Achieving Criticality” (Antares Nuclear, Mark-0), June 4, 2026.

[5] U.S. Department of Energy, “Department of Energy Celebrates Second Advanced Reactor Achieving Criticality” (Valar Atomics, Ward 250), June 18, 2026.

[6] Executive Order 14301, “Reforming Nuclear Reactor Testing at the Department of Energy,” May 23, 2025.

[7] U.S. Department of Energy, Office of Energy Dominance Financing, conditional loan commitment announcement, Westinghouse Electric Company, June 23, 2026.

[8] HEAL Utah; San Juan County Water Conservancy District records, Blue Castle Holdings water lease history, 2012-2021.

[9] Executive Order 14299, “Deploying Advanced Nuclear Reactor Technologies for National Security,” May 23, 2025.

[10] Executive Order 14300, “Ordering the Reform of the Nuclear Regulatory Commission,” May 23, 2025, 90 Fed. Reg. 22587 (May 29, 2025).

[11] Valar Atomics, “Project NOVA” technical materials, valaratomics.com.

[12] U.S. Air Force, 62nd Airlift Wing, Operation Windlord press materials, February 2026; Valar Atomics social media announcement, February 15, 2026.

[13] U.S. Department of Energy / Secretary Chris Wright, public statement on Aalo Atomics approval, late June 2026.

[14] Valar Atomics funding disclosures, various rounds, 2025-2026.

[15] U.S. Nuclear Regulatory Commission, press release No. 26-028, “NRC Issues First Commercial Reactor Construction Permit in Nearly a Decade,” March 4, 2026; U.S. Department of Energy, “NRC Issues Construction Permit for TerraPower’s Natrium Advanced Reactor.”

[16] State of Utah, Governor Spencer Cox, “Operation Gigawatt” initiative materials, 2024-2026.

[17] Cowboy State Daily, “Energy Secretary Tours 17 Labs And Urges Wyoming To Join ‘Nuclear Renaissance,’” December 9, 2025.

[18] Santee Cooper, V.C. Summer site status statements, 2025.

[19] Elementl Power, project announcement, Ohio, June 18, 2026.

[20] Kairos Power, Hermes reactor construction status, Oak Ridge, Tennessee.

[21] U.S. Department of Energy, Reactor Pilot Program fact sheet, June 2025.

[22] Brookfield Corporation, Westinghouse Electric Company ownership disclosures; Cameco Corporation, public filings.

[23] U.S. Nuclear Regulatory Commission, press release No. 26-035, “NRC Approves Final Rule for Advanced Reactor Licensing,” March 25, 2026; Perkins Coie, client alert, March 25, 2026.

[25] Blue Collar Analytics, “Where Fortress America Lands: The Southern Spine and the Race to Build It,” Fortress America Series, Paper III, Section 4.

[26] Idaho National Laboratory, reporting on Mark-0 zero-power criticality milestone, June 2026.

[28] Fulcrum Point Holdings / Blue Castle Holdings joint venture announcement, redesign to Holtec SMR-300, 2026.

[29] Circle of Blue, reporting on Great Salt Lake and Colorado River water decline, Utah.

[31] U.S. Nuclear Regulatory Commission, “Part 53” final rule, 91 Fed. Reg. 15696, March 30, 2026 (effective April 29, 2026); fee schedule update, effective October 2025

The Atom And The Chip Paper5 1

86KB ∙ PDF file

Download

Download

Equity, Speed, and the Limits of Disclosure

Author: Adam Wood | Publication: Blue Collar Analytics | Series: Fortress America, Paper IV | Date: June 2026 | Version: 1.0

EXECUTIVE SUMMARY

In a single five-month window beginning in March 2025, the federal government built the legal foundation for a fundamental shift in how it secures critical supply chains: from grants, tax credits, and loan guarantees toward direct equity ownership in private companies. Executive Order 14241 expanded Defense Production Act authorities and directed the Department of Defense and the U.S. International Development Finance Corporation (DFC) to create a joint mineral investment fund. The One Big Beautiful Bill Act, signed July 4, 2025, appropriated billions of dollars and gave the Department of Defense (DoD) explicit statutory authority to take equity positions in private companies. In the same legislative season, DFC’s own authority was reauthorized and dramatically expanded — its investment ceiling raised from $60 billion to $205 billion, its equity authority raised to a 40 percent ownership ceiling, and a statutory exclusion barring investment in the world’s wealthiest economies lifted for the first time since the agency’s creation, specifically for energy, critical minerals and rare earths, and information and communications technology.

Four federal entities — DoD’s Office of Strategic Capital, the Department of Energy’s newly renamed Energy Dominance Financing Program, the Department of Commerce’s CHIPS Program Office, and DFC — built or substantially empowered a dedicated equity-and-loan vehicle within roughly the same twelve-month period, then deployed those tools jointly against the same targets. The Pentagon’s $400 million equity stake in MP Materials, announced in July 2025, made DoD the company’s largest shareholder and remains the single most fully documented example. It is not an isolated case. At least a dozen domestic companies — among them Vulcan Elements, Trilogy Metals, Lithium Americas, and USA Rare Earth — have received some combination of federal equity, loans, price floors, or offtake guarantees since mid-2025, several from more than one agency simultaneously. The same toolkit now operates internationally as well, with documented federal capital commitments reaching at least eleven countries across four continents, including sovereign co-investment partnerships with the United Arab Emirates and Qatar.

This paper maps what has been disclosed. It does not claim to map what exists. Disclosure of these transactions follows no uniform federal transparency requirement; it depends almost entirely on whether the private party to a deal happens to be a publicly traded company subject to Securities and Exchange Commission (SEC) reporting rules, or whether the government and the company simply choose to announce. Where the counterparty is privately held, there is no legal mechanism requiring either party to disclose the size, terms, or even the existence of the arrangement. The dollar figures and deal counts that follow are therefore best understood as a documented floor, not a ceiling. What has not been voluntarily disclosed is, by definition, not counted here.

SECTION 1 — THE INSTRUMENTS AND THE AGENCIES

The legal foundation for this entire shift was built in a single five-month window. In March 2025, Executive Order 14241, “Immediate Measures to Increase American Mineral Production,” expanded the federal government’s authority under the Defense Production Act, reduced the approval and notification requirements that had previously slowed such actions, and directed DoD and DFC to establish a joint mineral investment fund. On July 4, 2025, the One Big Beautiful Bill Act became law. Section 20004 of that act appropriated $2 billion to expand the National Defense Stockpile, $5 billion to a new Industrial Base Fund with explicit statutory authority allowing DoD to take equity positions in private companies, $500 million to DoD’s Office of Strategic Capital for loans, guarantees, and technical assistance, and $1 billion for Defense Production Act financing through September 2027. In the same legislative season, Congress passed the DFC Modernization and Reauthorization Act of 2025 as part of the FY2026 National Defense Authorization Act, signed into law on December 18, 2025, raising DFC’s investment ceiling from $60 billion to $205 billion, increasing its equity authority to a 40 percent ownership ceiling, creating a new $5 billion equity revolving fund, and lifting, for the first time since DFC’s creation, the statutory exclusion that had barred the agency from investing in the world’s wealthiest economies — an exclusion now waived specifically for energy, critical minerals and rare earths, and information and communications technology.

What followed was not one agency acting alone but four building or empowering a dedicated internal vehicle for direct equity and loan deployment within roughly the same twelve-month period. DoD’s Office of Strategic Capital made its first direct loan — $150 million to MP Materials — in July 2025. The Department of Energy rebranded its Loan Programs Office as the Energy Dominance Financing Program in 2025–2026 and stood up a new Office of Critical Minerals and Energy Innovation. The Department of Commerce began exercising direct equity authority through its CHIPS Program Office, the same office originally built to fund semiconductor manufacturing incentives. DFC, already in existence since 2019, was simply handed a far larger ceiling and a wider mandate. Four different parts of the federal government did not coincidentally arrive at the same solution; they were directed to by the same six-month run of legislation and executive action.

The deals that followed show the toolkit in practice. The Department of Defense’s partnership with MP Materials, signed July 9–10 and closed July 11, 2025, is the clearest and most fully documented example: $400 million in newly issued Series A Cumulative Perpetual Convertible Preferred Stock, convertible at $30.03 per share; a separate $150 million unsecured loan to expand heavy rare earth separation capacity at Mountain Pass; a ten-year price floor of $110 per kilogram for neodymium-praseodymium products; and a ten-year offtake agreement guaranteeing the purchase of magnet output from a new production facility. The arrangement, together with an accompanying warrant, made DoD the company’s largest shareholder, at approximately 15 percent of outstanding shares on a fully diluted basis. Four months later, DoD’s Office of Strategic Capital and the Department of Commerce jointly backed Vulcan Elements, a North Carolina magnet manufacturer, with a $620 million loan and a $50 million equity stake respectively, as part of a package that also included $550 million in private capital. In October 2025, DoD acquired a 10 percent stake in Trilogy Metals, with warrants for an additional 7.5 percent, to advance copper and cobalt development in Alaska. That same month, the Department of Energy restructured a $2.2 billion loan to Lithium Americas, taking a 5 percent warrant-based equity stake in the company and its joint venture as a condition of the renegotiation.

These deals do not run through separate, walled-off channels. They overlap, often on the same company. Vulcan Elements drew simultaneously on DoD and the Department of Commerce. MP Materials drew on three distinct instruments — equity, a loan, and a price-floor guarantee — from DoD alone, later extended into an international joint venture with the same agency and Saudi Arabia’s Ma’aden. When four agencies build the same kind of tool within the same year and then deploy those tools jointly against the same targets, the overlap itself is a finding: this is not a single program responding to a single company’s need. It is a coordinated, multi-agency posture toward an entire sector, assembled in months, and it raises an immediate question the rest of this paper takes up directly — who is required to tell the public when it happens, and what happens when no one is?

SECTION 2 — THE TRANSPARENCY PROBLEM

The Department of Defense, the largest instrument in this entire toolkit, is notoriously bad at passing basic accounting practices, with formal outside independent audits still failing every year since they began in 2018. That is the baseline condition of the agency now taking equity stakes, setting decade-long price floors, and guaranteeing future purchases across a fast-growing list of private companies.

The legal and procedural avenues for outside oversight of these specific new actions have also narrowed rather than expanded. Executive Order 14241, issued in March 2025, reduced the approval and notification requirements that had previously applied to Defense Production Act actions and delegated to DFC certain authorities traditionally reserved for the president. The concern this raises is not speculative; it has been flagged from inside DFC itself. DFC’s own Inspector General has warned that the agency’s expanding domestic responsibilities risk distracting it from its statutory development mandate, and a prior Government Accountability Office review of DFC’s use of similar emergency lending authority during the COVID-19 pandemic found that the process was slowed by “more applications and more complex interagency involvement than DFC expected” — a finding that suggests the same agency now handling a far larger and faster-moving portfolio may not have the internal capacity to track it carefully. Separately, the Project On Government Oversight (POGO), a nonpartisan watchdog organization, has identified a specific legal ambiguity in how several of these deals have been justified more broadly: officials have described certain equity arrangements as voluntary agreements between consenting parties, a characterization that, if accurate, means no statute needs to be invoked to authorize them at all. An arrangement that requires no statutory authority to create also requires no statutory reporting obligation to disclose.

That gap matters most because of who is and is not required to say anything. Publicly traded companies must file a Form 8-K with the SEC within four business days of a material event, under Section 13(a) of the Securities Exchange Act of 1934 — the only reason any of these deals become visible to the public in specific dollar terms at all. Privately held companies carry no equivalent obligation: that statutory reporting duty attaches only once a company either lists its securities on an exchange or, under Section 12(g) of the same Act, exceeds 2,000 total shareholders (500 of whom are non-accredited investors) and $10 million in assets — thresholds a small, recently founded, venture-backed magnet manufacturer like Vulcan Elements does not come close to meeting. There is no form, no agency, and no statute that requires a private company below those thresholds to disclose that it has received a government loan, sold the government equity, or agreed to a price floor, unless the company or the government chooses to announce it.

And the deals themselves are not waiting around for any of that machinery to catch up. DoD’s $400 million equity stake in MP Materials, announced in July 2025, was negotiated in roughly eight weeks, beginning in late April 2025. Andrew Castaldo, J.P. Morgan’s co-head of Mid-Cap M&A and the deal’s lead advisor, called it “the most unique transaction I’ve been involved in throughout my career in M&A” — adding that, despite the deal’s novelty, it largely followed the standard diligence and structuring steps required of any complex M&A transaction. Four months later, DoD and the Department of Commerce backed Vulcan Elements, a privately held company with no SEC obligations at all, in a matter of weeks — a pace Pentagon officials described to journalists as a deliberate departure from the agency’s normal multi-month process, reportedly set in motion by a direct request from Peter Navarro, a senior White House official and friend of Donald Trump Jr., rather than the office’s standard review. Even after that deal was announced, basic terms went unanswered: when asked the size of the government’s stake, Vulcan referred the question to the Department of Commerce, which did not respond. It took a letter from Senators Elizabeth Warren, Richard Blumenthal, and Andy Kim to Secretary of Defense Pete Hegseth, dated January 22, 2026, to produce any further public accounting of the deal’s terms at all — and as of this writing, the questions the senators posed, including whether the loan followed a competitive process, remain unanswered beyond their February 5, 2026 deadline.

Both deals moved faster than the people executing them considered ordinary. The Vulcan case shows the only mechanisms that surfaced to catch a deal moving this fast were: a public company’s legal filing, a watchdog’s research, or a sitting senator’s letter, deployed only after the fact and only when someone happens to ask. None of those three is fast, and none is guaranteed to fire at all. A system of accountability built around mechanisms that move at the speed of ordinary bureaucracy is poorly matched to a set of transactions now closing faster than ordinary bureaucracy can track, especially when the company on the other side of the table has no legal obligation to say anything until someone outside the deal happens to ask the right question.

SECTION 3 — GOING GLOBAL

The toolkit documented in the first two sections of this paper does not stop at the U.S. border. Federal equity and loan commitments tied to critical minerals now reach at least eleven countries across four continents: Canada, Australia, Saudi Arabia, Mozambique, Brazil, South Africa, the Democratic Republic of Congo, Ukraine, Greenland, South Korea, and Jamaica. The mechanism behind that reach is the same one this paper has already traced domestically — equity, loans, price floors, and offtake guarantees — now extended outward through DFC and, increasingly, DoD acting abroad.

What makes the international footprint different from the domestic one is not the instrument. It is the partner. Domestically, the federal government takes a position in a private company. Internationally, in several of the largest and most recent deals, it takes a position alongside another government. In October 2025, DFC partnered with Orion Resource Partners, a private metals and materials investment firm, and ADQ, an Abu Dhabi sovereign wealth fund, to form the Orion Critical Mineral Consortium — an initial $1.8 billion commitment split evenly among the three partners, with a stated goal of growing to $5 billion. By January 2026, DFC had closed $600 million of its share. Qatar’s sovereign wealth fund occupies a similar position in a separate deal, holding $180 million in the critical minerals company TechMet alongside DFC’s own $105 million equity stake in the same firm. In both cases, the United States is not investing in a foreign asset. It is co-investing with a foreign government, as a financial peer, inside the same instrument.

That posture is recent enough to date precisely. The statutory authority that formally permits DFC to operate this way in wealthy economies — the exception carved out for energy, critical minerals and rare earths, and information and communications technology — was signed into law on December 18, 2025. The Orion Critical Mineral Consortium with the UAE’s ADQ was announced two months before that, in October 2025, and DFC’s portion of it closed the following January, under authority the agency held before its formal expansion took effect. The agency, in other words, was already operating at the edge of its mandate while Congress was still finishing the paperwork to expand it.

The Iran war supplied the clearest evidence yet of how quickly this toolkit can be redeployed when a new shock arrives. The conflict closed the Strait of Hormuz to shipping beginning March 4, 2026, cutting off a corridor that carries roughly half of the world’s seaborne sulfur trade — the feedstock for sulfuric acid, the reagent used to process copper, nickel, uranium, and rare earth ores. Mining companies producing those materials reported output reductions of 20 to 30 percent within weeks. The day before the United States and Israel launched their opening airstrikes on Iran on February 28, 2026, the Pentagon had already asked mining companies in its Defense Industrial Base Consortium for proposals to boost domestic supply of thirteen critical minerals. Days after the Strait closed, President Trump ordered DFC to insure maritime trade through the Gulf directly, and the agency announced a reinsurance facility covering losses of up to $20 billion on a rolling basis — a new, multibillion-dollar federal mandate, created by presidential order, inside a single week, layered on top of the equity and loan architecture already in place.

An interim memorandum between the United States and Iran was signed in mid-June 2026, calling for an end to hostilities and the reopening of the Strait without Iranian tolls for at least sixty days. As of this writing, that reopening is contested rather than settled: Iran’s military command has separately declared the Strait closed again, citing unrelated fighting in Lebanon, while U.S. Central Command maintains that commercial traffic continues to move and that the waterway has not, in practice, shut down. The legal and physical status of the Strait remains genuinely unresolved at the time of publication, which makes the underlying point of this section more durable, not less: the federal financial architecture built around this crisis — the insurance facility, the equity stakes, the supply chain mobilization — was constructed for a shock that has not actually concluded, and that architecture does not appear designed to unwind quickly even if a ceasefire eventually holds.

The pattern is the same one this paper has documented domestically. A shock arrives, and the response is not retreat or negotiation alone. It is an expansion of direct federal financial participation, assembled and deployed faster than the institutions involved consider normal, now extending to cover not just American companies but the shipping lanes and sovereign partners that connect them to the rest of the world.

SECTION 4 — THE SNOWBALL

The pattern across all three preceding sections has a shape, and the shape is acceleration. It did not begin large. DoD’s first rare earth investment in MP Materials, in 2020, was a $9.6 million Defense Production Act Title III award. The Biden administration continued the relationship at a similarly modest scale — tens of millions of dollars across individual awards, including $45 million for Mountain Pass processing and $35 million toward MP’s heavy rare earth separation project in 2022, plus a formal five-year mine-to-magnet investment strategy announced in 2024 rather than a financial instrument of comparable size to what followed. For most of this decade, the toolkit this paper documents existed in outline only: a recognition that China’s dominance of critical minerals was a problem, addressed through grants and loan guarantees of a size that drew little public attention because there was little, in dollar terms, to attract it.

That changed in 2025, and it changed in stages that compound rather than simply add. DoD’s $400 million equity stake in MP Materials in July was roughly forty times the size of its first investment five years earlier. The legislative foundation behind it — the One Big Beautiful Bill Act, the DFC reauthorization, Executive Order 14241 — arrived within the same five-month window, each piece expanding what the next deal could do. By the time Vulcan Elements closed its loan in November, the pace had compressed from months to weeks. By February 2026, DFC reported to a State Department-hosted gathering of senior officials from 54 countries — the 2026 Critical Minerals Ministerial — that the U.S. government had mobilized more than $30 billion in letters of interest, investments, loans, and other support over the preceding six months alone. None of this was one decision. It was the same decision, made repeatedly, at a larger scale and a faster pace each time it was made again.

Then the snowball stopped rolling and started falling. A snowball that rolls down a hill gathers mass gradually, picking up what is already in its path. What happened when the Iran war closed the Strait of Hormuz in March 2026 was not gradual. It was a shock that hit a structure already in motion and threw it forward at a different velocity entirely. Within a single week, a sitting president ordered a federal development agency to underwrite up to $20 billion in maritime insurance it had never been asked to provide before, for a purpose — protecting the shipping lanes that move the sulfur needed to process rare earths and other critical minerals — that did not exist as a stated mission until the war made it one. The Pentagon had already been moving in this direction; the war did not start the acceleration. It picked up something already rolling and threw it down the back half of the hill. As of this writing, the Strait’s reopening remains contested rather than confirmed, which means the federal apparatus built around that shock has not yet been tested by an actual, settled peace — only by a temporary truce that one side has already declared broken at least once.

But every figure in this section describes only the visible part of the snowball. Section 2 of this paper established that disclosure of these transactions is voluntary, uneven, and dependent on whether a counterparty happens to be publicly traded or whether someone chooses to announce. That limitation does not pause for this section’s argument. The $30 billion figure, the eleven-plus countries, the compressed timelines — all of it is a measure of what has surfaced, not a measure of the toolkit’s true size or true speed. If the pattern documented elsewhere in this paper holds, the gap between what is visible and what exists does not shrink as the activity accelerates. It is reasonable to expect the opposite: a faster-moving, less scrutinized process is, if anything, more likely to outrun disclosure than a slower one was. Whatever rate of change this section can document is therefore a floor on the true rate, not a ceiling. The snowball could be larger and moving faster than any figure in this paper is able to show.

CONCLUSION

This paper has argued that the federal government is moving with increasing speed to secure equity stakes and ensure the flow of the raw rare earth minerals necessary for advanced manufacturing, both at home and abroad. The pace and the dollar figures are now part of the public record: a $9.6 million Defense Production Act award in 2020 grew into a $400 million equity stake in 2025, the legal authority behind it built in a single five-month legislative window, and by early 2026 the government itself was reporting more than $30 billion in commitments mobilized in six months alone, reaching at least eleven countries on four continents. The mechanism for the public to learn any of this has not kept pace with how quickly it is happening. Disclosure depends on whether a company happens to be publicly traded, or on whether the government and its private counterparties simply choose to announce. What this paper has presented is therefore a floor, not a ceiling — the documented minimum of an activity whose true scope cannot currently be measured from the outside.

The entire series has built toward this point. Paper I traced the decades-long, multi-administration pattern of hemispheric consolidation. Paper II identified the technology, humanoid robotics, that makes domestic manufacturing an urgent defense question rather than an economic preference. Paper III mapped where the physical buildout is landing, in the fabs and processing facilities of the southern spine. This paper has shown how the government is paying for all of it: not primarily through grants and tax credits, but through direct ownership, in deals that move faster than the public can track and disclose less than a public company would be required to.

Recent commentary has rebranded hemispheric consolidation as one president’s personal project, the Don-roe Doctrine. The record this paper has assembled argues against that framing, but it does not argue from a position of full visibility. The cloak does not open on its own. It opens when the law requires it, as it does for the handful of publicly traded companies in this story, or when the government opens it for its own reasons, or, rarely, when a senator’s letter forces a partial look at what is underneath. Most of the time, by this paper’s own accounting, it stays closed. An institution that has failed eight consecutive financial audits and spent decades unable to account for trillions of dollars clearly knows how to move slowly when it wants to. It has chosen not to here. A boat moving fast leaves more wake than one moving slowly, whether or not the captain wants to be seen, and the cloak of empire slips the same way — not because anyone has chosen to reveal what is underneath, but because nothing moving this fast stays fully covered. The senators’ letter that followed the Vulcan deal, the watchdog testimony, the public company filings this paper has relied on throughout: all of it is wake. What is known is what the law required, what someone wanted credit for, or what speed left visible by accident. What remains hidden is not measured here, and there is no evidence that the rate at which the government is moving is making the covering any thinner.

Whether Fortress America is complete by the end of the decade is not something this paper can answer, and it would be dishonest to claim otherwise. The transparency gap this paper has documented means that no one outside the government currently has the information required to measure the buildout’s true size, let alone forecast its finish line. What can be said, on the evidence assembled across four papers, is narrower and more certain: every measurable point on this curve is larger and faster than the point before it, the most recent acceleration was triggered by a war whose own ending remains contested as this paper goes to press, and nothing in the record suggests that curve is bending the other way.

REFERENCES

[1] One Big Beautiful Bill Act, Public Law 119-21, Section 20004, signed July 4, 2025.

[2] Executive Order 14241, “Immediate Measures to Increase American Mineral Production,” 90 Fed. Reg. 13673, signed March 20, 2025, published March 25, 2025, Federal Register.

[3] One Big Beautiful Bill Act, Public Law 119-21, Section 20004; Brownstein Hyatt Farber Schreck, “U.S. Expands Critical Minerals Financing and Bilateral Partnerships Under Trump,” November 10, 2025.

[4] FY2026 National Defense Authorization Act, Public Law 119-60, Section 8701, “DFC Modernization and Reauthorization Act of 2025,” signed December 18, 2025; U.S. International Development Finance Corporation, “DFC Secures Expanded Authorities with FY26 NDAA Signed into Law,” press release, December 18, 2025; Haynes Boone, “Warfighting Procurement Authorization: 2026 NDAA Poised to Enact Sweeping Changes to Defense Contracting,” December 12, 2025; Mayer Brown, “US Government Equity and Equity-Linked Investments in Critical Minerals,” April 15, 2026.

[5] FTI Consulting, “The New U.S. Government Critical Minerals Playbook,” February 13, 2026; Department of Energy, “Office of Energy Dominance Financing,” energy.gov.

[6] MP Materials Corp., Form 8-K, July 10, 2025, U.S. Securities and Exchange Commission; CNBC, “Pentagon to become largest shareholder in rare earth miner MP Materials,” July 10, 2025.

[7] Mayer Brown, “US Government Equity and Equity-Linked Investments in Critical Minerals,” April 15, 2026; InvestorNews, “Follow the Money: The U.S. Government Funding Hit List for Critical Minerals Companies (2023–2026),” February 12, 2026.

[8] U.S. Department of State, “2026 Critical Minerals Ministerial,” Office of the Spokesperson, February 2026; DFC, “DFC Highlights Landmark Critical Minerals Investments to Strengthen U.S. National Security During State Department Ministerial,” press release, February 4, 2026.

[9] U.S. Securities and Exchange Commission, Form 8-K filing requirements, 17 CFR 249.308; Investopedia, “Form 8-K: Definition, What It Tells You, Filing Requirements, Deadline.”

[10] Brownstein Hyatt Farber Schreck, “U.S. Expands Critical Minerals Financing and Bilateral Partnerships Under Trump,” November 10, 2025.

[11] Department of Energy, “Office of Energy Dominance Financing,” energy.gov; McAllister & Quinn, “DOE Critical Minerals Initiatives: New Offices, Major Funding Programs, and Pending Battery Materials FOA,” February 5, 2026.

[12] FTI Consulting, “The New U.S. Government Critical Minerals Playbook,” February 13, 2026.

[13] MP Materials Corp., Form 8-K, filed July 10, 2025, U.S. Securities and Exchange Commission (Subscription Agreement, Transaction Agreement, Certificate of Designations); StockTitan, “MP Secures $550M DoD Funding & 10-Year Offtake, Dilution ~15%,” July 10, 2025; J.P. Morgan, “J.P. Morgan Leads Groundbreaking Rare Earth Magnets Deal,” insights, October 2025.

[14] U.S. Department of Defense, “Office of Strategic Capital Agrees to Joint $700M Conditional Loan Commitment with Vulcan Elements and ReElement Technologies,” press release, November 21, 2025; U.S. Department of State, “2026 Critical Minerals Ministerial,” February 2026.

[15] Investing News Network, “Trilogy Metals Shares Rocket as US Government Takes Stake in Alaska Project,” October 2025; Axios, “US to take 10% stake in Trilogy Metals,” October 2025.

[16] FTI Consulting, “The New U.S. Government Critical Minerals Playbook,” February 13, 2026 (citing DOE Loan Programs Office, Thacker Pass restructuring, October 2025).

[17] Reuters, “MP Materials to Build Saudi Rare Earths Refinery with Pentagon, Maaden,” U.S. News & World Report, November 19, 2025.

[18] U.S. News & World Report, “Pentagon Says It Fails Eighth Audit, Targets 2028 to Pass,” December 19, 2025; Project On Government Oversight, “Fact Sheet: Auditing the Department of Defense,” October 6, 2025.

[19] Congressional Research Service, “INSIGHTi: Executive Order 14241 and DFC Authorities” (citing DFC Office of Inspector General statements and a Government Accountability Office review of DFC’s Title III lending under Executive Order 13922, May 2020), congress.gov; Project On Government Oversight, “Fact Sheet: Auditing the Department of Defense,” October 6, 2025.

[20] U.S. Securities and Exchange Commission, Form 8-K filing requirements, 17 CFR 249.308.

[21] Securities Exchange Act of 1934, Section 12(g) and Section 13(a), 15 U.S.C. § 78m and § 78l; U.S. Securities and Exchange Commission, Investor.gov, “The Laws That Govern the Securities Industry.”

[22] J.P. Morgan, “Rare earths, real impact: Inside the MP Materials deal,” What’s the Deal podcast, recorded August 25, 2025, jpmorgan.com/insights/podcast-hub.

[23] ProPublica, reporting on Vulcan Elements loan approval process and Peter Navarro’s involvement, cited in: Cato Institute, “White House Intervention Bolsters Trump Jr–Connected Rare Earths Firm,” Cato at Liberty Blog, May 2026; Wikipedia, “Vulcan Elements,” accessed June 2026.

[24] The Wire China, “The Magnet Makers,” March 20, 2026.

[25] Senators Elizabeth Warren, Richard Blumenthal, and Andy Kim, letter to Secretary of Defense Pete Hegseth, January 22, 2026, warren.senate.gov; U.S. Senator Elizabeth Warren, “Warren, Blumenthal, Kim Sound Alarm on Potential for Donald Trump Jr.-Linked Companies to Profit Off Department of Defense Contract Awards, Loans,” press release, January 23, 2026.

[26] InvestorNews, “Follow the Money: The U.S. Government Funding Hit List for Critical Minerals Companies (2023–2026),” February 12, 2026; CSIS, “G7 Cooperation to De-Risk Minerals Investments in the Global South,” May 16, 2025; Mercuria, “Gécamines and Mercuria Launch Copper-Cobalt Joint Venture in DRC with Backing from U.S. International Development Finance Corporation (DFC),” December 2025.

[27] DFC, “DFC Joins $1.8 Billion Consortium to Secure Critical Mineral Supply Chains and Bolster U.S. Economic Growth and Security,” press release, October 23, 2025; Mining.com, “US, Orion partner on $5B critical minerals fund,” October 23, 2025.

[28] DFC, “DFC Highlights Landmark Critical Minerals Investments to Strengthen U.S. National Security During State Department Ministerial,” press release, February 4, 2026.

[29] CSIS, “How to Reform the DFC to Meet U.S. Critical Minerals Security Needs,” January 31, 2025; CSIS, “G7 Cooperation to De-Risk Minerals Investments in the Global South,” May 16, 2025.

[30] GIS Reports, “Iran war squeezes critical supplies, alliances,” June 2026; The Soufan Center, “The Iran War: A Crisis for the Defense Industrial Base Now Too,” March 25, 2026.

[31] GIS Reports, “Iran war squeezes critical supplies, alliances,” June 2026.

[32] The National, “Iran war exposes fragilities in global critical minerals supply chain,” March 17, 2026; The Soufan Center, “The Iran War: A Crisis for the Defense Industrial Base Now Too,” March 25, 2026; Wikipedia, “2026 Strait of Hormuz crisis,” accessed June 2026.

[33] Congressional Research Service, “Iran Conflict and the Strait of Hormuz: Impacts on Oil, Gas, and Other Commodities,” congress.gov, R45281; statement of President Donald Trump, March 3, 2026, cited in CRS report; CNBC, “Iran reportedly closes Strait of Hormuz again, casting shadow over nuclear talks,” June 20, 2026; Newsweek, “Iran Army Declares Strait of Hormuz ‘Closed’ Over MOU ‘Breach,’” June 20, 2026; Euronews, “Tehran says Strait of Hormuz remains open after confusion,” June 19, 2026.

[34] Heatmap News, “The Pentagon’s Rare Earths Deal Is Making Former Biden Officials Jealous,” July 2025.

[35] The White House (Biden-Harris Administration Archives), “Fact Sheet: Biden-Harris Administration Takes Further Action to Strengthen and Secure Critical Mineral Supply Chains,” September 2024; Global Policy Watch, “Made in America: The Outlook for Critical Minerals,” October 2025.

[36] U.S. Department of State, “2026 Critical Minerals Ministerial,” Office of the Spokesperson, February 2026.

FORTRESS AMERICA SERIES
Paper I: Fortress America — Hemispheric Consolidation, Infrastructure Cascades, and the Regional Transformation of the American Economy
Paper II: If It Can Work, It Can Fight — The Case for Domestic Humanoid Robotics Manufacturing
Paper III: Where Fortress America Lands — The Southern Spine and the Race to Build It
Paper IV: The Government Stake — Equity, Speed, and the Limits of Disclosure

DISCLAIMER: This paper represents independent analytical and systems research and is the fourth and final paper in the Blue Collar Analytics Fortress America series. Nothing contained herein constitutes financial advice, investment recommendations, or legal counsel. Readers should conduct their own due diligence and consult qualified professionals before making investment decisions.

The Government Stake Paper4 1

97.9KB ∙ PDF file

Download

Download

The Southern Spine and the Race to Build It

Author: Adam Wood | Publication: Blue Collar Analytics | Series: Fortress America, Paper III | Date: June 2026 | Version: 1.0

EXECUTIVE SUMMARY

The Fortress America framework established in Papers I and II predicted that hemispheric consolidation would produce a domestic infrastructure buildout traceable in real time through capital flows, regulatory actions, and physical deployment. Paper III maps where that buildout is physically landing — a California-Texas-Arizona southern spine emerging as the industrial core of the American economy for the next generation. The rare earth supply chain from mine to magnet to motor is now domestically traceable for the first time. More than a dozen semiconductor fabs are at some stage of production or active construction across this corridor. The binding constraint on the entire buildout is water in an arid region, and capital and engineering are already being deployed against it through closed-loop reclamation and dedicated municipal infrastructure built specifically to serve the new industrial demand.

SECTION 1 — THE SOUTHERN SPINE

Arizona is the most concentrated single-state semiconductor investment in American history. Since 2020, the Arizona Commerce Authority has documented more than 60 semiconductor expansions representing more than $205 billion in investment — a figure Governor Katie Hobbs cited directly at SEMICON West 2025, the semiconductor industry’s own annual trade conference, held in Phoenix for the first time. By early 2025, Industrial Info Resources tracked nearly $50 billion in active industrial construction across the state, with semiconductor fabs, data centers, power plants, and mining operations leading the pipeline.

Texas is building a different but complementary concentration — multiple companies across a contiguous corridor rather than one dominant anchor. Samsung’s Taylor fab on 1,200 acres is the largest foreign direct investment in Texas on record, targeting operational status at the end of 2026 with 1,500 permanent employees and production of 2-nanometer leading-edge chips. In July 2025, Samsung filed a formal regulatory disclosure on the Seoul Stock Exchange confirming a $16.5 billion semiconductor supply agreement running through December 31, 2033; Tesla CEO Elon Musk confirmed the same day that Samsung’s Taylor facility would produce Tesla’s next-generation AI6 chip, used across Tesla’s AI products including autonomous vehicles and the Optimus humanoid robot. The robotics manufacturer and the semiconductor fab are already contractually linked in the same state through the end of the decade.

Both states are absorbing pressure from different directions, and the pattern reveals which part of the buildout is durable. Semiconductor fabs faced real delays in 2023-2024 — Taiwan Semiconductor Manufacturing Company’s (TSMC) first Arizona fab slipped roughly a year due to skilled-labor shortages, and its second fab moved from a 2026 target to 2027-2028 — but both are now active, with TSMC reporting pilot production ahead of its most recent revised schedule as of April 2026. Data centers in the same state have faced a different kind of setback entirely: data center opposition nationwide blocked or delayed at least 75 projects worth roughly $130 billion in the first quarter of 2026 alone — a single-quarter total matching all of 2025 combined, and more than double the $64 billion blocked cumulatively between 2023 and early 2025. Arizona has been part of this wave throughout, where Chandler’s city council unanimously rejected a proposed AI data center in December 2025 and a $14 billion West Valley project was withdrawn after the city declined to approve rezoning. The fabs slipped on labor and equipment timelines and recovered. The data centers are being stopped by the communities they are built in.

The southern spine is the operational core of the domestic semiconductor buildout today, with Arizona in high-volume production and Texas ramping to production in 2026-2027. New York and Ohio represent a second wave of the same institutional pattern — committed capital, steel in the ground, state and federal funding locked in — coming online in the 2030-2032 timeframe. The fab-by-fab production status confirming this geography as the industrial core of Fortress America is detailed in Section 3.

SECTION 2 — MINE TO MAGNET TO MOTOR: THE DOMESTIC RARE EARTH SUPPLY CHAIN

Semiconductor capacity is the precondition for advanced manufacturing; without the chip, a robot has no intelligence to actuate in the first place. But intelligence without actuation is just as inert — a robot that can think but cannot move is a research demonstration, not a defense asset. The rare earth supply chain is the second non-negotiable bottleneck, and for the first time, it is now traceable end to end on American soil.

Rare earth elements are not a single-application input. They are foundational across the entire advanced manufacturing stack: permanent magnets containing neodymium, praseodymium, dysprosium, terbium, and samarium power the precision actuators inside semiconductor fabrication equipment itself, optical crystals doped with yttrium and lanthanum enable the lasers used in advanced lithography, and the same magnet chemistry that moves a wafer with micron-level precision inside a fab is what moves a robotic joint with the torque and control a humanoid platform requires. The supply chain risk is not confined to one product category. China accounts for over 90% of global rare earth production broadly, and approximately 85 to 90% of neodymium-iron-boron (NdFeB) magnet production specifically — concentration that touches the fabs documented in Section 1 as much as the robots documented in Paper II.

The chain that resolves this dependency, for the first time, exists entirely within U.S. borders. Mountain Pass, California, operated by MP Materials, is the only operational rare earth mining and processing facility of scale in the United States — the extraction point. Independence, MP Materials’ magnet manufacturing facility in Fort Worth, Texas, is already producing at an initial 1,000 metric ton annual capacity and expanding — the conversion point, where processed rare earth oxide becomes finished magnet. A robot built in Texas with American rare earth magnets processed from California ore is a fundamentally different strategic asset than one assembled with components Beijing can restrict on short notice.

The federal government’s commitment to this chain is structured with more depth than a typical subsidy. In July 2025, the U.S. Department of Defense (DOD) entered a multi billion-dollar agreement with MP Materials that made DOD the company’s largest shareholder, acquiring approximately 15% of outstanding shares through a $400 million purchase of newly created preferred stock, convertible at $30.03 per share. The package extends well beyond equity: a $150 million loan to expand heavy rare earth separation capacity at Mountain Pass; a 10-year price floor of $110 per kilogram for MP’s neodymium-praseodymium (NdPr) products, insulating the company from Chinese price manipulation; and a 10-year off take agreement under which DOD guarantees that 100% of magnet output from a new “10X” production facility — targeting 10,000 metric tons of annual capacity by 2028 — will be purchased by defense and commercial customers. MP secured an additional $1 billion in commercial financing from JP Morgan Chase and Goldman Sachs on the strength of the federal commitment. As a binding condition of the agreement, MP Materials ceased all sales of rare earth products to China in July 2025 — a cessation already reflected in the company’s own reported revenue — and committed not to renew its existing off take agreement with China’s Shenghe Resources at its January 2026 expiration, ending the one remaining contractual link between America’s primary rare earth producer and a Chinese state-affiliated buyer.

This is not an isolated transaction. It is the first and most fully developed example of a repeating federal instrument: the government taking direct equity positions in companies that control the extraction and early-stage processing of strategic minerals, rather than in the factories that turn those minerals into finished goods. The same Department of Defense office that structured the MP Materials deal has since taken a 10% stake, with warrants for an additional 7.5%, in Trilogy Metals to advance copper and cobalt development in Alaska, and a separate equity position in Lithium Americas. What this pattern signals about how the government is choosing to secure the upstream end of the supply chain, rather than only the downstream manufacturers, will be explored in full in the final paper of this series.

The one legislative risk to this chain is the Section 45X Advanced Manufacturing Production Credit, which subsidizes domestic production of critical minerals and components. For most 45X-eligible products, the credit phases down starting in 2030 and expires entirely in 2033. Critical minerals, however, sit on a separate track within the same credit: the phaseout for critical minerals does not begin until 2031, with full expiration in 2033 — and unlike solar and wind components, critical mineral production was carved out from the general sunset schedule precisely because of its strategic designation. Even within that more favorable schedule, non-renewal in 2033 would still represent a real long-term risk to the economics of expanding domestic magnet production at the pace the DOD off take agreement assumes.

But the DOD off take agreement itself is what makes that risk politically difficult to realize. The Pentagon is now MP Materials’ largest shareholder, has committed to a 10-year price floor, and has guaranteed purchase of 100% of a facility’s output through approximately 2035 — two years past the 45X expiration date. A Congress that allowed the critical minerals credit to lapse without renewal would be undermining an investment the Department of Defense itself structured, sized, and continues to hold equity in. Attacking the credit’s renewal becomes, in practice, attacking the defense department’s own balance sheet and its own stated rare earth independence strategy — a substantially higher political bar than allowing a generic manufacturing subsidy to expire on schedule.

The mine-to-magnet-to-motor chain described here is not a projection. Mountain Pass is producing today. Independence is producing today. The DOD’s equity, loan, and off take commitments are signed, public, and filed with the U.S. Securities and Exchange Commission (SEC). What remains is scale — the 10X facility in Northlake reaching its 10,000-metric-ton target by 2028, and the broader expansion of domestic magnet capacity beyond MP Materials alone, an effort already extending into Wyoming and Alaska on separate tracks. The chain that supplies the humanoid robotics manufacturing imperative identified in Paper II now exists, end to end, without crossing Chinese territory at any point.

SECTION 3 — THE SEMICONDUCTOR SPINE

The strategic vulnerability this buildout addresses is stark and well documented. The United States declined from producing roughly 37 to 40% of global semiconductors in 1990 to approximately 10 to 12% today, and as of the passage of the CHIPS and Science Act in 2022, none of the world’s most advanced chips were manufactured domestically. The correction underway since is not a slow policy response — it is one of the fastest industrial buildouts in American history, and its pace is best understood by naming what is actually under construction, what is already producing, and what remains delayed.

TSMC’s Arizona campus is the clearest evidence of acceleration. Fab 21’s first facility entered high-volume 4-nanometer production in Q4 2024 and is now manufacturing chips for Apple and Nvidia — the first time TSMC has produced cutting-edge AI silicon outside Taiwan. The second fab’s construction was completed in 2025, with equipment installation beginning in the third quarter of 2026 and 3-nanometer production targeted for 2027 — a full year ahead of the original schedule, according to TSMC’s own CEO. A third fab broke ground in April 2025, targeting 2-nanometer and A16 process technology by the end of the decade. Total committed investment has grown from an initial $12 billion in 2020 to $165 billion across six planned fabs, with TSMC describing long-term framework expansion reaching approximately $465 billion — the largest single foreign direct investment in American history.

Texas Instruments offers the clearest example of speed. Construction began on its Sherman, Texas site in mid-2022; the first fab, SM1, began production on December 17, 2025 — three and a half years from groundbreaking to output. The facility produces foundational analog and embedded processing chips, the components used in nearly every electronic device, including industrial robotics and automotive systems. SM2’s exterior shell is already complete, with cleanroom installation underway in 2026. The Sherman campus is part of a $60 billion investment across seven planned fabs in Texas and Utah.

Samsung’s Taylor, Texas fab presents a more complicated picture, and an honest accounting requires saying so. Originally targeted for 2024 production, the facility was delayed twice — once to upgrade its process technology from 4-nanometer to the more advanced 2-nanometer node, and again reportedly due to insufficient near-term customer demand. As of May 2026, Samsung’s own foundry leadership confirmed that customer production, including for Tesla, is scheduled to begin in 2027, with the facility’s third-generation 2-nanometer process now in installation. The delay is a demand and technology-upgrade story, not a capital withdrawal — Samsung’s total investment in the site grew from $17 billion to $44 billion over the same period, and the company received $4.75 billion in direct CHIPS Act funding to proceed.

Intel’s Chandler, Arizona campus adds two more fabs to the spine. Fab 52 is in high-volume production on Intel’s 18A process — the first U.S. facility to cross the 2-nanometer threshold, with Intel’s own chief technology officer describing it as capable of more than 10,000 18A wafer starts per week — while Fab 62 is under construction and expected to be ready around 2028. Intel’s commitment to Arizona has held even as the company cancelled comparable projects in Germany and Poland, consolidating its advanced manufacturing investment domestically rather than abroad.

Taken together, the southern spine’s fab buildout shows a pattern consistent with the rest of this paper’s findings: delays have been common, but they have been delays of one to two years driven by labor availability, technology upgrades, and customer demand — not cancellations, and not capital withdrawal. Every fab named above is either producing today or under active construction with a committed completion date. That distinguishes this buildout sharply from the data center cancellation wave described in Section 1, where the setbacks have come from local political rejection rather than engineering or market timing.

SECTION 4 — THE WATER CONSTRAINT AND THE ENGINEERING RESPONSE

The binding constraint on the southern spine is water, in a region already under acute and well-documented stress. Arizona’s water supply depends heavily on the Colorado River, which has experienced sustained shortage conditions for over two decades, and on groundwater aquifers facing long-term depletion as the state’s population and industrial base both grow. Texas faces a parallel and independently documented problem: the state’s own water planning authorities project a long-term supply deficit of nearly 6.9 million acre-feet, with officials warning that without expanded infrastructure, demand could outstrip supply during the next prolonged drought. Into both constrained systems, the fab buildout documented in Section 3 introduces a new and significant industrial water demand layered on top of an existing shortage.

The scale of that demand is best understood by direct comparison. A single semiconductor fabrication facility consumes roughly 10 million gallons of ultrapure water per day for wafer rinsing and cleaning, per the World Economic Forum’s commonly cited industry figure — though independent technical sources put a single fab’s daily draw anywhere from roughly 3 million to 10 million gallons depending on facility size and process node, with the largest individual facilities and multi-fab campuses running higher still. A single large data center, by contrast, consumes roughly 1.5 to 5 million gallons per day. The fab is the larger water consumer, by a factor of two to four, and the figure is not contested across sources the way some data center estimates are — fab water demand is a directly measured manufacturing input, not a variable cooling-technology estimate.

The engineering and capital response to this demand is already under construction, not theoretical, and it spans every major fab in the southern spine. TSMC operates a water reclamation system for its Arizona fabs modeled on its home-market facility, which already supplies 10,000 cubic meters of recycled water daily and is scheduled to scale to 36,000 cubic meters per day by 2026, against a company-wide target of sourcing 60% of fab water from reclaimed sources by 2030; TSMC’s process water recycling rate already stood at 90.3% company-wide in 2023. Intel’s Ronler Acres fab in Oregon installed a full zero-liquid-discharge wastewater plant and achieved “net positive” water status — returning more clean water to the local watershed than the facility withdraws — as of 2022, a model the company is extending to its Arizona campus. A dedicated third-party water-engineering industry has formed around this need: providers such as Gradiant have built treatment facilities achieving up to 99% water recycling rates for individual semiconductor clients, at investments running into the hundreds of millions of dollars per facility.

The two Texas sites show the same pattern with even harder capital evidence, because the surrounding municipalities had to rebuild their own infrastructure to accommodate the fabs. Samsung’s Taylor facility is served by a dedicated water system built specifically for it: EPCOR, a third-party utility, owns and operates both the Sandow Water Project, which treats groundwater drawn from the Carrizo-Wilcox Aquifer, and the Blue Sky Water Reclamation Facility, targeting 75% process water reuse within the fab itself. In Sherman, the city government restructured its entire water infrastructure plan around the Texas Instruments and GlobalWafers buildout — a plan originally built around 1.4% annual growth had to be reassessed entirely once the two manufacturers announced their facilities. The city approved more than $400 million in water system capital improvements and a separate $500 million infrastructure package tied directly to TI’s $30 billion plant, bringing a new wastewater treatment plant online that lifted Sherman’s total daily treatment capacity to 16 million gallons specifically to serve the new industrial demand.

None of this is a future commitment. These are operating systems, signed utility contracts, and completed or substantially funded municipal capital projects, built ahead of and alongside the fabs themselves.

The honest limit of this response must be stated plainly, because the engineering is real but incomplete. Industry-wide, an average of 76% of the water a fab withdraws is used directly in the manufacturing process, and most current recycling captures water for secondary systems — cooling towers, scrubbers — rather than returning it into the ultrapure water loop that touches the wafer itself. Closing that final loop, so water cycles through the core manufacturing process multiple times rather than once, remains unsolved at most facilities, even as it has been achieved at a handful of sites, including Intel’s Oregon plant. The fabs documented in Section 3 are being built faster than the water engineering needed to make their long-term operation fully sustainable in two of the most water-constrained states in the country — a gap the capital flows above show the industry is actively working to close, not one it has already closed.

CONCLUSION

Across this paper and the series it belongs to, the forces shaping the buildout of Fortress America are moving with a speed that is itself a finding. Section 1 documented more than $205 billion in announced semiconductor investment concentrated in Arizona alone, with Texas building a complementary corridor anchored by a $16.5 billion contractual link between Samsung’s Taylor fab and Tesla’s robotics and vehicle silicon. Section 2 traced the rare earth supply chain from Mountain Pass to Independence to a $400 million Department of Defense equity stake — the federal government’s largest and most structurally complete bet on any single company in this series, complete with a 10-year price floor and an off take agreement guaranteeing the purchase of an entire facility’s output. Section 3 showed the semiconductor spine itself: fabs producing today in Arizona, fabs ramping in Texas, delays measured in months and quarters rather than cancellations. Section 4 identified the constraint that ties all of it together — a fab’s daily water demand exceeds a large data center’s by a factor of two to four, in a region already short on both the Colorado River allocations and groundwater that the rest of the regional economy depends on.

What several billion-dollar projects already producing, and several more under active construction, demonstrate is that these are not speculative bets being made by capital hoping a policy environment holds steady. They are commitments structured to survive a change in administration, because the surrounding institutions — state legislatures funding water infrastructure before the fabs are finished, the Department of Defense taking equity rather than simply writing a grant, municipal governments rebuilding their utilities ahead of demand — are treating the underlying strategic problem as a fixed feature of the next decade, not a four-year policy preference.

The clearest evidence of that institutional permanence is the rare earth chain itself, and the timeline is worth stating plainly because it spans three administrations without reversing direction once. The Department of Defense’s first rare earth investment in MP Materials came in 2020, under the first Trump administration, a $10 million Defense Production Act award to a company struggling against Chinese retaliatory tariffs. The Biden administration not only continued the relationship but expanded it — $45 million for Mountain Pass processing, more than $288 million to a second rare earth company, and a formal five-year mine-to-magnet investment strategy announced in 2024. The second Trump administration then escalated that foundation into the $400 million equity stake, the price floor, and the off take guarantee detailed in Section 2 — by far the largest single commitment in the chain’s six-year history. Three administrations, two parties, one direction, each handoff larger than the last.

This is the new mechanism this series has identified — equity rather than grants, ownership rather than subsidy — and it will be explored in full in the final paper of this series, which examines how the federal government is securing the raw material end of the supply chain across rare earths, copper, cobalt, and lithium, and what that ownership costs in dollars and control.

The water constraint is the one piece of this picture without a finish line yet. Engineering is being actively deployed against it — TSMC’s reclamation plant, Intel’s zero-discharge facility, the dedicated water systems municipalities in Texas have built specifically to serve Samsung and Texas Instruments — but the gap between what is being recycled today and what full closed-loop production requires remains real, and semiconductor fabrication itself, not the data centers that draw more public attention, is the larger consumer of the water at stake. Closing that gap is necessary for the mine-to-magnet-to-motor chain to function entirely on American soil.

The supply chain is now visible, end to end, for the first time. The constraints are known and named. What remains is execution — and the next paper in this series shows how deliberately, and how far back, that execution has already begun.

REFERENCES

[1] Arizona Commerce Authority; Office of Governor Katie Hobbs, remarks at SEMICON West 2025, Phoenix, October 2025.
[2] Industrial Info Resources, Arizona Industrial Construction Pipeline, 2025.
[3] Office of Governor Greg Abbott, Texas Semiconductor Innovation Fund Grant Announcement, September 17, 2025.
[4] Samsung Electronics, Seoul Stock Exchange Regulatory Filing, July 28, 2025; Bloomberg, Reuters, CNN, July 28, 2025.
[5] Tom’s Hardware, “TSMC accelerates production timeline for new Arizona factory,” December 2025; Data Center Dynamics, “TSMC says Arizona fab is now ahead of schedule,” 2026.
[6] Data Center Watch (10a Labs), Q1 2026 Report, datacenterwatch.org; NBC News, “Study shows state and local opposition to new data centers is gaining steam,” 2026; Tom’s Hardware, “More than 75 data center build-outs worth $130 billion have been successfully blocked in the first three months of 2026,” 2026; Fox Business, “Chandler, Arizona, city council unanimously votes against AI data center,” December 2025.
[7] UltraFacility, “Semiconductor in numbers: Global fab construction timelines,” April 2026; Intel Newsroom, Ohio project status statements, 2025-2026.
[8] Rare Earth Exchanges, “How Rare Earth Elements Enable Modern Semiconductor Manufacturing Equipment,” January 2026.
[9] World Population Review / USGS rare earth production data, 2025; Optimusk, “Tesla Optimus Supply Chain,” 2026; 36kr English, April 2026.
[10] MP Materials Corp., Form 8-K, January 22, 2025, U.S. Securities and Exchange Commission, “MP Materials Restores U.S. Rare Earth Magnet Production”; Fort Worth Report, “Fort Worth manufacturer begins producing rare earth magnets,” January 2025.
[11] MP Materials Corp., Form 8-K, July 2025, U.S. Securities and Exchange Commission; CNBC, “Pentagon to become largest shareholder in rare earth miner MP Materials,” July 2025.
[12] MP Materials Corp., Form 8-K, July 2025, U.S. Securities and Exchange Commission; The Defense Post, “Pentagon Takes Stake in US Rare Earth Company,” July 2025.
[13] MP Materials Corp., Form 8-K, July 2025, U.S. Securities and Exchange Commission; Select Committee on the Chinese Communist Party, U.S. House of Representatives, witness testimony, November 2025; C&EN, “US invests in rare earth firm MP Materials,” July 2025.
[14] Investing News Network, “Trilogy Metals Shares Rocket as US Government Takes Stake in Alaska Project,” October 2025; Axios, “US to take 10% stake in Trilogy Metals,” October 2025; Mayer Brown, “US Government Equity and Equity-Linked Investments in Critical Minerals,” April 2026.
[15] Internal Revenue Code Section 45X, Advanced Manufacturing Production Credit, as amended.
[16] U.S. Department of the Treasury, guidance on Section 45X critical minerals phasedown schedule, 2025-2026.
[17] MP Materials Corp., Form 8-K filings, U.S. Securities and Exchange Commission, 2025.
[18] Council on Foreign Relations, “The CHIPS Act: How U.S. Microchip Factories Could Reshape the Economy,” 2024; Semiconductor Industry Association, Chip Incentives & Investments data.
[19] TSMC Arizona, company facility status page, 2026; Tech Insider, “TSMC’s $165B Arizona GigaFab: Reshaping US Chips,” 2026.
[20] Tom’s Hardware, “TSMC brings its most advanced chipmaking node to the US yet,” December 2025.
[21] TSMC Arizona, company facility status page, 2026.
[22] BlackRidge Research, “TSMC Arizona Fab: USD 165 Billion Semiconductor Project,” 2026.
[23] Texas Instruments, “Texas Instruments begins production at its newest 300mm semiconductor manufacturing facility in Sherman, Texas,” December 17, 2025.
[24] EE Times, “Inside Texas Instruments’ New 300mm Fab in Sherman, Texas,” December 2025.
[25] TI.com, “Sherman, Texas: 300mm wafer fabs,” company site, 2026.
[26] Tom’s Hardware, “Samsung delays $44 billion Texas chip fab,” July 2025; Electronics360, “Report: Full production of Samsung’s Texas fab possibly delayed to 2027,” March 2026.
[27] TechTimes, “Samsung Taylor Fab Production Confirmed for 2027,” May 2026.
[28] MLQ.ai, “Samsung Delays Completion of $44 Billion Texas Chip Plant,” July 2025.
[29] CNBC, “Intel aims to find clients and catch TSMC with new chip fab in Arizona,” December 19, 2025; Tom’s Hardware, “Intel’s fab roadmap examined — Arizona, Ohio, Ireland, and the two deadlines deciding 14A process node,” June 2026.
[30] UltraFacility, “Semiconductor in numbers: Global fab construction timelines,” April 2026.
[31] U.S. Bureau of Reclamation, Colorado River shortage condition declarations, 2022-2026.
[32] Texas 2036, “Foundation for Economic Growth: Assessing Texas’ Water Infrastructure Needs,” 2024.
[33] World Economic Forum, “Semiconductor manufacturing and big tech’s water challenge,” 2024; CWR, “8 Things You Should Know About Water & Semiconductors”; IDE Tech, “Water Sustainability in the Semiconductor Industry”; Semiconductor Engineering, “How Semiconductor Fabs Use Water,” August 2025.
[34] EESI, “Data Centers and Water Consumption”; MOST Policy Initiative, “Data Center Water Use,” April 2026.
[35] ScaleBan Equipments, “How Semiconductor Industry is Tackling Wastewater Challenges”; UltraFacility, “Semiconductor in numbers: Fab water sources,” April 2026.
[36] ScaleBan Equipments, “How Semiconductor Industry is Tackling Wastewater Challenges.”
[37] Manufacturing Dive / ESG Dive, “Semiconductor industry faces water, sustainability challenges,” August 2025.
[38] EPCOR USA, “EPCOR Tapped as Water Partner in Central Texas,” July 2023; 3A Makina, “Project Delivery: Texas, Samsung Semiconductor Fabrication Facility WWTP.”
[39] Freese and Nichols, “Award-Winning Water Expansion Powers Sherman’s Semiconductor Boom,” July 2025; Texas 2036, “Foundation for Economic Growth,” 2024; KTEN, “New wastewater plant adds to Sherman’s infrastructure,” October 2025.
[40] Manufacturing Dive, “Semiconductor industry faces water, sustainability challenges,” August 2025.
[41] Heatmap News, “The Pentagon’s Rare Earths Deal Is Making Former Biden Officials Jealous,” July 2025.
[42] The White House (Biden-Harris Administration Archives), “Fact Sheet: Biden-Harris Administration Takes Further Action to Strengthen and Secure Critical Mineral Supply Chains,” September 2024; Global Policy Watch, “Made in America: The Outlook for Critical Minerals,” October 2025; Vulcan Elements, “Money Finally Flowing to US Rare Earths Can’t Come Fast Enough,” August 2025.

FORTRESS AMERICA SERIES
Paper I: Fortress America — Hemispheric Consolidation, Infrastructure Cascades, and the Regional Transformation of the American Economy
Paper II: If It Can Work, It Can Fight — The Case for Domestic Humanoid Robotics Manufacturing
Paper III: Where Fortress America Lands — The Southern Spine and the Race to Build It
Paper IV: Forthcoming

DISCLAIMER: This paper represents independent analytical and systems research and is the third in the Blue Collar Analytics Fortress America series. Nothing contained herein constitutes financial advice, investment recommendations, or legal counsel. Readers should conduct your own due diligence and consult qualified professionals before making investment decisions.

Where Fortress America Lands Paper3 2

90.8KB ∙ PDF file

Download

Download

IF IT CAN WORK, IT CAN FIGHT

The Case for Domestic Humanoid Robotics Manufacturing

Author: Adam Wood | Publication: Blue Collar Analytics | Series: Fortress America, Paper II | Date: June 2026 | Version: 1.1

EXECUTIVE SUMMARY

The United States has seen this before. Corporate offshoring of semiconductor manufacturing created a strategic vulnerability that took decades to recognize and billions in legislative spending to begin correcting. The CHIPS and Science Act of 2022 was not innovation — it was remediation. The same structural error is now repeating in humanoid robotics, in accelerated form, with a state-backed Chinese competitor that has explicitly declared its intention to dominate the global market before American firms establish scale.

This paper argues that the Fortress America institutional framework — the same pattern of bipartisan policy response that produced the CHIPS Act — will produce an equivalent legislative and manufacturing response to the robotics challenge. It must. Because a robot that can work a factory floor can walk a battlefield, and that reality makes domestic manufacturing not an economic preference but a defense requirement.

The technology is here. The threat is documented in congressional testimony. Humanoid robots are already deployed on an active battlefield. What remains is execution — and this paper maps the argument for why execution is not only necessary but institutionally inevitable.

SECTION 1 — THE PRECEDENT: HOW AMERICA LEARNED THE HARD WAY

The pattern now emerging in humanoid robotics has played out before. For decades, American corporations offshored semiconductor manufacturing to cheaper economies, optimizing for cost per unit in an era when Pax Americana made strategic vulnerability feel theoretical. The semiconductor was invented in the United States, and America retained its lead in research and development — but ceded the physical manufacturing base that turns innovation into strategic power (White House, 2024).

The consequence took years to fully register. By the time policymakers acted, the United States had declined from producing roughly 40% of global semiconductors in 1990 to approximately 12% today, with none of the most advanced chips manufactured domestically (CFR, 2024). The remediation response was the CHIPS and Science Act of 2022 — $280 billion in authorized spending, combining the Endless Frontier Act’s investment in domestic research with the CHIPS for America Act’s manufacturing incentives, specifically designed to compete with China (Conference Board, 2025). The CHIPS R&D Office is investing $11 billion into a domestic R&D ecosystem while the CHIPS Program Office dedicates $39 billion to manufacturing incentives (NIST, 2026).

The lesson is not that offshoring is always wrong. The lesson is that offshoring strategic industries creates dependencies that adversaries can and will weaponize. China demonstrated this with precision when it imposed export controls on rare earth magnets in April 2025 — a licensing regime targeting high-performance magnet grades containing dysprosium and terbium, the specific compounds required for advanced robotics actuators, EV motors, and defense applications. This is not a blanket ban; it is a targeted chokehold on the performance tier advanced manufacturing actually requires, with a discretionary licensing process that functions as a second lever even on technically permitted shipments. Each Optimus humanoid robot requires approximately 3.5 kilograms of NdFeB magnets — with China controlling approximately 85 to 90% of global NdFeB production and approximately 70% of the Optimus component supply chain by value share — making the export control regime the single largest production risk in American humanoid robotics in 2026 (Optimusk, 2026; 36kr, 2026). The semiconductor playbook and the robotics playbook are identical. The question is whether America corrects the error before or after the damage compounds.

SECTION 2 — THE BOSTON DYNAMICS PROBLEM: WE BUILT IT AND SOLD IT

The convergence of artificial intelligence and humanoid robotics represents the most significant technological transition since the microprocessor. The defense establishment recognized this early. DARPA funded Boston Dynamics through 57 government contracts totaling nearly $200 million between 2008 and 2025, with Atlas — the primary American humanoid robotics platform — developed under direct DARPA oversight, with components produced by Sandia National Laboratories and iRobot (Science Arena, 2025; Wikipedia, 2026). This was not venture capital. This was public investment in a strategic defense technology.

Then the same pattern that hollowed out semiconductor manufacturing repeated itself. Boston Dynamics was sold to Google, then to SoftBank, and in June 2021 was acquired by Hyundai Motor Group for $1.1 billion — giving the South Korean automotive conglomerate an 80% controlling interest in the primary American defense-originated humanoid robotics platform (Hyundai, 2021). The technology American taxpayers funded through DARPA is now majority owned by a foreign corporation.

Unlike the semiconductor crisis, however, policymakers this time are sounding the alarm before the damage fully compounds. The Humanoid ROBOT Act of 2025 was introduced in the Senate, prohibiting defense contractors from using humanoid robots designed or manufactured by adversarial foreign entities and requiring the Secretary of Defense to submit a report on national security threats posed by humanoid robots in countries of concern (Congress.gov, 2025). The FY2026 National Defense Authorization Act includes provisions directing the Army to expand robotic automation in munitions manufacturing (Holland & Knight, 2025). The Special Competitive Studies Project issued a memo to the President explicitly calling for U.S. independent leadership in robotics and prevention of Chinese firms dominating the robotics stack (SCSP, 2025). Congressional testimony from March 2026 documented China’s robotics strategy from Made in China 2025 through the 14th Five Year Robotics Plan and the 2023 Robotics Plus action plan — all identifying robotics as strategic military-civil fusion technology (House Homeland Security Committee, Cybersecurity and Infrastructure Protection Subcommittee, 2026).

The institutional alarm is sounding. The legislative framework is forming. The question is whether it moves fast enough.

SECTION 3 — THE CHINESE STRATEGIC THREAT: MARKET DOMINANCE AND SUPPLY CHAIN COERCION

This paper does not argue that conflict with China is inevitable or desirable. The economic entanglement between the two nations is real and mutual destruction is not a rational outcome for either. What this paper argues is simpler and more documented: competitive desire to dominate strategic technology markets does not yield to common sense, and China has demonstrated both the intent and the capability to pursue robotics dominance through means that go beyond market competition.

The scale of China’s state-directed robotics push is without precedent in this technology sector. More than 140 Chinese manufacturers launched over 330 humanoid robot models in 2025 alone, described by Chinese authorities as the first year of mass production (Xinhua, 2026). China released its first national standard system for humanoid robots on March 3, 2026, covering the full industrial chain and lifecycle of humanoid robotics — a regulatory architecture designed to entrench domestic manufacturers and marginalize foreign competitors (AI Insider, 2026). The Jamestown Foundation documented that a single Beijing-registered humanoid robotics firm can simultaneously access over a dozen distinct policy support tools from one local government entity alone (Jamestown, 2026).

The price signal is the most visible manifestation of this strategy. Unitree Robotics launched its R1 humanoid at $5,900 in July 2025 — a price point previously considered years from viability — backed by RMB 76 million ($11 million) in tax incentives in the first nine months of 2025 alone, and RMB 32 million ($4.7 million) in direct government grants received cumulatively between 2022 and September 2025, as disclosed in Unitree’s own IPO prospectus (Jamestown, 2026). This is not organic market competition. It is state-subsidized price destruction designed to capture market share before American firms reach manufacturing scale.

The coercive dimension of the strategy is more significant than the price competition. China’s imposition of export controls on rare earth magnets in April 2025 demonstrated that supply chain weaponization is not a theoretical risk — it is an executed policy. Chinese manufacturers currently constitute approximately 70% of Tesla’s Optimus supply chain by component share (36kr, 2026). The export control regime did not just threaten future production — it created an immediate production constraint on the leading American humanoid robotics manufacturer. Market strategy and supply chain coercion operating simultaneously represent a more sophisticated competitive threat than American firms faced in semiconductors. The window for a corrective legislative response is open. It will not remain open indefinitely.

SECTION 4 — IF IT CAN WORK, IT CAN FIGHT

Mainstream media coverage of humanoid robotics has framed this technology primarily as a factory worker story — a labor displacement narrative focused on manufacturing productivity. That framing is not false. It is incomplete in ways that matter enormously for policy.

Humanoid robots are already deployed on a battlefield. Two Phantom MK-1 units — developed by San Francisco-based startup Foundation — were deployed to Ukraine in February 2026 for frontline reconnaissance and logistics support in hazardous areas, in trials conducted with direct U.S. government support (Interesting Engineering, 2026). Foundation has secured $24 million in confirmed contracts across the U.S. Army, Navy, and Air Force to evaluate robotic systems for inspection, logistics, and weapons-related tasks (Interesting Engineering, 2026). Foundation’s CEO has stated publicly that the company’s goal is for humanoid robots to be the first to enter dangerous missions, potentially carrying lethal weapons while keeping human soldiers out of immediate harm’s way (Reuters, 2026).

The Pentagon is not waiting for the technology to mature before committing resources. In July 2025, Defense Secretary Pete Hegseth issued a memorandum directing all service branches to accelerate acquisition of drone and robotic systems, establish dedicated robotic units within each service branch by FY2027, and increase funding for human-machine teaming research by 40% (Robozaps, 2026). The Pentagon’s Defense Autonomous Working Group budget is proposed to increase from $226 million to $54 billion under the FY2027 spending proposal — an increase of nearly 240 times in a single budget cycle (Defense One, 2026).

The historical pattern is consistent and unbroken. Military technology does not wait for perfection before deployment. The airplane was fabric and wood when armies mounted guns on it and began dropping bombs. The energy density limitations of current humanoid robot batteries, the payload constraints, the endurance gaps — these are the fabric and wood of 1914. They will be solved or worked around because the strategic imperative demands it. Tethered power systems and high-capacity battery loads represent near-term viable solutions for fixed battlefield and logistics support roles even before full autonomous energy independence is achieved.

What the deployment record in Ukraine and the Pentagon budget signals confirm is that this technology has already crossed the threshold from research to operational reality. The question for American policymakers is not whether humanoid robots will serve on a battlefield. The documented record confirms they already do. The question is whether the robots deployed for America will be built in America — or whether they will be built with Chinese components, in Chinese-owned factories, dependent on Chinese rare earth supply chains that Beijing has already demonstrated willingness to restrict.

SECTION 5 — THE AMERICAN MANUFACTURING IMPERATIVE

The legislative framework for a domestic response is already forming. The Humanoid ROBOT Act, the FY2026 NDAA robotics provisions, and the Special Competitive Studies Project presidential memo collectively represent the early institutional signal pattern that preceded the CHIPS Act — the same Fortress America bipartisan convergence around a recognized strategic vulnerability.

The domestic manufacturing anchor most capable of responding at scale is Tesla. Tesla’s SEC filings confirm that preparations for the first large-scale Optimus production facility at Fremont are underway, with a first-generation line designed for one million robots per year, and Gigafactory Texas being prepared for a second-generation line designed for long-term annual production capacity of ten million robots (Tesla SEC Filing, 2026). Over 1,000 Gen 3 Optimus units are already deployed internally across Tesla’s own manufacturing operations — the most advanced real-world deployment of humanoid robots in American industrial history (Tesla SEC Filing, 2026).

However, the supply chain reality must be stated plainly. Chinese manufacturers currently constitute approximately 70% of the Optimus component supply chain. The Shanghai Gigafactory — not Fremont or Texas — has been identified by Tesla’s own China president as the primary production site for achieving Optimus scale, citing China’s manufacturing system as the key to solving mass production challenges (BigGo Finance, 2026). This is not a solved problem. It is the problem the legislative response must address — and it mirrors precisely the dynamic that made the CHIPS Act necessary.

Tesla has already begun responding to supply chain weaponization without waiting for legislation. The Terafab initiative — announced March 21, 2026 — is Tesla’s in-house semiconductor fabrication program, a direct response to chip supply risk for both Cybercab and Optimus production (Optimusk, 2026). Alternative NdFeB magnet supply chains are being pursued in the United States and Australia in direct response to China’s April 2025 export controls (Optimusk, 2026). The corporate response to supply chain coercion is already underway. Legislative support — on the CHIPS Act model — would accelerate what market forces alone cannot complete fast enough.

The domestic rare earth supply chain is the foundation everything else rests on. MP Materials, currently the only operational rare earth mining and processing facility of scale in the United States, sits at the critical intersection of the Fortress America resource consolidation thesis and the humanoid robotics manufacturing imperative. A robot built in Texas with American rare earth magnets processed in California is a fundamentally different strategic asset than a robot assembled in Fremont with Chinese components that Beijing can restrict on short notice.

The manufacturing imperative is not complicated. Build the robot. Source the components domestically. Secure the supply chain. The CHIPS Act demonstrated that American institutional will can accomplish this when the strategic necessity is recognized. The deployment of humanoid robots in Ukraine, the $54 billion autonomous systems budget proposal, and the Chinese rare earth export control regime have collectively made the strategic necessity undeniable. The only remaining question is the speed of execution.

CONCLUSION

The United States does not have a robotics problem. It has a pattern problem. The same sequence that created semiconductor dependency — offshore for cost, cede manufacturing, discover the strategic vulnerability too late — is repeating in humanoid robotics with one critical difference: the warning signs are visible in real time and the legislative response has already begun.

A humanoid robot is already deployed on a battlefield in Ukraine. The Pentagon is proposing a $54 billion autonomous systems budget. China has imposed export controls on the rare earth materials that power humanoid actuators and state-subsidized competitors to a $5,900 price point that American manufacturers cannot match without domestic supply chain support. The Humanoid ROBOT Act is in the Senate. The NDAA has robotics provisions. The SCSP has sent a memo to the President.

The Fortress America institutional framework — the bipartisan convergence of defense necessity, legislative response, and domestic manufacturing investment — is already initiating around robotics exactly as this paper’s framework predicts. The CHIPS Act was remediation for a strategic error that took decades to correct. The robotics response has the opportunity to be prevention.

The institutional response already reflects a single coherent logic. The CHIPS Act, the Humanoid ROBOT Act, the FY2026 NDAA robotics provisions, and the $54 billion DAWG budget are collectively the documented expression of an institutional judgment that what is deployed for America’s defense must be built on American supply chains. This paper has mapped the pattern, the vulnerability, and the response already forming across successive administrations and both political parties. Whether the response moves fast enough is the only remaining open question.

If it can work, it can fight. The institutions are already acting on that logic. This paper has simply read the blueprint.

REFERENCES

U.S. Legislation & Federal Policy

CHIPS and Science Act (2022). Public Law 117-167. U.S. Congress. https://www.congress.gov/crs-product/R47523

Conference Board (2025, March 13). The Future of the CHIPS and Science Act. https://www.conference-board.org/research/ced-policy-backgrounders/the-future-of-the-CHIPS-and-Science-Act

Congress.gov (2025). S.3275 - Humanoid ROBOT Act of 2025. 119th Congress. https://www.congress.gov/bill/119th-congress/senate-bill/3275/text

Holland & Knight (2025, December). FY2026 National Defense Authorization Act: A Comprehensive Analysis. https://www.hklaw.com/en/insights/publications/2025/12/fy-2026-national-defense-authorization-act

NIST (2026, January 21). CHIPS for America. National Institute of Standards and Technology. https://www.nist.gov/chips

White House (2024, August 9). Fact Sheet: Two Years after the CHIPS and Science Act. https://bidenwhitehouse.archives.gov/briefing-room/statements-releases/2024/08/09/fact-sheet-two-years-after-the-chips-and-science-act

Defense & Military Sources

Defense One (2026, May 4). Pentagon seeks smarter, self-organizing drones as autonomous-warfare budget is poised to skyrocket. https://www.defenseone.com/technology/2026/05/pentagon-drones-autonomous-warfare/413323/

House Homeland Security Committee, Cybersecurity and Infrastructure Protection Subcommittee (2026, March 17). Examining the National Security Risks of PRC Artificial Intelligence. Congressional Testimony. https://docs.house.gov/meetings/HM/HM08/20260317/118982/HHRG-119-HM08-Wstate-DoshiPhDR-20260317.pdf

Interesting Engineering (2026). Humanoid robot trialed as soldier in deadly Ukrainian battlefield. https://interestingengineering.com/ai-robotics/humanoid-robot-battlefield-ukraine

Interesting Engineering (2025, December 17). US firm Foundation plans to build 50,000 humanoid robots by 2027. https://interestingengineering.com/military/us-foundation-build-50000-humanoid-robots

Reuters / CNBC (2026, May 30). This Trump-linked startup plans to put humanoid robots in the military. https://www.cnbc.com/amp/2026/05/30/humanoid-robots-ukraine-war-foundation-military-ai.html

Robozaps (2026). Humanoid Robots in Military and Defense. https://blog.robozaps.com/b/humanoid-robots-in-military-and-defense

SCSP / Special Competitive Studies Project (2025, February). Memos to the President: National Robotics Strategy. https://www.scsp.ai/wp-content/uploads/2025/02/Robotics-Memo.pdf

China & Geopolitical Sources

AI Insider (2026, March 1). China Releases National Standards for Humanoid Robotics and Embodied AI. https://theaiinsider.tech/2026/03/01/china-releases-national-standards-for-humanoid-robotics-and-embodied-ai/

Jamestown Foundation (2026, April 18). Policy Support for Robotics Firms Shows Defense Integration. https://jamestown.org/policy-support-for-robotics-firms-shows-defense-integration/

36kr English (2025, October 15). Elon Musk Places $685M Order with Sanhua: China Becomes Key Supplier for Optimus. https://eu.36kr.com/en/p/3510288514980998

36kr English (2026, April 24). Elon Musk’s Plan to Build One Million Robots: How Many Motors, Reducers, and Lead Screws Are Made in China? https://eu.36kr.com/en/p/3780414717129481

Boston Dynamics & Robotics History

Hyundai Motor Group (2021, June 21). Hyundai Motor Group Completes Acquisition of Boston Dynamics from SoftBank. https://www.hyundai.com/worldwide/en/newsroom/detail/hyundai-motor-group-completes-acquisition-of-boston-dynamics-from-softbank-0000000516

Science Arena (2025, September 8). Invisible state: How government contracts created Boston Dynamics. https://www.sciencearena.org/en/columns/invisible-state-how-government-contracts-created-boston-dynamics/

Wikipedia (2026). Atlas (robot). https://en.wikipedia.org/wiki/Atlas_(robot)

Tesla & Domestic Manufacturing

BigGo Finance (2026, April 16). Tesla Plans Mass Production of Humanoid Robots at Shanghai Gigafactory. https://finance.biggo.com/news/fmJ4lp0BZk7xib5fQwHY

CFR / Council on Foreign Relations (2024). The CHIPS Act: How U.S. Microchip Factories Could Reshape the Economy. https://www.cfr.org/articles/chips-act-how-us-microchip-factories-could-reshape-economy

Optimusk (2026). Tesla Optimus Supply Chain: Who Makes the Parts? https://optimusk.blog/blog/tesla-optimus-suppliers/

Tesla SEC Filing (2026, Q1). Form 8-K, Quarterly Results. U.S. Securities and Exchange Commission. https://www.sec.gov/Archives/edgar/data/0001318605/000162828026026551/exhibit991.htm

Tesla SEC Filing (2026, FY2025 Annual). Form 8-K, Annual Results. U.S. Securities and Exchange Commission. https://www.sec.gov/Archives/edgar/data/0001318605/000162828026003837/exhibit991.htm

FORTRESS AMERICA SERIES
Paper I: Fortress America — Hemispheric Consolidation, Infrastructure Cascades, and the Regional Transformation of the American Economy
Paper II: If It Can Work, It Can Fight — The Case for Domestic Humanoid Robotics Manufacturing
Paper III: Forthcoming

DISCLAIMER: This paper represents independent analytical and systems research and is the second in the Blue Collar Analytics Fortress America series. Nothing contained herein constitutes financial advice, investment recommendations, or legal counsel. Readers should conduct their own due diligence and consult qualified professionals before making investment decisions.

If It Can Work It Can Fight Adam Wood

78.4KB ∙ PDF file

Download

Download

Note: This piece was revised on 6/17/2026 after additional fact-checking — a sourcing error in the opening statistic was corrected, one low-quality citation was replaced, and the water/PFAS section was updated to reflect a regulatory change that occurred after initial publication. Full text and PDF below reflect v1.1.

FORTRESS AMERICA

Hemispheric Consolidation, Infrastructure Cascades, and the Regional Transformation of the American Economy

Author: Adam Wood | Date: June 2026 | Version: 1.1

EXECUTIVE SUMMARY

U.S. trade as a share of GDP peaked at 31% in 2011 and has fallen to roughly 25% by 2024 (World Bank; U.S. Bureau of Economic Analysis). This is not a cyclical fluctuation — it marks a structural break in the post-war economic order that is now reshaping how the United States positions itself for the next half century.

To maintain financial and technological superiority in a fragmented world, the United States is pursuing what this paper terms Fortress America: the systematic consolidation of the Western Hemisphere into an economically and resource self-sufficient bloc. This strategy is not a declared policy but an observable pattern of institutional behavior visible across Treasury documents, bipartisan legislation, and physical capital deployment.

The implementation follows a logical sequence. Friendshoring — the realignment of allied manufacturing into the Western Hemisphere — is already explicit in U.S. policy language. Reshoring that manufacturing base creates cascading infrastructure demands: $1.4 trillion in energy grid investment is committed through 2030, with major natural gas turbine deliveries to regional utilities confirming the buildout is underway. Industrial reshoring at scale will simultaneously create acute wastewater and water treatment demands that regulators, states, and courts are already mandating solutions for.

The result will be a measurable geographic transformation of American regions that can be tracked in real time through capital flows, regulatory actions, and physical infrastructure deployment. The Gulf Coast energy export corridor is already emerging as one of the first visible nodes of this realignment.

This paper maps that transformation, identifies where capital is flowing ahead of public awareness, and argues that the infrastructure bottlenecks created by Fortress America represent the defining investment and policy terrain of the next decade. The framework’s predictive power is already visible — U.S. energy export infrastructure is being tested and expanded in real time as global supply routes face unprecedented disruption.

SECTION 1 — GLOBAL TRADE FRAGMENTATION

U.S. trade as a share of GDP peaked at 31% in 2011 and has fallen to roughly 25% by 2024 (World Bank; U.S. Bureau of Economic Analysis). This single data point marks more than a cyclical downturn — it establishes a structural break in the post-war economic order. The IMF estimates that full fragmentation into competing trading blocs could reduce global GDP by 0.2 to 7%, with a two-bloc scenario alone producing a 5% contraction (IMF, 2023). The globalized order that served American interests since the end of the Cold War is not in decline — it is ending. A multipolar world has arrived, and American institutional behavior suggests adaptation is already underway.

SECTION 2 — FORTRESS AMERICA: THE HEMISPHERIC CONSOLIDATION STRATEGY

That adaptation is what this paper terms Fortress America: the systematic consolidation of the Western Hemisphere into a self-contained economic, technological, and resource bloc capable of producing and exporting energy and technology on American terms.

The evidence for this consolidation is behavioral rather than declarative. No administration has announced Fortress America as policy. What exists instead is a consistent pattern of institutional action across successive administrations and both political parties that produces the same directional outcome regardless of stated intent. Whether these actions reflect a coordinated grand strategy or the convergent self-interest of competing institutional factions is ultimately unknowable from the outside. What is knowable is the outcome — and the outcome follows a single coherent logic.

That logic is visible in a sequence of hemispheric actions. On January 3, 2026, U.S. Delta Force operators captured Venezuelan President Nicolás Maduro in Operation Absolute Resolve, removing a hostile government from the country holding the largest known oil reserves in the world (Task & Purpose, 2026). The official justification was narco-terrorism. The structural consequence — access to Venezuelan oil on terms favorable to U.S. interests — is identical regardless of which explanation one accepts.

Sustained pressure on the Cuban government, including an oil blockade in effect since January 2026, has produced a documented humanitarian crisis. UN officials report that humanitarian needs on the island remain acute and persistent as a direct result of the fuel shortage, with rolling blackouts, disrupted water delivery, and delayed medical care affecting millions of residents (UN News, 2026). As of this writing, the blockade remains in effect alongside ongoing social unrest on the island. Whatever its ultimate resolution, the blockade has already secured effective control over Caribbean maritime transit corridors critical to hemispheric energy and commercial shipping.

The ongoing attempt to acquire Greenland signals a strategic calculation about Arctic shipping routes and rare earth mineral deposits as polar regions become economically accessible.

Collectively these moves follow one logic: no hostile regime will be permitted to control a critical resource or transit chokepoint within the Western Hemisphere. This is not a new American instinct — it is the Monroe Doctrine updated for the resource and technology competition of the 21st century.

At the economic level the same pattern holds across administrations. Treasury Secretary Janet Yellen formally introduced friendshoring in April 2022 — the policy of favoring allied nations for critical supply chains — explicitly acknowledging that the old model of open global trade had become a strategic liability (U.S. Treasury, 2022). The CHIPS and Science Act committed $280 billion to reshore semiconductor production (CHIPS Act, 2022), directly addressing America’s decline from producing roughly 37 to 40% of global semiconductors in 1990 to about 10 to 12% today, with none of the most advanced chips manufactured domestically (CFR, 2024; Semiconductor Industry Association). The Infrastructure Investment and Jobs Act and the Inflation Reduction Act layered further capital into domestic industrial capacity.

Taken together, these are not isolated policy decisions responding to individual crises. They are the institutional expression of a coherent hemispheric strategy — executed incrementally, justified variously, but directionally consistent across every administration that has touched them.

SECTION 3 — THE ENERGY SPINE: GULF COAST AS GLOBAL SUPPLIER

The backbone of the Fortress America bloc is energy — its production, its distribution, and increasingly, its export. This is no longer a projection. Live data confirms the transition is already underway.

U.S. crude oil and petroleum product exports reached a record of nearly 12.9 million barrels per day in 2026, while LNG exports set an all-time high in March, as buyers across Asia and Europe turned to American supplies to offset shortages caused by Middle East conflict (Energy News Beat, 2026). Crude exports alone rose from 3.92 million barrels per day in January 2026 to 5.44 million barrels per day by April — a roughly 39% increase in three months (EIA, 2026). A volume increase of that scale mechanically requires a substantial, sustained surge in tanker traffic calling at U.S. Gulf ports, consistent with the elevated vessel activity widely observed on AIS tracking during this period. In a striking illustration of the reversal underway, Corpus Christi shipped more gasoline, diesel, and jet fuel to the Middle East in the first quarter of 2026 alone than in all of 2025 combined (EIA, 2026).

Iran’s closure of the Strait of Hormuz on March 2, 2026, drove Middle East tanker rates to all-time highs while simultaneously driving U.S. Gulf Coast rates to record levels — the same disruption that crippled one supply source made the alternative more valuable (EIA, 2026).

Even a rapid end to hostilities cannot quickly reverse these dynamics. Energy infrastructure damaged or idled during conflict does not return to operational capacity on a ceasefire timeline — it returns on an engineering and capital investment timeline measured in months to years. The extent of infrastructure damage across Gulf producing nations remains unknown at the time of writing. A conservative estimate places meaningful Middle East supply recovery at six to twelve months at minimum. The market share and buyer relationships the U.S. Gulf Coast captures in that window do not automatically return to prior suppliers when the window closes — global energy procurement officers now carry institutional memory of Hormuz closure that will drive supply diversification regardless of how quickly the conflict resolves.

This export surge does not sustain itself without a corresponding domestic investment in the energy infrastructure required to produce and move supply at scale. Here the Fortress America strategy reveals its second layer. The $1.4 trillion committed to U.S. energy grid investment between 2025 and 2030 — double the total investment of the prior decade — is not a response to the current conflict (Morningstar DBRS, 2025). It was already underway: individual utilities including American Electric Power ($72 billion through 2030), Duke Energy (over $100 billion), Southern Company ($81 billion), and Dominion Energy ($50 billion, with nearly 10 gigawatts of demand already locked into take-or-pay contracts) have board-approved or regulator-approved capital plans already being executed on a multi-year disbursement schedule. The conflict has validated the investment thesis, but the infrastructure buildout preceding it suggests institutional actors anticipated exactly this kind of supply disruption and positioned accordingly.

At the regional level this is already physically visible. Major natural gas turbines weighing over one million pounds are being delivered to regional grid operators, committing energy infrastructure to 30 to 40 year operational horizons. These are not hedges — they are irreversible bets on American energy dominance at the regional scale. The Department of Energy has warned that absent new firm capacity additions, blackout frequency could increase by up to 100 times by 2030 (DOE, 2026) — a projection that has drawn methodological push back from FERC-aligned analysts and clean-energy groups, who argue it undercounts the contributions of wind, solar, and storage. Even critics of the report’s methodology do not dispute the underlying driver: demand growth from data centers and reshored industry is real, accelerating, and creating the regulatory and political pressure that ensures the investment cycle cannot be stopped by any single administration.

The Gulf Coast is emerging as the spine of this system — the export corridor through which Fortress America projects energy power into global markets. That corridor is being physically expanded in real time. Houston’s crude export capacity increased from 2.15 to 2.42 million barrels per day between early 2025 and early 2026, while the Houston Ship Channel was widened from 530 to 700 feet to accommodate increased tanker traffic (RBN Energy, 2026). The channel widening and the broader grid investment cycle described above are separate undertakings — the former authorized under the 2020 Water Resources Development Act and executed across the Biden and second Trump administrations, the latter a distinct utility capital cycle — but both reflect the same underlying pattern of infrastructure expansion that outlasts any single administration. These are infrastructure commitments that will define the region’s economic character for a generation.

SECTION 4 — THE WATER CASCADE: DECONTAMINATION AS STRATEGIC INFRASTRUCTURE

The Fortress America buildout described in the preceding sections produces a cascading infrastructure consequence that has received less analytical attention than it deserves: water. The simultaneous reshoring of semiconductor fabrication, chemical manufacturing, and heavy industry — combined with the power demands of AI data center expansion and the population growth that follows industrial development — will place acute stress on American water systems that were not designed for this scale or this chemistry.

The media narrative has focused primarily on water consumption — specifically the volume data centers use for cooling. That framing is not false but it is misleading in proportion. Modern data center facilities increasingly use closed-loop cooling systems that recycle the majority of their water internally. The evaporative loss that generates headlines is real but secondary to the industrial water challenge that reshoring creates at scale. Semiconductor fabrication requires ultrapure water in large volumes and produces toxic process effluent containing PFAS compounds, heavy metals, and industrial solvents. Reshored chemical and battery manufacturing introduces additional hazardous waste streams that municipal water systems were never built to absorb. The water story of Fortress America is not consumption — it is decontamination.

The regulatory signal confirming this is on record, though it is now more complicated than a single rule. On April 10, 2024, the EPA finalized the first legally enforceable federal drinking water standards for PFAS compounds — setting maximum contaminant levels for six PFAS substances under the Safe Drinking Water Act (EPA, 2024). The rule affects an estimated 6 to 10 percent of the 66,000 covered public water systems nationally and originally required compliance by 2029 (EPA, 2024). As of May 2026, however, EPA has proposed rescinding four of the six regulated compounds — PFHxS, PFNA, HFPO-DA, and the Hazard Index mixture rule — retaining only PFOA and PFOS, with a delayed compliance deadline of 2031 (EPA, 2026).

The federal retreat has not stopped the underlying compliance dynamic; it has redistributed it. California, Massachusetts, New York, and Vermont already maintain PFAS drinking water standards stricter than the federal rule, several enacted explicitly in anticipation of federal rollback — California’s Assembly Bill 794, for instance, directs the State Water Board to maintain standards at least as protective as federal levels regardless of what Washington does. A federal court separately declined EPA’s request to fast-track its own rescission in March 2026, leaving the original rule’s legal status unresolved. The mandate-to-spending mechanism this paper describes is therefore running increasingly through states and courts rather than through the federal rule alone — arguably a stronger illustration of this paper’s central claim than reliance on a single federal rule would have been: institutional direction persists independent of any single administration’s choices.

The scale of this opportunity is not speculative. Tetra Tech’s CEO stated publicly that approximately 150,000 U.S. utilities will need to evaluate their water systems for PFAS compliance under the new federal mandate, and that the firm intends to serve that demand through existing municipal contracts (Citrini Research, 2024). Tetra Tech has held the number one ranking in water engineering by Engineering News-Record for twenty consecutive years (ENR, 2023). This is not a company positioning speculatively for future demand — it is the dominant incumbent in a regulated market where federal and state law have created mandatory spending. The distinction matters: this is not growth capital chasing an opportunity. It is essential infrastructure spending that must happen regardless of economic conditions.

The private sector more broadly has already identified and positioned around this dynamic. Engineering and environmental services firms specializing in water treatment, PFAS remediation, and industrial wastewater management are reporting record backlogs driven by government contract awards tied directly to federal and state environmental mandates. Tetra Tech alone reported record annual revenue of $5.2 billion in fiscal 2024 — up 15% year over year — with a $5.4 billion project backlog (Tetra Tech SEC Filing, 2024). PFAS-specific awards within that backlog include an $800 million U.S. Army Corps of Engineers PFAS remediation contract and a $464 million U.S. Army Environmental Remediation Services contract covering PFAS investigation and remediation at Army installations nationwide (Tetra Tech SEC Filing, 2024). Their own leadership stated publicly that they anticipate increased demand specifically tied to “water-reliant infrastructure, including data centers and industrial manufacturing” (Tetra Tech SEC Filing, 2025). Companies operating in adjacent sectors — hazardous waste disposal, industrial decontamination, and smart water infrastructure — are reporting similar dynamics. The revenue of firms in this space is not primarily dependent on market sentiment. It is dependent on federal and state law and the physical reality of toxic industrial byproducts that must be treated regardless of broader economic conditions.

Water in Fortress America is therefore not a crisis to be managed after the fact. It is a known bottleneck in a known buildout that regulators, at the federal level where politically tenable and at the state and judicial level where it is not, are already mandating solutions for, and that industry is already positioning around. The question is not whether the spending happens — the regulatory and industrial logic makes it structurally inevitable. The question is which regions bear the burden of that buildout, and which capture the economic benefit of hosting it.

SECTION 5 — REGIONAL TRANSFORMATION: READING THE BLUEPRINT IN REAL TIME

The preceding sections have established the strategic logic, the policy framework, and the capital commitment behind Fortress America. What follows from that logic is not clairvoyance — it is straightforward systems analysis. Regions that sit at the intersection of energy infrastructure investment, manufacturing reshoring, and water treatment demand will experience sustained, asset-backed economic transformation over the next decade. Unlike speculative capital chasing market sentiment, the investment now flowing into American infrastructure is anchored in physical assets with 30 to 40 year operational horizons. This is shovel and pick investing — slower, less meteoric than technology speculation, but backed by real assets performing real functions in the real world.

The regional transformation is already physically visible to anyone paying attention at ground level. In March 2026, Xcel Energy transported two GE-manufactured natural gas turbines from a rail yard in Loveland, Colorado to the Fort St. Vrain Generating Station in Platteville — part of a $500 million investment, approved by the Colorado Public Utilities Commission as part of the state’s 2024 Clean Energy Plan, that will add 200 megawatts of capacity and make Fort St. Vrain Xcel’s largest plant in Colorado (CBS Colorado, 2026). Each turbine weighs approximately 1.25 million pounds and required a specialized crew of 13 from Mammoet — a Dutch company specializing in engineered heavy lifting for the energy and petrochemical sectors — operating dual trailers with 166 tires each, moving at 5 miles per hour with Interstate 25 closed in both directions during crossings (CBS Colorado, 2026; Colorado Department of Transportation, 2026). This is part of Colorado’s 2024 Clean Energy Plan, approved by the Colorado Public Utilities Commission, targeting 6,100 megawatts of new generation capacity for the region (Lyons Today, 2026).

This is not a regional anomaly. It is a local expression of a national pattern. The $1.4 trillion in energy grid investment committed between 2025 and 2030 — double the prior decade’s total — is being deployed region by region through exactly these kinds of projects (Morningstar DBRS, 2025). The Department of Energy has warned that without new firm capacity additions, blackout frequency could increase by up to 100 times by 2030 — a contested projection, as noted above, though the underlying demand pressure behind it is not seriously disputed (DOE, 2026). That pressure ensures the investment cycle cannot be interrupted by any single administration. Where that investment lands first signals where manufacturing will cluster next, where water treatment infrastructure will follow by regulatory necessity, and where regional economies will transform as a result.

The Gulf Coast represents the most advanced expression of this regional transformation currently underway. As established in Section 3, U.S. crude and petroleum product exports have reached record levels as Middle East supply disruption redirected global energy demand toward American producers. The physical infrastructure confirms this shift is being expanded in real time — Houston’s crude export capacity increased from 2.15 to 2.42 million barrels per day between early 2025 and early 2026, and the Houston Ship Channel was widened from 530 to 700 feet to accommodate increased tanker traffic (RBN Energy, 2026). These are generational infrastructure commitments. Even if Middle East hostilities end tomorrow, damaged production infrastructure across Gulf producing nations cannot recover on a ceasefire timeline — engineering and capital investment timelines measured in months to years govern that recovery, not diplomatic ones. A conservative estimate places meaningful competitive supply recovery at six to twelve months at minimum, during which U.S. Gulf Coast export relationships and market share continue to deepen.

The method for tracking this regional transformation in real time is available to any analyst willing to use it. Capital flows through regulatory approvals, utility commission filings, corporate backlog disclosures, congressional STOCK Act trading records, and federal contract awards before it becomes visible in economic data. The turbines moving through northern Colorado at 5 miles per hour on a closed interstate are not a traffic story — they are a leading indicator of regional economic transformation written in steel and concrete. Multiply that observation across every regional grid operator in the country and the map of Fortress America’s domestic buildout becomes legible.

CONCLUSION

No smoke-filled room produced the realities described in this paper. What drives the Fortress America buildout is not a secret cabal but something at once more mundane and more powerful — the convergence of institutional self-interest, geopolitical necessity, and physical reality. Governments respond to fragmentation. Capital follows infrastructure. Regulators mandate what engineers then build. The actors are competing factions pursuing their own interests, and the outcome looks coordinated because the underlying pressures point every serious institutional player in the same direction.

These are not abstractions. They are observable in policy documents, corporate earnings filings, regulatory dockets, congressional trading records, and in the physical world itself. The analysis underlying this paper began not in a financial terminal or a think tank but on a highway in northern Colorado, watching two GE-manufactured turbines — each weighing 1.25 million pounds — move through the author’s own community at 5 miles per hour on a closed interstate. That observation was not a curiosity. It was a data point. A $500 million capital commitment with a 30 to 40 year operational horizon does not move through your backyard by accident. It moves because someone made an irreversible bet on the future of that region — and bets of that size, made by institutions with access to the best available information, are among the most reliable signals available to any analyst paying attention.

The thesis of this paper is not that the future is certain. Institutions are adaptive but not omniscient. Chaos is real and often profitable without being terminal. The Middle East conflict could resolve faster than projected. Regulatory frameworks could shift. Reshoring timelines could slip. These are genuine risks that any honest analysis must acknowledge.

What this paper argues is more durable than any single prediction: that an investor, policymaker, or analyst who understands the structural logic of Fortress America — hemispheric consolidation, energy spine development, infrastructure cascades in water and semiconductors, and the regional transformation those cascades produce — is better positioned to navigate the instability now baked into the global system than one who reads only the headlines. The headlines are written for mass consumption. The map is written in steel, concrete, regulatory mandates, and capital flows.

A new world is being built. This paper is an attempt to read the blueprint while the concrete is still being poured.

REFERENCES

International Sources & Policy Documents

IMF (2023). World Trade Can Still Drive Prosperity. International Monetary Fund Finance & Development. https://www.imf.org/en/publications/fandd/issues/2023/06/world-trade-can-still-drive-prosperity-georgievaokonjo-iweala

World Bank. World Bank Open Data — Trade (% of GDP), United States. https://data.worldbank.org/indicator/NE.TRD.GNFS.ZS?locations=US

U.S. Bureau of Economic Analysis. National Income and Product Accounts.

U.S. Legislation & Federal Policy

CHIPS and Science Act (2022). Public Law 117-167. U.S. Congress. https://www.congress.gov/crs-product/R47523

EPA (2024). PFAS National Primary Drinking Water Regulation. Federal Register, April 26, 2024 (89 CFR 32532). https://www.federalregister.gov/documents/2024/04/26/2024-07773/pfas-national-primary-drinking-water-regulation

EPA (2026, May 18). EPA Advances Comprehensive PFAS Strategy with Legally Defensible, Practical, Scientifically Sound Drinking Water Protections. https://www.epa.gov/newsreleases/epa-advances-comprehensive-pfas-strategy-legally-defensible-practical-scientifically

U.S. Treasury / Yellen, J. (2022). Address to the Atlantic Council on Friendshoring, April 2022. Referenced in USITC Executive Briefings on Trade, April 2023. https://www.usitc.gov/sites/default/files/publications/332/executive_briefings/ebot_friendshoring_ree.pdf

Energy & Infrastructure Data

CBS Colorado (2026, March 14). One million pound combustion turbine to be moved through Northern Colorado. https://www.cbsnews.com/colorado/news/combustion-turbine-northern-colorado-traffic-disruptions/

CBS Colorado (2026, March 23). Second massive turbine transported across Northern Colorado. https://www.cbsnews.com/colorado/video/second-massive-turbine-transported-across-northern-colorado/

Colorado Department of Transportation (2026, March 13). Northeastern Colorado — delays on US287, CO66, I-25, March 14. https://www.codot.gov/news/2026/march-2026-news/northeastern-colorado-delays-us287-co66-i25-march14

DOE / U.S. Department of Energy (2026). Speed to Power: SPARK Funding Opportunity. Office of Electricity. https://www.energy.gov/speed-to-power

EIA / U.S. Energy Information Administration (2026, March 26). Middle East crude oil tanker rates reached a multi-decade high in March. https://www.eia.gov/todayinenergy/detail.php?id=67386

EIA / U.S. Energy Information Administration (2026, April). Short-Term Energy Outlook, monthly crude export data. https://www.eia.gov/outlooks/steo/

Energy News Beat (2026, April 25). US Energy Exports Hit Records as Energy Markets Shift. https://energynewsbeat.co/crude-oil/us-energy-exports-hit-records-as-energy-markets-shift/

Lyons Today (2026, March 5). Xcel Energy to partially close northern Colorado highways for turbine transport. https://nationaltoday.com/us/co/lyons-co/news/2026/03/05/xcel-energy-to-partially-close-northern-colorado-highways-for-turbine-transport/

Morningstar DBRS (2025, October). US Electric Utilities Entering Investment Super-Cycle. Referenced in Utility Dive. https://www.utilitydive.com/news/us-electric-utilities-investment-super-cycle-morningstar/803841/

RBN Energy (2026, May 11). Surge in US Crude Exports Ups Estimates of What Gulf Coast Terminals Can Handle. https://rbnenergy.com/daily-posts/blog/surge-us-crude-exports-ups-estimates-what-gulf-coast-terminals-can-handle

Humanitarian & Geopolitical Sources

UN News (2026, April 6). Cuba energy crisis: Humanitarian needs remain despite fuel supplies. https://news.un.org/en/story/2026/04/1167254

Task & Purpose (2026, January 3). Delta Force and other special operations soldiers carried out Venezuela raid. https://taskandpurpose.com/news/venezuela-maduro-delta-force-absolute-resolve/

CFR / Council on Foreign Relations (2024). The CHIPS Act: How U.S. Microchip Factories Could Reshape the Economy. https://www.cfr.org/articles/chips-act-how-us-microchip-factories-could-reshape-economy

Semiconductor Industry Association. Chip Incentives & Investments. https://www.semiconductors.org/chips/

Corporate & Industry Sources

Citrini Research (2024, June 18). In Conversation: Tetra Tech Inc. — CEO Dan Batrack on PFAS regulation and municipal demand.

Citrini Research

In Conversation: Tetra Tech Inc.

We spoke with Dan Batrack, CEO of one of our favorite stocks in our thematic basket on Water, Tetra Tech Inc. (TTEK US…

Read more

2 years ago · 40 likes · 1 comment

ENR / Engineering News-Record (2023). Tetra Tech Ranked #1 in Water for 20th Year in a Row. https://markets.financialcontent.com/clarkebroadcasting.mymotherlode/article/bizwire-2023-5-9-tetra-tech-ranked-1-in-water-by-engineering-news-record-for-20th-year-in-a-row

Tetra Tech SEC Filing (2024). Form 8-K, Fiscal Year 2024 Annual Results, November 13, 2024. U.S. Securities and Exchange Commission. https://www.sec.gov/Archives/edgar/data/0000831641/000110465924117683/tm2428315d1_ex99-1.htm

Tetra Tech SEC Filing (2025). Form 8-K, Fiscal Year 2025 Annual Results, November 2025. U.S. Securities and Exchange Commission. https://www.sec.gov/Archives/edgar/data/0000831641/000110465925110537/tm2530975d1_ex99-2.htm

DISCLAIMER: This paper represents independent analytical and systems research. Nothing contained herein constitutes financial advice, investment recommendations, or legal counsel. Readers should conduct their own due diligence and consult qualified professionals before making investment decisions

Fortress America White Paper Adam Wood

87.1KB ∙ PDF file

Download

Download

Subscribe now