Tag: AI Data Center Power

  • Google’s 890 MW Across 11 Reactors Means AI Nuclear Uprates Are Bought by the Fleet

    Google’s 890 MW Across 11 Reactors Means AI Nuclear Uprates Are Bought by the Fleet

    TL;DR · 30-second read

    The Short Version

    Google and Amazon are paying to get more electricity out of nuclear plants that are already running, rather than waiting for new ones, to feed the computers behind artificial intelligence.

    Google’s deal adds 890 megawatts, close to what one large reactor produces. Getting there takes upgrades at 11 different reactors, because each one can only be pushed about a fifth harder at most.

    The extra power arrives from 2028 into the 2030s. It flows into the shared regional grid, not straight into the companies’ computer buildings.

    Data Center Frontier reported that Google and Amazon have signed long-term agreements with Constellation Energy that support about 1,080 MW of additional generating capacity from existing nuclear reactors. Under an October 6 deal, Google signed a 20-year power purchase agreement covering 890 MW of new capacity from more than $4.3 billion in upgrades at 11 operating reactor units in Illinois, Pennsylvania and New Jersey, with first deliveries expected in 2028. Google also signed a separate 15-year agreement for 2,700 MW of existing output. Amazon’s September 30 agreement covers 690 MW from Maryland’s Calvert Cliffs, including about 190 MW of new capacity due between 2030 and 2032 and backed by more than $3 billion in upgrades.

    The deals follow Meta’s January 2026 agreements with Vistra, which cover 2,609 MW, including 433 MW of planned uprates at Perry and Davis-Besse in Ohio and Beaver Valley in Pennsylvania. The Nuclear Regulatory Commission expects 31 uprate applications through 2032, totaling about 2,421 MW of potential added capacity.

    Executive Summary

    Hyperscalers, the largest cloud and AI operators, are moving beyond buying the output of existing nuclear plants. They are now paying to increase that output through power uprates. An uprate is a regulator-approved increase in the power level at which a licensed reactor may run. Combined, the Google and Amazon agreements with Constellation underpin more than $7 billion in planned investment. The new capacity they add to the grid runs from 2028 into the early 2030s.

    The approach has clear advantages. These plants already hold licenses, grid connections and experienced workforces, so the companies avoid much of the permitting and construction burden of building new reactors. It also has a structural ceiling. Uprates are bounded by the reactor and equipment already on site, typically from under 2% up to about 20% of a plant’s output. Volume therefore has to be assembled across many reactors, and owners with large operating fleets are best placed to supply it.

    The deals also expose a separate problem. New generation entering the PJM regional grid does not guarantee power at a specific data center. FERC’s October 9 rejection of key elements of PJM’s large-load framework sharpened that gap between financing megawatts and delivering them to a site.

    Why 890 Megawatts Takes 11 Reactors

    The arithmetic of Google’s deal shows how this market works. Constellation is upgrading 11 operating units to deliver 890 MW of new capacity. Spread evenly, that is about 81 MW per unit. The companies have not given a per-unit split, but the national figures follow the same pattern. The NRC’s 31 anticipated uprate applications total about 2,421 MW, an average of roughly 78 MW each. These additions are meaningful, but they come in increments of tens of megawatts, not gigawatts.

    The ceiling comes from the engineering. The NRC recognizes three uprate classes:

    • Measurement uncertainty recapture: generally under 2%, achieved through more precise instrumentation such as feedwater flow measurement.
    • Stretch uprates: about 2% to 7%, drawing on margins already designed into the plant.
    • Extended power uprates: sometimes up to 20%, but requiring work on turbines, condensate pumps, generators, transformers and other systems.

    Amazon’s 190 MW at Calvert Cliffs is about 10.6% above the plant’s current 1,790 MW, which puts it beyond stretch territory and well into extended-uprate work. Even at the top of that range, a single plant contributes a bounded slice of its existing output.

    As a result, a buyer seeking hundreds of megawatts of new nuclear output has to aggregate across many reactors. Google did that within Constellation’s fleet. Meta did it across three Vistra plants: 213 MW at Perry, 80 MW at Davis-Besse and 140 MW at Beaver Valley. The suppliers with an advantage are owners of many operating units. For the hyperscaler, the practical limit on any one deal is the number of uprate-ready reactors a single counterparty can bring to the table.

    Headline Dollars Bundle Preservation With Growth

    Every one of these agreements pairs a modest block of new capacity with a much larger purchase of existing output. Google’s 890 MW of new capacity sits alongside a separate 15-year deal for 2,700 MW of existing generation. Amazon’s 690 MW at Calvert Cliffs includes only about 190 MW of new capacity. Meta’s 2,609 MW with Vistra includes 433 MW of uprates and 2,176 MW of existing generation.

    The contracts also support plant life extensions. Amazon’s commitment gives Constellation revenue certainty to seek another 20 years of operation at Calvert Cliffs, whose two reactor licenses expire in 2034 and 2036. Meta’s agreements support Vistra’s plans to seek 20-year license extensions at all three of its plants. Part of what these hyperscalers are financing, then, is keeping existing carbon-free generation online.

    This matters when reading the investment figures. Dividing Amazon’s more than $3 billion by 190 MW produces a cost per new megawatt far above Google’s more than $4.3 billion for 890 MW. The comparison is not like-for-like, because neither company has said how much of its spend goes to new output versus modernization and life extension. Investors and analysts should treat the dollar totals as program budgets, not as a price for incremental capacity.

    Generation Gets Financed; Delivery Remains Unsolved

    None of this power is dedicated to a specific campus. Amazon’s Calvert Cliffs agreement replaced a proposed on-site data center that was withdrawn amid local opposition. All of the plant’s output will continue to flow into PJM, the 13-state grid and wholesale market, and Amazon also signed a related retail supply agreement across that territory. Meta has linked its Prometheus AI supercluster in New Albany, Ohio to its nuclear procurement, but the Vistra plants have no dedicated transmission to the campus.

    This grid-supply model avoids the regulatory and community friction that has surrounded co-locating data centers at nuclear plants. In exchange, it leaves the hardest local question open: whether transmission and distribution can carry the power to where the servers are. FERC’s October 9 rejection of key elements of PJM’s large-load framework is a reminder that the rules for connecting very large loads are still unsettled.

    Timelines add a further constraint. Large uprates have to be coordinated with scheduled refueling outages, equipment procurement and NRC safety review. Google’s first added output is expected in 2028. Amazon’s new capacity is due from 2030 to 2032, and Meta’s full 433 MW is not expected until the end of 2034. For developers short of power today, uprates are a source of supply in the next decade, not a near-term fix.

    Background

    Nuclear uprates are an established practice. According to the NRC, uprates approved over the past 20 years have added about 6 GW of capacity to the U.S. grid, roughly the output of six large reactors. Reactors are licensed by thermal power, measured in megawatts thermal, so a formal uprate raises the licensed heat output. Turbine-efficiency improvements, by contrast, can raise electrical output without changing the license.

    As AI data center demand has grown, hyperscalers have signed long-term contracts for the output of existing nuclear plants, many of them in the PJM market that spans 13 states. The latest agreements extend that model. Instead of only buying the electricity existing plants already produce, Google, Amazon and Meta are now financing increases in that output.

    Sources

    Source: Big Tech’s Next Nuclear Bet: More Power From Reactors Already Running (Data Center Frontier) covers Google, Amazon and Meta agreements to finance power uprates at operating U.S. nuclear plants.

  • Kentucky’s 6.6 GW Nuclear Plan Has Land and a Design. Its AI-Scale Test Is the Buyer

    Kentucky’s 6.6 GW Nuclear Plan Has Land and a Design. Its AI-Scale Test Is the Buyer

    TL;DR · 30-second read

    The Short Version

    A company called The Nuclear Company wants to build up to six large nuclear reactors beside the Ohio River in rural Lewis County, Kentucky. The state has never had a nuclear power plant.

    If all six were built, it would be the biggest nuclear power site in the country. Site preparation is planned to start in 2028.

    Why it matters: tech companies building huge computer centers for artificial intelligence want steady, around-the-clock electricity at exactly this size. The company has the land and has chosen a reactor. It has not said who would buy the power or who would pay to build it.

    On October 21, 2026, The Nuclear Company announced Prosperity One, according to a release distributed through PR Newswire. Prosperity One is a proposed nuclear energy campus on the Ohio River in Lewis County, eastern Kentucky. It could host up to six Westinghouse AP1000 reactors totaling about 6.6 gigawatts (GW). A gigawatt is 1,000 megawatts, the scale of a large power station. If fully built, the campus would be Kentucky’s first commercial nuclear plant. The company says it would also be the largest nuclear generating complex in the United States and one of the largest in the world.

    The company says it controls the site, which is owned by lifelong Lewis County resident Sam Howard. It plans to begin site development work in 2028 and is aiming to secure the federal approvals needed to start construction on that timeline. Separately, Lewis County is applying to the Kentucky Nuclear Energy Development Authority for designation as a ‘nuclear-ready community.’

    Executive Summary

    Prosperity One is a site announcement for a multi-reactor nuclear campus, not a construction decision. The Nuclear Company has secured control of land on the Ohio River and named its reactor technology: Westinghouse’s AP1000, the large pressurized-water design now running at Georgia’s Plant Vogtle. It has also lined up visible local support. County leaders launched a nuclear-readiness initiative unanimously in July 2026, held public hearings, and this month voted by a wide margin to apply for state designation.

    The scale is what makes the announcement notable. At roughly 6.6 GW, a fully built campus would deliver firm power on the order now being discussed for the largest new electricity users, a group that increasingly includes AI data-center campuses. Firm power means electricity available around the clock regardless of weather. The company presents the project around reliable power for Kentucky, local jobs and a stronger tax base. It has not named a customer, utility partner, cost estimate or financing plan.

    The next milestones are therefore commercial and regulatory, not geographic. They include an offtake agreement (a long-term contract to buy the plant’s output), a financing structure, a licensing filing with the Nuclear Regulatory Commission, and a transmission interconnection plan. Those will show whether Prosperity One becomes a buildable project or remains a well-sited option.

    What Prosperity One Already Has

    The announcement covers the early, local layer of a nuclear project. The company has site control, and the landowner says he joined the effort so his nine grandchildren can stay in a county he describes as among Kentucky’s highest in unemployment. The county has also run a public process and voted to seek the state’s nuclear-ready designation. Many energy projects stall at the local layer. Prosperity One, by the company’s account, has cleared that step early.

    The team also carries relevant experience. Chief Nuclear Officer Joe Klecha is a former licensed senior reactor operator with 33 years in nuclear power, from US Navy submarines to operations leadership at large reactor fleets. He held construction and operational-readiness leadership roles at Vogtle Units 3 and 4, the most recent AP1000s completed in the US. The company says its services division has performed maintenance at more than 40 percent of America’s operating nuclear plants. Together these support the claim that the team knows how existing reactors are run and maintained. Executing a six-unit new build is a different and much larger undertaking.

    The early work program is a standard list for a greenfield nuclear site. It includes detailed site characterization, geotechnical surveys (testing the ground the reactors would sit on), environmental reviews, transmission interconnection planning and preparation for federal licensing. The company also says it will address transportation, housing, emergency preparedness and water use. For a campus drawing cooling water from the Ohio River, water use is a material planning issue, not a footnote.

    Why 6.6 GW Makes the Buyer the Deciding Question

    Six AP1000 units at the stated total works out to roughly 1.1 GW each. A campus of that size is beyond what a rural county or a single regional load would absorb on its own. It is the size class at which the conversation turns to the largest new loads on the US grid, and the largest new loads being planned today are AI data-center campuses measured in hundreds of megawatts up to gigawatts. That is the AI-infrastructure relevance of Prosperity One. The Nuclear Company has not linked the campus to data centers or named any prospective customer.

    Here is the mechanism that puts the buyer at the center. Reactors of this size cost far more up front than gas plants and take years to build. Lenders and investors fund them against predictable revenue over decades. That revenue can come from a regulated utility recovering costs from ratepayers, as at Vogtle, from long-term contracts with large power users, or from some combination. Whichever route applies, someone has to commit to buying the output before anyone commits capital to the concrete. The site is controlled, local support is documented, and the reactor design is already licensed and operating at Vogtle. Of the elements a nuclear project needs, the commercial commitment is the one that remains entirely undisclosed.

    Several groups are directly affected by how that question resolves. Data-center developers looking for firm, carbon-free power in the Ohio Valley gain a possible long-dated supply option, but not a near-term one. Utilities and the regional grid operator would need to plan for, and potentially build transmission around, a very large new source of supply. Kentucky ratepayers have a stake in whether any of the cost lands in regulated rates. Until a buyer appears, Prosperity One signals where large-scale nuclear could go. It is not yet capacity anyone can plan around.

    The Vogtle Credential and the 2028 Clock

    Vogtle Units 3 and 4 are the reference point for any AP1000 project in America. They proved the design can be built and operated in the US: the units entered commercial service in 2023 and 2024. They also ran years behind schedule and well over original cost estimates. Industry advocates have argued that later AP1000s should be cheaper and faster because the design is finished and the workforce has now done it once. Prosperity One’s leadership includes people who worked on Vogtle, which supports that argument. Proving it requires a contracted build, and the company has not yet described one.

    The 2028 date needs careful reading. The company says it plans to begin site development work in 2028 and aims to obtain the federal approvals needed to start construction on that timeline. It says it will engage the Nuclear Regulatory Commission ‘through established licensing processes.’ It has not stated what it has filed, which license pathway it will use, how six units would be phased, or when the first unit would produce power. For a buyer weighing the campus against near-term alternatives, the first-power date matters more than the groundbreaking date, and that date is not in the announcement.

    Gas Bridges, Small-Reactor Pilots and the Case for Large Reactors

    Large power users now have a few broad options for firm power: natural gas plants deployed as a bridge, small modular reactor (SMR) pilots that promise factory-built units in smaller increments, and large conventional reactors. Prosperity One is a clear bet on the third option. It uses a design that already has a US operating record and stacks enough units in one place to spread site, licensing and workforce costs across a lot of output.

    The trade-off is concentration. A single six-unit campus puts a great deal of capital, schedule risk and regional grid dependence in one location. SMR proponents argue that smaller units reduce exactly those risks, though at commercial scale those designs remain earlier in development. Neither path has settled the debate. What Prosperity One adds is a concrete, sited proposal for the large-reactor approach at a scale few US projects have attempted, and a test of whether that approach can attract the long-term commitments it needs.

    Background

    The United States has built very few new reactors in recent decades. Vogtle Units 3 and 4 in Georgia were the first AP1000s completed in the country. They entered service in 2023 and 2024 after significant delays and cost overruns, and they are now the reference case for anyone proposing new large reactors. Interest in nuclear power has since grown, driven by forecasts of rising electricity demand, much of it tied to data centers, and by buyers seeking firm, carbon-free supply.

    Kentucky has long been a coal and energy-producing state, and it lifted its moratorium on new nuclear construction in 2017. The state has since established the Kentucky Nuclear Energy Development Authority, which handles the nuclear-ready community designations that counties such as Lewis County are now seeking.

    Sources

    Source: Die Nuclear Company kündigt Prosperity One an, einen 6,6 Gigawatt starken Kernkraftwerkscampus in Lewis County, Kentucky: The Nuclear Company’s announcement of a proposed campus of up to six AP1000 reactors on the Ohio River.

  • Quanta Services, High-Voltage Breakers and the Grid Bottleneck

    Quanta Services, High-Voltage Breakers and the Grid Bottleneck

    TL;DR · 30-second read

    The Short Version

    Building a data center is the easy part. Connecting it to electricity is not. That needs new power lines, new substations — the yards full of equipment that step electricity down to usable voltages — and the specialist crews who build them.

    Quanta Services is one of the largest of those construction firms. It has now tied itself to an equipment maker, Hyosung HICO, in a venture making high-voltage breakers: the giant switches that cut power in a fraction of a second when something goes wrong.

    Quanta’s shares have risen sharply over five years. The debate now is whether that price already assumes a decade of building.

    Simply Wall St reported on September 6, 2026 that Quanta Services (NYSE: PWR) is back in investor focus following updates on analyst sentiment, earnings estimate revisions and the HYOSUNG HICO BREAKER joint venture, a partnership tied to grid modernization and the manufacture of high-voltage circuit breakers.

    The shares closed at US$624.41. That price followed a 3.6% gain over seven days and a 7.1% decline over 30 days, set against a one-year total shareholder return of 67.8% and a five-year total shareholder return of 443.5%. The most-followed valuation narrative on the stock puts fair value at US$710.00, implying the shares are 12.1% undervalued, while Quanta’s trailing price-to-earnings multiple of 70.7x sits more than double the 32.5x construction industry average.

    Executive Summary

    Strip away the valuation debate and the operationally interesting item is the joint venture itself. Quanta Services is a contractor: it employs the linemen, the substation electricians and the heavy-equipment operators who physically build electric transmission and distribution systems. A venture in high-voltage breaker manufacturing moves it, at least partially, from installing grid hardware to having a stake in producing it.

    That direction of travel matters because the constraint on connecting new load to the grid — whether that load is a hyperscale data center, a reshored factory or a wind farm waiting in an interconnection queue — is increasingly physical rather than financial. Capital is abundant. Skilled crews, energized substations and long-lead switchgear are not. A contractor that controls both the labor to build a substation and a share of the equipment that goes inside it is positioned at the narrowest point in that chain.

    The market has already noticed. Quanta’s five-year return of 443.5% reflects an investor base that has repriced electrical contracting from a cyclical, low-margin trade into a scarce structural asset. The open question is not whether the demand is real — utility hardening, transmission expansion and AI-driven power demand are all visible — but how much of that decade is already in the price at 70.7x earnings.

    The Binding Constraint Is Construction, Not Capital

    The public conversation about artificial intelligence infrastructure has largely been about semiconductors and, more recently, about electricity generation. The less-discussed choke point sits between them: the wires, transformers, breakers and substations that carry power from where it is generated to where it is consumed, and the finite pool of qualified crews that install them. A gigawatt of contracted generation is worthless to a data center operator if the interconnection — the physical and contractual connection to the grid — cannot be built for several years.

    Quanta’s business is precisely that middle layer. Its stated markets span electric and gas utilities, power generation, load centers, manufacturing, communications and pipelines. That breadth is the point: the same craft labor pool serves grid replacement, utility storm hardening, renewable interconnection and industrial reshoring simultaneously. When several demand waves draw on one labor pool, the contractor’s scarcity value rises independently of any single end market. This is the structural case behind the stock’s re-rating, and it does not depend on any particular AI capital-expenditure forecast holding.

    The risk embedded in that same logic is symmetry. Diversified demand cushions a downturn in any one vertical, but it does not protect against a broad deferral of utility capital programs, nor against the execution risk that comes with very large, complex fixed-price programs. Simply Wall St’s own summary flags both: a slowdown in new grid or AI-related awards, and execution setbacks on large projects.

    Moving Upstream Into the Equipment It Installs

    A high-voltage circuit breaker is the protective switch at the heart of a substation. When a fault occurs — a downed line, a short circuit — the breaker interrupts a current that would otherwise destroy transformers and start fires, and it does so in milliseconds. It is unglamorous, safety-critical, heavily engineered equipment, and the number of qualified manufacturers is small.

    For a contractor, an interest in breaker manufacturing changes the negotiating posture on schedule. Contractors are typically at the mercy of equipment lead times: the crew is mobilized, the site is ready, and the project waits on switchgear. A manufacturing venture offers, in principle, visibility into that queue and a hedge against it. It also opens a second revenue line with a different margin profile from installation work — equipment manufacturing carries capital intensity and inventory risk that labor-led contracting does not.

    Whether this venture delivers those benefits at meaningful scale is genuinely unknown from what has been disclosed. The joint venture is described in terms of its purpose — grid modernization and high-voltage breaker manufacturing — rather than its economics. Investors should treat it, for now, as directional evidence about strategy rather than as a quantifiable earnings contributor.

    Two Valuation Frameworks, Opposite Conclusions

    The stock currently supports two defensible and contradictory readings, and the release presents both. On a forward-looking narrative basis, fair value is pegged at US$710.00 against a US$624.41 close, implying 12.1% upside; the company also trades at a discount to analyst price targets. On a ratio basis, the picture inverts: a 70.7x price-to-earnings multiple compares with 32.5x for the construction industry, 33.8x for named peers and a 38.6x “fair ratio” benchmark.

    These are not really disagreements about Quanta’s business. They are disagreements about classification. A multiple near the construction industry average assumes Quanta is a contractor whose earnings will eventually revert with the building cycle. A multiple near 70x assumes it is something closer to a bottleneck asset — a firm whose scarce labor base and customer relationships let it hold pricing power across a multi-decade grid rebuild. The peer set the ratio comparison uses, including Comfort Systems USA and EMCOR Group, are themselves specialty contractors that have re-rated on similar logic, which makes the “peer average” a moving target rather than a fixed anchor.

    The practical implication is that the compression risk is arithmetic, not speculative. If sentiment cools and the multiple drifts toward the 38.6x fair ratio without a compensating jump in earnings, the required earnings growth to defend the current price is substantial. The 7.1% decline over 30 days, against a 67.8% one-year return, suggests the market is already testing that question rather than settling it.

    Background

    Quanta Services is one of the largest specialty infrastructure contractors in North America, providing engineering and construction services across electric and gas utilities, power generation, load centers, manufacturing, communications, pipelines and energy. Its core asset is not equipment but people: a large, trained and licensed craft workforce whose skills — high-voltage line work, substation construction — take years to develop and cannot be scaled quickly in response to demand.

    That workforce has become unusually valuable. Utility capital spending has risen simultaneously across grid replacement, storm hardening, renewable interconnection and, most recently, the electrical infrastructure required to serve large computing loads. The result has been a sustained re-rating of listed specialty contractors, Quanta among them, from cyclical builders toward companies valued for structural scarcity — a shift reflected in Quanta’s 443.5% five-year total shareholder return and in the valuation debate that now surrounds the stock.

    Sources

    Source: Quanta Services (PWR) Gets A Grid Venture Boost, Is The Stock Too Expensive? — Simply Wall St, September 6, 2026, reporting on Quanta Services’ HYOSUNG HICO BREAKER joint venture, recent share price performance and the competing valuation cases for the stock.