jain.com — data centers, connectivity and security, analyzed daily

Tag: Nvidia

  • AWS and NVIDIA’s 2 Million GPUs: Power Is the New Constraint

    AWS and NVIDIA’s 2 Million GPUs: Power Is the New Constraint

    NVIDIA and Amazon Web Services have announced an expanded partnership to deliver 2 million additional GPUs and next-generation infrastructure aimed at agentic AI (software that plans and executes multi-step tasks rather than just answering prompts) and physical AI (robotics, autonomous machines and industrial systems). Both companies published the news through their own newsrooms.

    The announcement lands alongside two related data points: TechCrunch reports that Amazon has tripled its order of Nvidia chips, citing “surging demand,” and the Associated Press reports that Nvidia’s second-quarter results came in well beyond Wall Street’s expectations on the strength of AI chip demand. Together they describe one buyer, one supplier, and a step-change in contracted volume.

    Executive Summary

    The headline number — 2 million GPUs — matters less for what it says about Nvidia’s order book than for what it implies about the physical plant required to land it. A GPU is a graphics processing unit: a chip built for massively parallel math, and the workhorse of AI training and inference. Two million of them is not a purchase order; it is a multi-year industrial programme that has to be matched by buildings, substations, transformers, switchgear, water or refrigerant loops, and fibre.

    Read together with Amazon’s tripled chip order and Nvidia’s Q2 beat, the pattern is a shift in how hyperscalers buy. Opportunistic, quarter-by-quarter allocation chasing has given way to committed, long-horizon supply agreements — the procurement posture of an airline ordering airframes, not a retailer restocking shelves. That change is rational when lead times on the surrounding infrastructure run longer than the lead time on the chips themselves.

    For anyone who builds, powers or cools digital infrastructure, the strategic reading is straightforward: the scarce input is migrating downstream. When silicon supply is contracted years ahead, the question that determines whether capacity actually arrives on schedule is no longer “can you get the accelerators?” but “where will you land them, what feeds them, and what carries the heat away?”

    Procurement Has Gone Industrial

    A commitment expressed in millions of units, spanning generations of hardware, behaves differently from a spot purchase. It requires the supplier to reserve foundry capacity, advanced packaging and high-bandwidth memory allocation well in advance, and it requires the buyer to commit capital before the demand it serves is fully booked. Both sides are trading flexibility for certainty — the classic structure of industrial supply contracts in aerospace, energy and heavy manufacturing.

    That framing explains why Amazon tripling its order and Nvidia beating expectations are the same story told from two ends of the same contract. The supplier’s revenue recognition and the buyer’s capital plan are now coupled over a multi-year horizon. The upside is predictability: fabs can plan, and data centre teams can sequence construction against known delivery windows. The downside is that a demand forecast, once converted into contracted volume, is expensive to be wrong about.

    It also raises the entry price for everyone else. When a large share of leading-edge accelerator output is spoken for by a handful of buyers with balance sheets to match, smaller clouds, enterprises and national programmes are not competing on price so much as on queue position — and increasingly on whether they can offer the supplier something the hyperscalers cannot.

    The Binding Constraint Moves From Silicon to the Envelope

    AI accelerators concentrate far more power into a rack than the general-purpose servers most existing data centre halls were designed around. That concentration is what forces the shift from air cooling to liquid — direct-to-chip cold plates or immersion — and what turns electrical distribution, from the utility interconnect down through transformers, switchgear and busway, into the pacing item of a build. None of that is fast. Utility interconnection studies, transformer manufacturing and high-voltage equipment orders routinely take longer than a chip generation.

    This is the practical significance of a 2-million-GPU commitment for infrastructure operators. The chips have a delivery schedule; the power envelope has a permitting, procurement and construction schedule; and the two only intersect if someone sequenced them together years earlier. Capacity that cannot be energised and cooled on time is not capacity — it is inventory.

    The physical-AI element of the announcement adds a second dimension. Robotics and autonomous systems generate inference demand at the edge and in regional facilities, not only in a handful of mega-campuses. If that materialises at scale, it argues for distributed, latency-sensitive capacity in metros — a different real-estate and connectivity problem from the remote gigawatt campus, and one where existing colocation footprints and dense fibre routes have a genuine structural advantage.

    Who Benefits, and Where the Risk Sits

    The clearest beneficiaries beyond the two named parties are the suppliers of the envelope: power developers and independent producers, electrical equipment manufacturers, liquid-cooling vendors, mechanical and electrical contractors, and colocation operators with energised, high-density-ready shells. Scarcity in those categories is not a temporary shortage caused by one deal; it is a structural mismatch between how quickly chips can be fabricated and how slowly grid infrastructure can be built.

    The risk is concentration and timing. A programme sized in millions of units assumes sustained demand for agentic and physical AI workloads that are, today, earlier in commercial adoption than large language model inference. If adoption arrives more slowly than the delivery schedule, the exposure is not primarily in the chips — which can be redeployed to other workloads — but in the long-lived, single-purpose assets built to host them, and in the power contracts signed to feed them.

    For enterprise buyers, the near-term implication is capacity planning, not panic. More contracted supply should, over time, ease the availability constraints that have shaped GPU cloud pricing. But it will not ease them uniformly: availability will follow where power and cooling land first, which makes region selection, interconnection and committed-use terms more consequential in procurement than headline instance pricing.

    What These Announcements Do and Do Not Substantiate

    It is worth being precise about the evidentiary base. What is on the record is a stated intent to deliver 2 million additional GPUs and next-generation infrastructure, a reported tripling of Amazon’s chip order attributed to surging demand, and a quarterly result that exceeded analyst expectations. Those are meaningful, and the financial result in particular is an audited, externally verifiable data point rather than a marketing claim.

    What is not established by these announcements is the delivery schedule, the capital commitment, the split between training and inference capacity, the regions involved, or the power procurement behind them. “Additional” is doing real work in the headline and is not defined against a stated baseline. A vendor-and-customer joint announcement is, by construction, the parties’ own account of their arrangement; it is a statement of direction, not a disclosure document.

    None of this makes the announcement thin — the direction it signals is consistent with the independently reported financial results. But the useful posture for infrastructure planners is to treat the 2-million figure as a demand signal for power, cooling and land, and to wait for filings, permit applications, interconnection queue entries and utility disclosures for the details that determine when and where the capacity actually appears.

    Background

    NVIDIA designs the GPUs and accompanying networking and software that underpin most large-scale AI training and a growing share of inference. Amazon Web Services is the largest public cloud provider and has long combined third-party accelerators with silicon of its own design. The two have partnered on AI infrastructure for years; this announcement extends that relationship rather than establishing it.

    The context is a multi-year build-out in which cloud providers have committed unprecedented capital to AI capacity. Early in that cycle, the scarce resource was the accelerators themselves, and access to allocation was a competitive differentiator. As supply agreements have lengthened and volumes have grown, attention across the infrastructure industry has moved to the constraints that cannot be solved by a purchase order: grid capacity, interconnection queues, long-lead electrical equipment, and the retrofit or replacement of facilities designed for a lower power density than AI hardware demands.

    Source: Strong AI chip demand fuels Nvidia’s Q2 results well beyond Wall Street’s expectations — AP News reporting on Nvidia’s quarterly results, read alongside the AWS–NVIDIA announcement of 2 million additional GPUs and reports of Amazon tripling its chip order.

  • NVIDIA Takes Minority Stake in Cloverleaf Infrastructure to Speed AI Factory Sites

    NVIDIA Takes Minority Stake in Cloverleaf Infrastructure to Speed AI Factory Sites

    Cloverleaf Infrastructure, a Houston-based data center site developer founded in 2024, announced on August 21, 2026 a strategic partnership with NVIDIA that includes a minority equity investment from the chipmaker. The investment amount was not disclosed.

    Under the partnership, Cloverleaf will apply the NVIDIA DSX platform to integrate site, power, cooling, computing, and facility decisions earlier in the design phase, and Cloverleaf customers will gain access to NVIDIA’s full AI factory stack. The company says it has delivered multiple gigawatt-scale projects across North America since its founding.

    Executive Summary

    The world’s dominant AI chip supplier just bought a piece of a company that doesn’t make chips, servers, or software — it develops land, power, and grid connections. NVIDIA’s minority investment in Cloverleaf Infrastructure, announced jointly from Santa Clara and Houston, is framed by both companies as a way to accelerate the buildout of “AI factories,” the industry’s term for data centers purpose-built to train and run artificial intelligence models at industrial scale.

    The logic is stated plainly in the release itself: “land, power and shell are their foundation,” in the words of NVIDIA vice president Nico Caprez. Access to powered, shovel-ready sites — parcels that already have utility-scale electricity secured and permits in hand — has become the pacing constraint on how fast new AI computing capacity can come online. By taking an equity position in a site developer, NVIDIA is extending its reach beyond the server rack and down into the physical and electrical foundations of the industry it supplies.

    What the announcement does not include is as notable as what it does: no investment figure, no named customers, no specific sites, and no committed capacity or timelines. It is a directional signal backed by real money of undisclosed size, and it should be read that way.

    NVIDIA Keeps Reaching Further Down the Stack

    NVIDIA’s core business is selling GPUs — the specialized processors that power AI training and inference. But a GPU generates no revenue sitting in a warehouse; it needs a building, a cooling system, and above all a grid connection capable of delivering tens or hundreds of megawatts. This deal shows NVIDIA working to de-bottleneck its own demand pipeline: every powered site Cloverleaf brings to market faster is a site that can absorb NVIDIA hardware sooner. The release makes the linkage explicit, noting that Cloverleaf customers “will be able to engage with NVIDIA across the full AI factory stack,” from accelerated computing and networking down through infrastructure software.

    There is a coherent strategic pattern here. A chip vendor that influences site selection, power procurement, and facility design early in a project’s life is well positioned to shape what gets deployed inside that facility later. That is not sinister — vertical coordination is common when supply chains strain — but it does mean the partnership serves NVIDIA’s commercial interests as much as Cloverleaf’s, and prospective customers should evaluate the integrated offering on its merits rather than its branding.

    Powered Land Is the New Scarce Resource

    For most of the cloud era, the binding constraint on data center growth was capital or construction labor. Today it is increasingly electricity — specifically, the interconnection process by which a new large load gets permission and physical equipment to draw power from the grid. Utility interconnection studies, transmission upgrades, and substation construction can take years, which is why a “shovel-ready” site with power already secured commands a premium. Cloverleaf’s entire business model, per its own description, is partnering with utilities and energy innovators to deliver exactly those sites.

    Seen through that lens, NVIDIA’s investment is a bet that site development — not silicon supply — is where AI capacity growth will be won or lost over the next several years. It also validates the developer category itself: Cloverleaf was formed only in 2024, with initial backing from Sandbrook Capital and NGP Energy Capital, and claims multiple gigawatt-scale project deliveries already. If the claim holds up, that is a remarkably fast ramp; the release, however, offers no project names, locations, or customer identities against which to check it.

    What DSX Integration Actually Changes

    The operational substance of the partnership is Cloverleaf’s adoption of the NVIDIA DSX platform, which the release describes as bringing “site, power, cooling, computing and facility decisions together earlier in the design phase.” In plain terms: instead of designing a building first and figuring out later what computing it can support, developers would co-optimize the facility and the hardware from the start, evaluating tradeoffs against available power, water, and grid capacity. Once a facility is running, DSX software is pitched as helping operators squeeze more useful AI output from every megawatt.

    If it works as described, this addresses a genuine industry pain point — AI-era facilities differ radically from traditional data centers in power density and cooling, and retrofitting mismatched designs is expensive. But the release offers no performance data, deployment examples, or quantified efficiency gains for DSX at Cloverleaf sites, so the benefit remains a stated intention rather than a demonstrated result. Buyers should also weigh whether design-phase integration with one vendor’s platform preserves flexibility to deploy other vendors’ hardware later; the release does not address exclusivity in either direction.

    Winners, Losers, and Open Questions for the Market

    The clearest winner is Cloverleaf, which gains capital, the credibility of NVIDIA’s endorsement, and a channel to customers making multi-billion-dollar deployment decisions. Its private equity backers gain a marquee validation event. Utilities partnered with Cloverleaf may benefit from better-engineered load forecasts. Competing site developers and master-planned data center campus firms now face a rival with privileged access to the industry’s most important technology supplier.

    The unresolved question is what this consolidation of influence means for the broader ecosystem. When the dominant chip supplier holds equity positions across the infrastructure chain, the industry gains coordination speed but concentrates dependency on a single vendor’s roadmap. That tradeoff has served fast-growing industries well in some eras and poorly in others — and with no disclosed deal terms, outside observers cannot yet judge how much influence this particular investment buys.

    Background

    Cloverleaf Infrastructure is a young company in an old-fashioned business: assembling land, permits, and — critically — electric power for others to build on. Formed in Houston in 2024 with backing from Sandbrook Capital and NGP Energy Capital, it targets the pinch point of the AI buildout, where demand for computing capacity has outrun the grid’s ability to connect new large loads quickly. Its customers are the technology companies that construct and operate data centers, the facilities behind the internet, cloud services, and AI.

    NVIDIA, headquartered in Santa Clara, California, is the dominant supplier of the GPUs that power modern AI, and has increasingly involved itself in the layers surrounding its chips — networking, software platforms, and now, through this investment, the land-and-power development stage where AI facilities begin. The partnership reflects a broader industry shift: as AI computing scales, electricity availability and site readiness, rather than chip supply alone, increasingly determine how fast new capacity comes online.

    Source: Cloverleaf Infrastructure Forms Strategic Partnership with NVIDIA to Accelerate Data Center Infrastructure Development — PR Newswire release of August 21, 2026 announcing NVIDIA’s minority investment in the Houston-based data center site developer.

  • OpenAI’s 8 GW Ohio Campus: Legacy Grid Covers 800 MW, New Power Plants Gate the Rest

    OpenAI’s 8 GW Ohio Campus: Legacy Grid Covers 800 MW, New Power Plants Gate the Rest

    TL;DR · 30-second read

    The Short Version

    • OpenAI, the maker of ChatGPT, will rent space in a huge computer campus planned on the grounds of an old Cold War uranium plant in rural southern Ohio.
    • At full size, the campus would use about as much electricity as eight large nuclear reactors produce.
    • Only about a tenth of that can come from today’s power lines, starting in 2028. The rest needs new power plants, including gas-fired ones, that have not been built yet.
    • The developer promises to pay for the grid upgrades itself, so local electric bills do not cover them.

    OpenAI announced on August 17 that it has agreed to secure approximately 8 gigawatts of IT capacity at the PORTS-Pike Technology Campus in Pike County, Ohio, working with developer SB Energy, NVIDIA and the U.S. Department of Energy. The campus is being developed on private land and on remediated federal land at the former Portsmouth Gaseous Diffusion Plant. SB Energy will build, own and operate the data center under a 20-year lease to OpenAI, and the site will exclusively host NVIDIA AI computing hardware.

    The first 800 megawatts are expected to be available in 2028, largely using existing AEP infrastructure. Later phases will require new grid-connected power plants, including natural gas generation, plus new transmission lines. NVIDIA will invest $1.5 billion in SB Energy and provide credit support for the land, power and shell buildout of the initial 4.25 gigawatts. The companies project 35,000 construction jobs over a six-year buildout through 2032 and 2,500 long-term operating jobs.

    Executive Summary

    The PORTS-Pike agreement is one of the largest single-site AI capacity commitments announced to date: roughly 8 gigawatts of power delivered to computing equipment, on a former uranium enrichment site in southern Ohio. OpenAI is the tenant, SB Energy is the developer and owner, NVIDIA is both the exclusive hardware supplier and a financial backer, and the Department of Energy is contributing federal land and an existing onsite water system.

    The more telling number is the smaller one. Only the first 800 megawatts, about a tenth of the headline figure, is expected to come online in 2028 on largely existing grid infrastructure. Everything after that depends on new power plants and new transmission lines, along with permits, environmental reviews and financing that the companies say must still be in place. In practice, this is a power-development project with a data center attached, and its pace will be set by generation and wires rather than by chips.

    The package also includes a pledge that SB Energy will pay the full cost of grid upgrades and new transmission rather than passing them to Ohio ratepayers, a closed-loop cooling design meant to limit ongoing water use, and more than $160 million in community grants and student software credits.

    Nine-Tenths of the Campus Waits on Power That Does Not Yet Exist

    The headline figure is 8 gigawatts-IT, a measure of the electricity delivered to servers and networking gear rather than the site’s total draw, which is higher once cooling and power conversion are counted. Against that, the companies say the first 800 megawatts should be available in 2028 “largely using existing AEP infrastructure,” meaning the transmission and substation capacity American Electric Power already has in the area. That is roughly 10% of the campus. The remaining 7 gigawatts or so, per OpenAI’s announcement, “will require new power plants connected to the grid, including natural gas generation, as well as new transmission lines and associated infrastructure.”

    That split is the operational core of the deal. Existing wires are the fastest path to power for any large data center, because new generation and new high-voltage lines each require siting, permitting, interconnection studies and equipment procurement before a single rack can be energised. Gas turbines, transformers and high-voltage switchgear have all been in heavy demand as utilities and data center developers compete for the same supply chains. The six-year buildout window through 2032 signals that the companies expect most of the capacity to arrive only as that new infrastructure is completed.

    The parties most exposed to this sequencing are OpenAI, whose compute roadmap depends on the later phases; SB Energy, which carries the development risk; AEP and PJM, the regional grid operator covering Ohio, which must study and integrate the new plants and lines; and NVIDIA, whose hardware sales at the site scale with each tranche of power that comes online.

    Why a Former Uranium Plant Works as a Power-First AI Site

    The Portsmouth Gaseous Diffusion Plant was built in the 1950s to enrich uranium, a process that consumed enormous amounts of electricity. That history left the area with heavy grid infrastructure, a large industrial footprint and federal land that has been under cleanup for years. For an AI campus, those legacy assets matter more than proximity to cities or fiber hubs: they are what allows the first 800 megawatts to arrive in 2028 instead of later.

    The Department of Energy’s role extends beyond land. PORTS-Pike is working with DOE to use the site’s existing onsite water system, and the data center will use closed-loop, air-cooled systems that recirculate water rather than evaporative cooling towers, which continuously consume water. OpenAI says ongoing water use, once the system is filled, should be comparable to an office building supporting a similar number of people, and well below what the enrichment plant historically used. It has committed to publish expected water use once the design is final.

    The model is replicable only where similar conditions exist: large parcels of already-disturbed land, legacy high-voltage infrastructure, and a federal owner willing to redevelop. Retired industrial and federal sites with those traits are a finite set, which is why they are becoming contested real estate for AI developers.

    Who Carries the Financial Risk

    The deal is structured to keep capital off OpenAI’s balance sheet up front. SB Energy builds and owns the facility, and OpenAI begins paying only as completed capacity becomes available for lease, funding those payments from revenue, cash flow and investor capital. That protects OpenAI from paying for capacity that is delayed, but it shifts construction and power-development risk to SB Energy and its financiers.

    NVIDIA fills part of that gap. Its $1.5 billion investment in SB Energy and its credit support, a guarantee-like backing that lowers borrowing costs, cover land, power and shell buildout for the initial 4.25 gigawatts. Because the site will exclusively host NVIDIA hardware, the chipmaker is effectively helping finance demand for its own products. That aligns incentives among the three companies, but it also concentrates exposure: the project’s financing, customer and supplier are tightly linked, and the backing described covers just over half of the 8 gigawatts.

    The Ratepayer Pledge Addresses the Biggest Local Objection

    Data center growth has become a flashpoint over whether households end up paying for grid expansion that serves large corporate loads. SB Energy’s commitment to pay the full cost of grid upgrades and new transmission lines, without shifting those costs to Ohio ratepayers, speaks directly to that concern, and the companies argue the new lines will also strengthen regional reliability.

    The community package, which includes $40 million from OpenAI on top of SB Energy’s earlier $40 million and $84 million in Codex credits for about 844,000 Ohio students, adds up to more than $160 million by OpenAI’s count. Alongside a union memorandum of understanding and annual public reporting on hiring, water and energy use, it is designed to build local consent for a project whose largest phases will require new permits and reviews over the coming years.

    Background

    The Portsmouth Gaseous Diffusion Plant near Piketon, Ohio, was built in the 1950s to enrich uranium for national defense and later for nuclear power fuel. Gaseous diffusion was extremely electricity-intensive, which left the region with substantial grid infrastructure. Enrichment ended in the early 2000s, and the Department of Energy has since overseen decontamination and cleanup, with parts of the site now available for redevelopment.

    SB Energy, a SoftBank-backed energy and infrastructure developer, is leading development of the PORTS-Pike campus. OpenAI has been expanding its computing capacity through long-term agreements with developers and hardware partners to support training of frontier AI models and growing demand for products such as ChatGPT and Codex.

    Sources

    Source: OpenAI joins PORTS-Pike project, OpenAI’s announcement of an approximately 8 gigawatt capacity agreement with SB Energy, NVIDIA and the U.S. Department of Energy in Pike County, Ohio.

  • Nvidia’s 4.25-GW Ohio Guarantee Puts Chipmaker Credit Behind AI Land and Power

    Nvidia’s 4.25-GW Ohio Guarantee Puts Chipmaker Credit Behind AI Land and Power

    TL;DR · 30-second read

    The Short Version

    Nvidia, the company whose chips run most of today’s artificial intelligence, has agreed to financially vouch for the land, power and buildings of a huge computing campus in southern Ohio. The campus is on the site of a closed Cold War uranium plant.

    In return, only Nvidia’s equipment can go inside. OpenAI, the maker of ChatGPT, has signed a 20-year lease to use it.

    At full size, the computers would use roughly the output of eight large nuclear reactors. The first sections are due in 2028. The unusual part is the financing: a chip seller is helping pay for the buildings its chips will sit in.

    Nvidia announced on August 17, 2026 that it will provide credit support for the land, power and shell buildout at SB Energy’s PORTS-Pike Technology Campus in Pike County, Ohio. In this context, credit support means Nvidia’s financial backing stands behind the project’s obligations. The support secures an initial 4.25 gigawatts of IT capacity, meaning power delivered to computing equipment, and Nvidia holds an option on the remaining 3.75 IT-GW. In exchange, Nvidia will be the exclusive AI compute provider on the site. OpenAI will be the customer for the full 8 IT-GW. SB Energy will build, own and operate the data center under a 20-year lease to OpenAI.

    Nvidia will also invest $1.5 billion in SB Energy, joining existing investors SoftBank Group and OpenAI. SB Energy and SoftBank committed to build at least 10 gigawatts of new generation and to invest at least $4.2 billion in regional grid infrastructure with AEP Ohio. Capacity is expected to come online in phases beginning in 2028.

    Executive Summary

    The headline numbers are 8 gigawatts of AI capacity, a 20-year lease and a $1.5 billion equity check. The structural news is who is standing behind the real estate. Data center developers typically finance land, power hookups and building shells against the creditworthiness of the tenant. At PORTS-Pike, Nvidia is supplying credit support for the first 4.25 IT-GW. That makes the chipmaker a financial backer of the physical site as well as the vendor of the hardware inside it.

    What Nvidia gets in return is explicit: exclusivity. Every AI system on the campus will be Nvidia’s, deployed on its DSX AI factory platform of GPUs, CPUs and networking. Jensen Huang framed it as securing “long-lived infrastructure for NVIDIA compute” that can be upgraded with each new chip generation. That implies a site designed to take several future Nvidia generations over the life of the lease.

    The deal matters beyond Ohio for a simple reason. If a chip supplier’s balance sheet can make a gigawatt-scale site financeable, then access to land, power and shells becomes something hardware vendors compete for directly, not just something their customers arrange.

    Why a Chipmaker Is Backing Buildings

    A data center campus has three expensive layers that come before any chip is installed: land, power (generation, substations and grid connections) and the shell, meaning the building, cooling plant and electrical rooms. The industry abbreviates these as LPS. Whoever finances that layer wants confidence that someone creditworthy will keep paying for the space for the life of the investment. At PORTS-Pike, SB Energy owns the asset and OpenAI is the 20-year tenant. Nvidia is adding a third layer of assurance: credit support on the LPS buildout covering the first 4.25 IT-GW.

    That is the mechanism behind the headline. Nvidia is using its own credit to reduce the financing risk of a site it does not own. In return, it gets a guaranteed home for its hardware at a scale of more than four gigawatts, with an option on 3.75 more. Nvidia’s statement makes the logic explicit: “land, power and shell have become vital in the age of AI.” A supplier that is worried about customers finding buildings and electricity for its systems fast enough is now helping underwrite the buildings and electricity itself.

    Several groups are affected. Developers gain a new kind of counterparty whose backing may unlock project finance. Lenders get another large name standing behind the obligations. Competing accelerator vendors face a campus that is closed to them by contract. For OpenAI, the arrangement ties a very large block of future capacity to a single supplier’s roadmap. This is one deal, not yet a market pattern. It does show that chipmaker credit can serve as a financing input for AI sites, and other developers will take note.

    Exclusivity, Upgrades and Concentration

    The exclusivity clause is the price of the guarantee, and it cuts both ways. Nvidia describes the site as upgradeable “with each new generation delivering more intelligence and better economics.” A 20-year lease spans many hardware cycles, so the value to Nvidia is not a single sale. It is a standing claim on refresh after refresh. For OpenAI, a site built around one vendor’s full stack can simplify design and speed deployment. It also narrows negotiating leverage over future generations at this location.

    A fair question is how tightly interlinked the parties now are. SoftBank Group and OpenAI were already investors in SB Energy. Nvidia is now investing $1.5 billion and backing the site’s buildout. OpenAI is the tenant and will run Nvidia hardware. Arrangements like this are not unusual in early infrastructure markets, and nothing in the announcement suggests anything improper. Still, it means the campus’s economics rest on a small group of counterparties whose fortunes are correlated with continued AI demand. The announcement asserts that demand is “growing at an extraordinary pace.” Investors will want to see it in contracted revenue as phases come online.

    Power-First, With Overhead Built In

    SB Energy and SoftBank committed to at least 10 gigawatts of new generation to support 8 IT-GW of AI capacity. The gap between those numbers is instructive. Roughly a fifth of the generation is not delivered to computing equipment. It covers cooling, power conversion and other facility overhead, plus any margin the developers are holding. That is a useful reminder that AI capacity figures quoted in IT-GW understate the electricity a campus actually requires.

    The grid commitment is at least $4.2 billion in regional infrastructure through a partnership with AEP Ohio that the companies say is “designed to protect ratepayers.” The release does not explain how that protection works: who pays for which upgrades, and what happens to those costs if demand falls short. Those questions matter to Ohio households as much as to investors. Reusing the decommissioned Portsmouth Gaseous Diffusion Plant site, with the Department of Energy and Department of Commerce involved, may offer existing industrial land and transmission advantages. Spanning federal and private land also adds approval steps that the companies have not detailed.

    Background

    SB Energy describes itself as an integrated data center and power infrastructure company that develops, builds and operates gigawatt-scale data center campuses and utility-scale power generation. Its investors before this deal included SoftBank Group and OpenAI. Its pitch is a “power-first” model: securing electricity before building compute, on the view that power is the industry’s primary bottleneck.

    The PORTS-Pike site sits on the former Portsmouth Gaseous Diffusion Plant in Appalachian Ohio. The plant was a Cold War-era uranium enrichment facility that has since been decommissioned. Redeveloping it as an AI campus places some of the country’s largest planned computing capacity in a region that SB Energy says has long shaped American industry, and it involves both federal and private land.

    Sources

    Source: NVIDIA Guarantees SB Energy’s PORTS-Pike Technology Campus in Ohio to Exclusively Host NVIDIA AI Compute. Nvidia Newsroom, August 17, 2026: Nvidia’s announcement of credit support, exclusivity and a $1.5 billion investment in SB Energy for OpenAI’s 8 IT-GW Ohio campus.

  • IBM’s $240M Together AI Deal Treats AI Inference Like Reserved Capacity

    IBM’s $240M Together AI Deal Treats AI Inference Like Reserved Capacity

    TL;DR · 30-second read

    The Short Version

    IBM, the century-old computing company, has signed a $240 million, multi-year deal with Together AI, a younger firm that runs freely available artificial intelligence programs for other businesses.

    IBM’s cloud, meaning its rentable computers in data centers, will host a large set of powerful Nvidia chips set aside for one job: answering people’s questions to those programs, the everyday work of a chatbot.

    The surprising part is how it is bought. Instead of paying per use, the computing power is booked years ahead, like leasing a whole building instead of renting hotel rooms by the night.

    IBM and Together AI have signed a $240 million multi-year agreement for an Nvidia-powered AI inference cluster that will run on IBM Cloud, Reuters reported on August 11, 2026. Inference is the stage where a trained AI model answers requests, as opposed to training, where the model is built.

    Together AI operates a cloud platform that serves open-weight AI models, meaning models whose parameters are publicly released, to developers and enterprises. Under the deal, a dedicated cluster of Nvidia accelerators on IBM’s infrastructure is committed to that inference work for several years.

    Executive Summary

    The headline number is $240 million, but the structure matters more. This is not a customer swiping a card for GPU hours. It is a multi-year commitment to a specific, dedicated cluster, which is how data center space and power have long been contracted, and increasingly how AI compute is contracted too.

    For IBM, the deal places IBM Cloud in the business of hosting third-party AI inference at scale, a market where the large hyperscalers and specialist GPU clouds have been most visible. For Together AI, it secures a block of capacity for serving open models at a time when access to Nvidia hardware still depends heavily on who has committed to it in advance.

    The companies have not disclosed the hardware generation, the size of the cluster, where it sits, or how the $240 million is paid over time. Those details determine whether this is a modest capacity reservation or a meaningful new source of inference supply.

    Inference Is Being Booked Like a Lease

    Most people meet AI inference as a metered service: a developer sends a prompt, a provider returns an answer, and the bill is calculated per token, a token being a small chunk of text. That pricing looks elastic, but the hardware underneath it is not. Nvidia accelerator clusters are expensive to buy, take time to deliver and install, and lose value as each new chip generation arrives. Someone has to commit to them before the first token is served.

    A $240 million, multi-year contract for a dedicated cluster is that commitment made explicit. It shifts inference from the on-demand model of cloud computing toward the reserved-capacity model long familiar from colocation and wholesale data center leasing, where a tenant contracts a defined block of infrastructure for a defined term. The provider gets revenue visibility that justifies deploying hardware; the tenant gets guaranteed supply it can resell to its own customers with predictable performance.

    This is one deal, and it does not prove an industry-wide shift on its own. But it shows clearly that open-model inference, often described as a commodity service racing to the lowest per-token price, is being underwritten by the same kind of long-dated capacity contracts as training clusters. The per-token price a developer sees sits on top of a fixed, multi-year obligation somewhere in the stack.

    What IBM Gains by Hosting Someone Else’s Models

    IBM has spent recent years positioning around open and hybrid AI: its Granite models are openly released, and its Red Hat business is built on open-source software. Hosting a platform whose entire proposition is serving open-weight models fits that posture more naturally than competing head-on with the largest clouds for frontier-lab training contracts.

    It also gives IBM Cloud a substantial, recurring AI workload with an anchor tenant. For a cloud provider, a committed customer reduces the risk of buying accelerators that might otherwise sit underused. For enterprise buyers who already run on IBM infrastructure, the arrangement could, in principle, place open-model inference closer to their existing systems, though neither company has described how enterprise customers would access the cluster.

    The Risk Sits in the Utilisation Curve

    Reserved capacity moves risk; it does not remove it. Inference demand can shift quickly as new models arrive, and per-token prices for open models have been falling as providers compete. A multi-year commitment to a fixed pool of hardware exposes whichever party carries it to two pressures: whether the cluster stays busy, and whether the hardware stays competitive as Nvidia releases newer, more efficient generations.

    There is also a physical side. Dense Nvidia clusters draw significant power and increasingly require liquid cooling, which constrains which data halls can host them and how quickly. The value of a contract like this depends on the cluster actually being energised and in service on schedule, and none of those milestones have been published.

    Background

    IBM has operated a public cloud for more than a decade and in recent years has concentrated its AI strategy on enterprise and hybrid deployments, including its watsonx platform, its openly released Granite models and the open-source software business of Red Hat. IBM Cloud offers Nvidia GPU infrastructure alongside its traditional enterprise services.

    Together AI emerged during the generative AI boom as a platform for running open-weight models, giving developers an alternative to closed models from the largest AI labs. Like other inference providers, it depends on securing large volumes of Nvidia accelerators, whose supply has been shaped by long-term commitments from cloud providers and AI companies.

    Sources

    Source: IBM, Together AI ink $240 million deal for Nvidia-powered AI inference cluster (Reuters): report on a multi-year agreement for a dedicated Nvidia inference cluster on IBM Cloud.

  • Anthropic Joins 5 Custom-Chip Builders: Nvidia’s Exposure Is Inference, Not Training

    Anthropic Joins 5 Custom-Chip Builders: Nvidia’s Exposure Is Inference, Not Training

    TL;DR · 30-second read

    The Short Version

    Anthropic, the company behind the Claude chatbot, is starting to design its own computer chips for running its AI. Samsung is reportedly in talks to build them.

    Today most AI runs on chips from Nvidia, which are expensive and use a lot of electricity. Google, Amazon, Meta, Microsoft and OpenAI already make their own chips.

    What most people miss is that these custom chips are built to answer people’s questions, not to create new AI models. Creating new models still mostly happens on Nvidia hardware. Nvidia’s rivals are winning ground on the everyday work of answering requests.

    Anthropic is building an in-house chip development team to co-design custom ASICs (application-specific integrated circuits, chips built for one kind of job) to run its Claude models. The company told Business Insider it is co-designing the chips so that Claude can run faster and more efficiently at the scale its customers need. It is hiring engineers now, and the job listing stresses delivering the design on schedule. Tom’s Hardware reported the development on August 7, 2026. It added that The Information had reported in July that Anthropic was in talks with Samsung to manufacture the chips.

    Anthropic has not named its design partner. No chip specifications, production volumes or deployment dates have been disclosed.

    Executive Summary

    Anthropic is the latest frontier AI developer to pursue its own silicon. Google, Amazon, Meta, Microsoft and OpenAI already have custom chip programs. The chips are aimed at inference, which is the work of running a trained model to answer requests. They are not aimed at training, the far more compute-intensive process of building the model in the first place.

    That distinction matters more than the headline rivalry with Nvidia. Training advanced models still depends heavily on Nvidia GPUs. Inference can run on a much wider range of hardware, and its cost recurs with every prompt a customer sends. For a company whose usage is growing quickly, especially from agentic AI that consumes many more tokens per task than a single chat exchange, inference is the line item that scales. It is therefore the one worth engineering down.

    The reported Samsung link adds a supply-chain dimension. Samsung is a small player in custom chip co-design, but it brings manufacturing capacity and memory. Memory is one of the scarcest inputs in AI hardware right now.

    Why Nvidia’s Exposure Sits in Inference, Not Training

    Every custom AI chip program disclosed so far aims first at running models, not building them. The economics explain why. Training a frontier model is an episodic, enormous job, and Nvidia’s GPUs and software ecosystem remain the default for it. Inference is continuous. Each question a user asks, and each step an AI agent takes, consumes compute. As usage grows, inference spending grows with it. Agentic workloads multiply the token count per task, so the growth is faster still.

    Inference also tolerates specialization. A chip that only needs to execute a known family of models efficiently can drop much of a general-purpose GPU’s flexibility. In exchange it gains lower power draw and lower cost per request. Estimates put custom silicon’s total cost of ownership advantage at up to 65% compared with Nvidia GPUs. That figure is a ceiling across the industry, not a disclosed number for Anthropic’s chip, which has no published specifications. Anthropic’s own stated goal is narrower and consistent with this reasoning: making Claude run faster and more efficiently at customer scale.

    For Nvidia, this means the pressure is concentrated in inference deployments. Each custom chip installed for inference displaces a GPU that would otherwise have done that work. Training, where Nvidia’s position is strongest, is not what these programs target. The large AI labs are also unlikely to become fully self-sufficient. Demand is growing too fast, and Nvidia retains strong control over access to the supply chain. The realistic outcome is mixed fleets, with GPUs for training and flexible workloads and in-house chips for high-volume serving.

    The Real Cost Is the Software, and the Years

    Designing a chip is only the visible part of the effort. A custom accelerator needs a full software stack, including compilers, runtimes and kernels, before it can run production models. The models themselves need tuning to exploit the hardware. That burden is why most companies sensibly buy general-purpose GPUs. Custom silicon only pays back at very large scale, where small per-request savings multiply across billions of requests.

    The reference point is sobering. Google has built its Tensor Processing Units for 12 years, with Broadcom as co-design partner on each generation. Amazon runs parallel Trainium and Inferentia lines. Meta has announced several MTIA designs for deployment through 2027. Anthropic is at the hiring stage, and its first chip will arrive into a market where rivals are several generations in. The job listing’s emphasis on schedule suggests the company understands this, but no delivery target has been disclosed.

    Samsung, Memory and a Concentrated Co-Design Market

    Custom AI chips are rarely designed alone. Broadcom and Marvell account for roughly 95% of the ASIC co-design market. Broadcom works with Google, OpenAI and Meta, as well as ByteDance and Fujitsu. It claims a $73 billion order backlog and expects more than $100 billion in annual AI chip revenue by the end of 2027. Marvell supports Amazon’s Trainium and Microsoft’s Maia and is expected to earn upward of $11 billion from co-design work in 2026. Whichever partner Anthropic has chosen for design, that partner is serving a crowded order book.

    Samsung’s reported role is manufacturing, and its appeal is not mainly design share. Samsung would bring fabrication capacity and chip design expertise. It would also bring access to memory, which is in short global supply. Modern AI accelerators depend on high-bandwidth memory (HBM) stacked alongside the processor. That memory, along with advanced packaging such as TSMC’s CoWoS, has been a practical limit on how many AI chips the industry can ship. A manufacturing partner that also makes memory could ease one of the tightest constraints a new chip program faces. Neither company has confirmed such an arrangement.

    What Changes for Data Center Operators

    For the facilities that host this hardware, custom inference chips add variety to planning. GPUs are power-hungry, and inference ASICs are built specifically to cut power per request. A fleet that mixes Nvidia GPUs with several in-house accelerators could bring different rack densities, cooling needs and networking layouts to the same campus. Operators serving the large AI labs should expect less standardization over the next few hardware cycles, not more.

    Background

    Anthropic develops the Claude family of AI models and sells access to consumers, businesses and government customers. It has grown quickly over the past year, and agentic AI, where models carry out multi-step tasks, has sharply increased the volume of computation each customer uses. Like other frontier labs, Anthropic runs its models on hardware from several suppliers, with Nvidia GPUs the industry’s default.

    Custom AI silicon has become standard practice among the largest AI companies and cloud providers, often called hyperscalers because of the scale of their data center fleets. The market for designing these chips is dominated by Broadcom and Marvell. Much of the manufacturing and advanced packaging runs through TSMC, and memory supply has become a key constraint.

    Sources

    Source: Anthropic co-designing custom AI inference chips to bypass costly Nvidia GPUs — Samsung reported as manufacturing partner for Claude maker (Tom’s Hardware): Tom’s Hardware’s report on Anthropic’s in-house inference chip team and its reported manufacturing talks with Samsung.

  • Nvidia’s 16 Zayo Routes Show New AI Fiber Gets Built Around an Anchor Tenant

    Nvidia’s 16 Zayo Routes Show New AI Fiber Gets Built Around an Anchor Tenant

    TL;DR · 30-second read

    The Short Version

    Nvidia, the company whose chips run most of today’s artificial intelligence, is the main customer for 16 new fiber-optic cable routes that Zayo is building across the United States.

    These cables are bundles of glass strands buried underground that carry data between cities. Zayo is laying them where there is enough electricity for the giant computer warehouses that AI needs.

    A rival network company says new long-distance cable usually loses money. Zayo’s answer was to sign a very large customer before digging. Neither company has said what it costs.

    Zayo Group has confirmed that Nvidia is the anchor tenant behind the “monster deal” it first described in April, when chief product and strategy officer Bill Long said the company was building 8,000 more route miles of fiber. Zayo executives told Fierce Network on August 7 that, in partnership with Nvidia, Zayo is building 16 new long-haul fiber routes in high-growth AI corridors across the U.S.: six entirely new (greenfield) routes and ten overbuilds that add fiber along existing Zayo paths. Nvidia is also leasing fiber on eight other Zayo long-haul routes that already had enough capacity.

    The program runs for several years, with the first routes expected as early as 2027 and the full set by 2030. Zayo declined to disclose the cost of the build or how much Nvidia is investing, saying only that it is “good business with an attractive return profile.”

    Executive Summary

    The world’s dominant AI chip supplier has become the named anchor customer for one of the largest new long-haul fiber builds in the United States. Long-haul fiber is the cable that links cities and regions, as opposed to the metro fiber inside a single city. Nvidia’s involvement spans 24 Zayo routes in total: six new, ten upgraded and eight leased as they stand.

    The announcement matters because it arrived the same week Lumen Technologies’ CFO, Chris Stansbury, said the economics of building new long-haul routes “stink” and that Lumen had walked away from several such deals. Lumen’s one exception, a Seattle-to-Minneapolis route called NorthLine, works financially because a partner is sharing it. Zayo’s Nvidia deal follows the same pattern. New intercity fiber for AI is getting built where a large, committed customer underwrites the route before construction begins.

    Zayo also says the routes were chosen by mapping where electrical power is available for future data centers, and that the conduit is sized well beyond Nvidia’s own needs. That makes the build relevant to every data center developer and cloud operator siting capacity along these corridors.

    Why an Anchor Tenant Changes the Math

    Building a new long-haul fiber route is expensive long before it earns anything. Zayo’s executives described what the work involves: dealing with hundreds of municipalities, regulators and permitting bodies, and coordinating multiple construction contractors. The money goes out during construction, while lease revenue arrives gradually over years. If a route is built on speculation and demand arrives slowly, the builder pays to finance an asset that is sitting idle.

    Lumen’s comments show this risk plainly. Stansbury said Lumen has “stepped away from a number of deals” because shareholder returns on greenfield routes are “awful.” Yet Lumen is building NorthLine because a partner, which it has not named, makes “the economics for that actually very good.” Zayo’s position rests on the same mechanism. With Nvidia committed as anchor tenant across 16 new or upgraded routes plus eight existing ones, Zayo starts construction with a large share of demand already contracted. It describes the result as an attractive return. The two companies reach opposite verdicts on greenfield fiber in general, but they agree on the condition that makes it work: a committed buyer before the trench is dug.

    This is a pattern in two data points, not a law of the market. Zayo also argues that its experience building new routes over the past seven or eight years gives it a cost and execution edge competitors lack. Because neither the cost nor Nvidia’s commitment has been disclosed, outsiders cannot check the return claim. What they can see is that the one greenfield build each company is pursuing has a large partner behind it.

    Fiber Is Now Following the Power Map

    Long said Zayo began seeing sharply higher long-haul demand about two years ago, in the form of orders for far higher fiber counts than before. The company responded by analysing where electrical power was available across the U.S. and concluding that those corridors were where bulk fiber would be needed. Bulk fiber means large quantities of strands sold to a few big users rather than small circuits sold to many.

    This reverses the traditional logic of long-haul networks, which were drawn between population centers and internet exchange points. AI data centers need hundreds of megawatts of electricity, and that power is scarcer than land or fiber, so it now decides where these facilities go. Fiber is being planned to meet them there. For developers, a site with both secured power and a new high-count fiber route nearby becomes more valuable. For utilities and regional economic planners, a new route can signal that data center projects are expected to follow.

    Built Bigger Than One Customer

    Zayo CEO Steve Smith said popular routes will carry at least 1,728 fibers, and that the conduit going into the ground can support up to 20,000. Conduit is the protective pipe that cables are pulled through, and it can be filled with more fiber over time. Long said Nvidia wanted a build package that would serve “the rest of the ecosystem” as well as its own use.

    For Zayo, oversizing the conduit turns an anchor-backed build into a multi-tenant asset. Pulling additional cable through existing conduit is typically far cheaper than new trenching, so every later customer on the route improves returns on the original construction. Cloud providers, newer GPU-cloud operators and other carriers would all be potential buyers. The main risk is timing. With completion stretching to 2030, the build depends on AI data center demand in these corridors holding up for years.

    Mostly Upgrades, Not Blank-Slate Builds

    Of the 24 routes Nvidia is using, only six are true greenfield builds. Ten are overbuilds on existing Zayo fiber, and eight needed no new construction at all. Overbuilds can typically draw on existing paths and permissions, which reduces permitting risk and cost compared with new routes. That mix suggests the deal leans heavily on Zayo’s existing footprint. It also suggests the greenfield portion, which Lumen called hardest to finance, is the smaller share of the program.

    Background

    Zayo Group is a U.S. fiber infrastructure provider that sells bandwidth and bulk fiber to carriers, cloud providers and large enterprises. Its executives say it has been the only company building new long-haul routes at scale in the U.S. over the past seven or eight years. It has also grown through acquisitions, including a $4.25 billion fiber deal with Crown Castle. Zayo reports that AI has driven a sharp rise in long-haul demand, with customers ordering much higher fiber counts than in the past.

    Nvidia is the leading supplier of the processors used to train and run AI models, and its customers are building large data centers wherever power is available. Lumen Technologies, formerly CenturyLink, operates one of the largest U.S. fiber networks and competes with Zayo for AI-driven long-haul demand.

    Sources

    Source: Zayo confirms Nvidia as its new ‘monster’ customer (Fierce Network): Zayo executives on the 16-route long-haul fiber build anchored by Nvidia, and Lumen’s view of greenfield route economics.

  • NVIDIA’s Reported $50B Lease Is Really a $19.6B Bet on Powered AI Capacity

    NVIDIA’s Reported $50B Lease Is Really a $19.6B Bet on Powered AI Capacity

    TL;DR · 30-second read

    The Short Version

    NVIDIA, the company whose chips run most artificial intelligence, has reportedly agreed to rent a huge computer campus in Texas from a firm called Hut 8. NVIDIA has not confirmed it.

    The deal is described as worth $50 billion, but only $19.6 billion is locked in for the first 15 years. The rest depends on renewals after that.

    What is really being rented is electricity: a guaranteed grid connection roughly equal to what one large nuclear reactor produces. For artificial intelligence, that has become as important to secure as the chips themselves.

    Data Center Frontier reported on August 3, 2026 that NVIDIA has emerged as the tenant behind Hut 8’s lease commitments at its one-gigawatt Beacon Point AI data center campus in Nueces County, Texas. Hut 8 has disclosed two 15-year leases covering 704 megawatts of IT capacity with a combined base-term value of $19.6 billion. That figure rises to as much as $50.2 billion if all renewal options are exercised. The Financial Times identified NVIDIA as the tenant. Reuters said it could not independently verify that, and NVIDIA neither confirmed nor denied its role, saying only that it works with ecosystem partners to deploy AI infrastructure through its DSX AI factory architecture.

    In a separate announcement, Aalo Atomics and Crusoe said they will partner on a nuclear-powered AI data center. They plan to demonstrate it at Idaho National Laboratory in 2027 and then deploy 50-megawatt Aalo Pod reactors at Crusoe data centers by the end of 2029.

    Executive Summary

    Hut 8 has fully commercialized Beacon Point with two 352-megawatt phases, each carrying a base-term contract value of about $9.8 billion. The campus runs on a one-gigawatt interconnection agreement with AEP Texas and is being designed around NVIDIA’s DSX reference architecture. If the reported tenant is confirmed, NVIDIA would move beyond supplying GPUs and become an anchor tenant whose credit helps make a multibillion-dollar campus financeable.

    The headline number overstates what is firm. Only $19.6 billion of the $50.2 billion is committed. The remaining $30.6 billion depends on renewals beyond the initial 15 years. What the firm commitment secures is 704 megawatts of capacity that is already power-connected, and in the current AI buildout that is the scarcest input.

    The Aalo–Crusoe partnership takes the opposite approach to the same constraint. Instead of locking up a gigawatt of grid power, it aims to manufacture both a small reactor and a modular data center and to add them 50 megawatts at a time. Beacon Point itself is grid-connected, not nuclear-powered.

    Most of the $50 Billion Sits in Renewal Options

    The $50.2 billion figure is a ceiling, not a check. Hut 8 has disclosed two 15-year leases with a combined base-term contract value of $19.6 billion, about $9.8 billion for each 352-megawatt phase. The remaining $30.6 billion, roughly 61 percent of the headline, materializes only if every renewal option beyond the initial terms is exercised.

    The firm commitment is still very large. Spread evenly, $19.6 billion over 15 years comes to about $1.3 billion a year, or roughly $1.9 million per megawatt of IT capacity annually. That is a simple average and not a rent schedule, because payment start dates and escalators have not been published.

    Whether the renewals happen depends on something no one can promise today: that a campus designed in 2026 will still be competitive decades from now. Computing systems turn over rapidly, while the land, the shell and the grid connection last far longer. Power density and cooling will need repeated upgrades to keep the site useful. The base term prices the next 15 years of AI demand, and the option value prices how long the power connection itself stays valuable.

    What the Lease Actually Buys Is Powered Capacity

    Set aside the headline value, and the scarce asset at Beacon Point is electricity that has already been promised. The campus holds one gigawatt of utility capacity under an interconnection agreement with AEP Texas. An interconnection agreement is the utility contract that lets a large load plug into the grid, and obtaining one has become as important as obtaining GPUs. The two leases commit 704 megawatts of IT capacity, meaning the power delivered to servers, networking and storage. The gap between that figure and the 1 GW connection typically covers cooling and electrical overhead.

    Hut 8’s own disclosure shows which variable is fixed. The company said it redesigned the first Beacon Point data hall around NVIDIA’s architecture and increased capacity by 57 percent within the same land and utility footprint. The grid connection did not grow. What changed was how much compute the design could extract from it. NVIDIA pitches DSX on the same logic, emphasizing token performance per megawatt (AI output per unit of power) rather than floor space.

    The parties most affected are those who cannot secure this kind of power allocation on their own. Some reporting has speculated that NVIDIA could sublease capacity. If it does, smaller GPU cloud providers, often called neoclouds, would gain access to power and space they could not finance independently. For Hut 8, the leases bring its contracted AI portfolio to 949 megawatts on 1,330 megawatts of utility capacity, with $26.6 billion of aggregate base-term value. Beacon Point accounts for roughly three-quarters of that value.

    A Chip Supplier’s Credit Becomes Construction Finance

    If the Financial Times identification holds, the bigger shift is NVIDIA’s role in the market. Campuses at this scale are typically financed against the creditworthiness of the tenant, and Hut 8 has described its counterparty as investment-grade. An NVIDIA lease would make the company an anchor tenant whose balance sheet helps get a multibillion-dollar campus built. It would no longer be only the supplier of the GPUs, networking, systems and software inside.

    The strategic logic is clear. A supplier that underwrites capacity can pull forward projects that might otherwise stall on financing. It can also standardize those projects around its own reference design and widen the pool of customers able to rent its hardware. The open question, which only NVIDIA can answer, is how much of the demand behind such a campus comes from end customers and how much is the supplier’s own commitment. A lease anchored by the chip vendor and sublet to the vendor’s customers concentrates risk around a single forecast of compute demand. That is the tradeoff the model accepts in exchange for speed.

    The Other Model: Putting the Reactor Next to the Racks

    The nuclear model in this news cycle comes from Aalo Atomics and Crusoe, not from Beacon Point. The two companies plan to run a Crusoe Spark modular data center on power from an Aalo reactor at Idaho National Laboratory in 2027. They then intend to deploy Aalo Pods, which are 50-megawatt-electric nuclear plants, at Crusoe data centers by the end of 2029. Crusoe puts its Spark factory lease, buildout and initial module fleet at more than $200 million. Aalo has secured more than $136 million in funding.

    The milestone so far should be read precisely. On July 4, 2026, Aalo’s Critical Test Reactor reached criticality, meaning it sustained a nuclear chain reaction. It did so at zero power, before sodium coolant or any electricity-generating systems were added. The 10-megawatt-electric Aalo-X power reactor is being built next door. Commercial sites will still require regulatory approval, fuel, waste-management plans, insurance, financing and community support. For that reason, 2027 and 2029 are better read as ambitious targets than as delivery dates.

    Side by side, the two projects answer the same constraint in opposite ways. Beacon Point locks up a gigawatt of grid power and fills it in 352-megawatt blocks. Aalo and Crusoe aim to manufacture both the power plant and the data center and to add capacity 50 megawatts at a time. In both cases, securing power comes first in the contract.

    Background

    Hut 8 built its business in bitcoin mining and has been repositioning toward AI data center development, leasing power-connected capacity to large tenants under long-term contracts. Its Beacon Point campus in Nueces County, Texas, is now fully contracted and sits within a 949-megawatt contracted AI portfolio supported by 1,330 megawatts of utility capacity. NVIDIA supplies the GPUs, networking, systems and software that many AI data centers run on. It promotes DSX as a reference architecture for designing entire facilities around its platforms.

    Crusoe is an AI cloud and data center company that sells computing through Crusoe Cloud and builds prefabricated Spark data center modules. Aalo Atomics is developing small advanced reactors, including the 10-megawatt-electric Aalo-X, a sodium-cooled design, and the 50-megawatt-electric Aalo Pod. The company is expanding toward a one-million-square-foot manufacturing operation, and its planned demonstration with Crusoe is sited at Idaho National Laboratory.

    Sources

    Source: NVIDIA’s Reported $50B Lease and the Nuclear-Powered AI Factory: Data Center Frontier’s August 3, 2026 analysis of Hut 8’s Beacon Point leases and the Aalo–Crusoe nuclear data center partnership.

  • LG’s 600kW Nvidia-Certified CDU Shows Why AI Cooling Is a Vetted-Vendor Market

    LG’s 600kW Nvidia-Certified CDU Shows Why AI Cooling Is a Vetted-Vendor Market

    TL;DR · 30-second read

    The Short Version

    LG, best known for televisions and fridges, just won approval from Nvidia, the company whose chips run most artificial intelligence systems, to supply equipment that keeps those chips cool.

    One cabinet of Nvidia’s newest chips uses roughly as much electricity as a hundred homes. That makes it far too hot for fans, so liquid has to be pumped right up against the chips.

    LG is the first Korean company approved for this. The approval matters because buyers rarely risk a leak next to millions of dollars of hardware.

    Startup Fortune reported that LG Electronics’ 600kW-class coolant distribution unit (CDU) has passed Nvidia’s final quality certification. That makes LG the first Korean company approved in this category of liquid cooling equipment for Nvidia-based AI server racks. LG announced the approval in late July 2026.

    LG says it will next seek certification for 1MW, 2.5MW and 4MW CDU models. It sells the unit as part of a ‘Chip to Chiller’ package spanning server-level cold plates through building-scale chillers.

    Executive Summary

    A coolant distribution unit is the pump-and-heat-exchanger that moves liquid between a data center’s cooling plant and the cold plates mounted on GPUs and CPUs. LG’s 600kW unit is now on Nvidia’s approved list. LG says it holds coolant temperature within ±0.25°C, runs redundant pumps and power supplies, and includes virtual sensors and leak detection.

    The approval matters because of rack density. Nvidia’s DGX GB200 NVL72 system draws about 120kW per rack, roughly four times the level at which air cooling alone stops working. At that density, liquid cooling is mandatory. It also sits millimetres from very expensive silicon, so buyers screen suppliers before any price is discussed.

    For LG, the certification also serves its shift toward business customers. It gets LG into the running for Nvidia-based deployments. It does not yet deliver orders: LG has named affiliate projects and earlier regional references, but no large external cloud contract.

    At 120kW a Rack, Air Is Out of the Picture

    Data Center Engineering puts the practical limit for air handling at roughly 30kW per rack. Above that, heat removal has to move closer to the chip. Nvidia’s own DGX GB200 system guide describes the NVL72 rack at approximately 120kW, with 72 Blackwell GPUs and 36 Grace CPUs in a single rack-scale design. For operators, liquid cooling at that density is a requirement, not a design choice.

    The CDU is the working heart of that liquid loop. It takes heat from coolant returning off the chips’ cold plates and passes it to the facility’s water system, while holding flow and temperature steady. On paper, a 600kW unit covers the heat of about five 120kW racks. That is a row-level building block, not a whole-hall system.

    Why Approval Comes Before Price

    This is the mechanism behind the headline. When the equipment being cooled costs millions of dollars per rack, a coolant leak or pump failure is a business interruption plus a large hardware bill, not a routine repair. Buyers therefore lean on Nvidia’s qualification as a first screen. The approved list narrows the field before bidding starts, so a cooling vendor’s first hurdle is getting qualified, and competing on price comes after.

    LG’s stated specifications match what that screen looks for: tight temperature control (±0.25°C), redundant pumps and power so cooling continues through a component failure, and leak detection. None of this is novel on its own. Together, with Nvidia’s approval attached, it lets LG compete for business it previously could not bid on.

    The effects run in several directions. Established cooling vendors gain a large, well-capitalised new competitor. Hyperscalers, the largest cloud operators, and other data center builders gain another qualified source at a time when liquid cooling demand is rising fast. Uncertified suppliers face a clearer barrier. One certification does not prove the whole market works this way, but it shows plainly where the entry gate for Nvidia-based cooling sits.

    From 600kW to 4MW: The Roadmap Is the Real Test

    LG plans to seek certification for 1MW, 2.5MW and 4MW CDUs next. The scale jump is significant: a 4MW unit could, in principle, serve the heat of more than thirty NVL72-class racks. That is the size at which large AI campuses buy cooling. The 600kW approval gets LG into the running. The megawatt-class approvals will decide whether LG can supply those campuses.

    LG frames the opportunity with a market forecast of about 26 percent annual growth, reaching roughly $133.5 billion (about 185 trillion won) by 2034. This is LG’s own projection, and it has not published the underlying methodology. Readers should treat it as the company’s view of market size, not an independent estimate. The direction is less in doubt than the number: Nvidia’s published rack power figures already require liquid cooling at the top end.

    An Appliance Maker’s Business-Customer Arithmetic

    The certification also fits LG’s broader repositioning. At a January 7 briefing, LG said business lines including B2B, webOS and direct-to-consumer grew from about 29 percent of revenue in 2021 to 45 percent in the second half of 2025. Their share of operating profit rose from 21 percent to about 90 percent. Data center cooling draws on work LG already does: HVAC, heat transfer and precision temperature control.

    The execution risk is real. LG’s stated pipeline includes AI data centers for group affiliate LG CNS, along with earlier cooling references in North America and Indonesia. These are credible starting points. They do not show that global cloud buyers, who usually demand worldwide service coverage and proven delivery at volume, will commit to LG for large deployments.

    Background

    LG Electronics built its brand on televisions, refrigerators and washing machines. For several years it has been shifting its growth strategy toward business customers, including commercial HVAC, and this year it said those business lines generate most of its operating profit. Precision cooling for data centers is a natural extension of that heat-transfer engineering.

    Nvidia’s Blackwell-generation rack-scale systems have made liquid cooling routine at the high end of AI infrastructure. The GB200 NVL72 combines 72 GPUs and 36 CPUs in one rack drawing about 120kW, far above the roughly 30kW at which air cooling reaches its limit. That has created fast-growing demand for direct-to-chip cooling components such as cold plates and coolant distribution units. Nvidia’s supplier qualification acts as the main gate into that supply chain.

    Sources

    Source: LG Electronics wins Korea’s first Nvidia liquid cooling certification as AI data centers hit a heat wall (Startup Fortune): LG’s 600kW coolant distribution unit clears Nvidia’s quality certification.

  • Japan’s 27,500-GPU National AI Factory Is Really a 140-Megawatt Power Project

    Japan’s 27,500-GPU National AI Factory Is Really a 140-Megawatt Power Project

    TL;DR · 30-second read

    The Short Version

    Japan’s government is backing a giant computer center, built with chipmaker NVIDIA and a Japanese company called Noetra, to train artificial intelligence for robots, factories and hospitals.

    It will hold 27,500 of NVIDIA’s newest high-end chips and be able to draw 140 megawatts of electricity, roughly what a small city uses.

    The chips get the headlines, but lining up that much power and cooling is the harder job. The announcement does not yet say where it will be built, when, or who is paying.

    NVIDIA announced on July 16, 2026 that it is working with Noetra Corp. to build an NVIDIA Vera Rubin “AI factory” in Japan with 13,750 Vera CPUs and 27,500 Rubin GPUs, delivering 140 megawatts of data center capacity on NVIDIA’s DSX reference platform and Spectrum-X Ethernet networking. The facility will provide the computing foundation for the FRONTia Project, a program launched by Japan’s Ministry of Economy, Trade and Industry (METI) to develop multimodal foundation models for robotics and “physical AI.”

    NVIDIA calls it the world’s first national AI infrastructure for physical AI. Noetra will share the pretrained weights of its models with Japanese developers and enterprises, alongside NVIDIA’s open model families and software libraries.

    Executive Summary

    The announcement pairs a state industrial program with a single, very large training cluster. METI’s FRONTia Project sets the mission: build reliable multimodal foundation models (large AI models that learn from text, images, video and sensor data together) that can underpin robots, digital twins and industrial automation. Noetra will build and run the hardware, and NVIDIA supplies the full stack from chips and racks to networking and software.

    It matters for two reasons. First, it is an explicit bet that Japan’s manufacturing base, and the real-world industrial data it generates, can be turned into a competitive advantage in AI for machines that act in the physical world. Japan’s AI Robotics Strategy targets more than 30% of a global AI robotics market it estimates at $133 billion by 2040. Second, the size of the build means this is a utility-scale infrastructure project as much as a computing one: 140 megawatts is the scale of a large hyperscale campus, and the announcement leaves open where that power and the cooling to go with it will come from.

    The Unit That Matters Is the Megawatt

    The headline numbers are chips, but the figure that governs how this project gets built is 140 megawatts. Divide it by 27,500 GPUs and the facility is provisioned at roughly 5 kilowatts per GPU once cooling, networking, storage and power-conversion losses are counted. The release describes the build as Vera Rubin NVL72 racks, which pair 72 GPUs with 36 CPUs, the same 2-to-1 ratio as the 27,500 and 13,750 figures announced. That works out to roughly 380 racks sharing 140 megawatts, or about 365 kilowatts of facility capacity per rack. The release does not separate IT load from cooling and overhead, so the per-rack draw at the chip level will be lower. Even so, the order of magnitude is far beyond the 10 to 20 kilowatts a conventional enterprise rack uses.

    Density at that level has direct operational consequences. Air cannot remove that much heat, so racks of this class are liquid-cooled, which means pumps, heat exchangers and a heat-rejection plant sized for a small power station’s output. A 140-megawatt load also needs a high-voltage grid connection and on-site substation capacity, which in most markets are the longest-lead items in any data center schedule. NVIDIA itself frames the platform this way: its DSX reference design is pitched on raising “token throughput per megawatt,” a metric that treats power, not silicon, as the scarce input.

    For NVIDIA as a named partner, securing the GPUs is the part most within the project’s control. Site selection, grid interconnection, permitting and cooling construction are the parts that determine when the factory actually trains its first model. The parties most affected are the utility that serves the site, the electrical and mechanical contractors, and cooling-equipment suppliers. The release names none of them.

    What “National” Means in This Model

    The novel element is less the hardware than the arrangement around it. METI supplies the policy mandate through FRONTia, Noetra builds and operates the cluster, and the output is intended as a shared national asset. Pretrained model weights will be “broadly available” to domestic developers and enterprises. That is closer to how governments treat roads or research reactors than to a commercial GPU cloud, where capacity is rented by the hour to whoever pays.

    The “world’s first national AI infrastructure” claim deserves a careful reading. Japan already operates government-backed AI computing, such as the ABCI supercomputer run by the national research institute AIST, and METI has previously subsidized domestic cloud providers to expand GPU capacity. The release qualifies its claim in the body text as the first national AI infrastructure “for physical AI.” That narrower framing, a national-scale system dedicated to robotics and industrial foundation models, is more defensible than the headline version.

    Physical AI and the 2040 Target

    “Physical AI” refers to models that perceive and act in the real world: robots on a factory line, autonomous logistics systems, digital twins that simulate a plant before it is changed. Training them is data-hungry in a different way from chatbots, because they need video, sensor and motion data from real industrial settings. METI’s pitch is that Japan’s manufacturers hold exactly that data and that pooling it around a common model gives domestic firms a head start they could not buy individually. Noetra’s CEO, Hironobu Tamba, makes the same argument: these are “challenges no single company can solve alone.”

    The link from one cluster to a 30% global market share by 2040 is a long one, and the release does not attempt to draw it. The $133 billion figure is an estimate attached to a policy goal, not a forecast tied to this facility. The release says the factory will support trillion-parameter-scale models “as the AI factory expands,” which suggests a phased build. The claims of “lower token costs” and “breakthrough AI performance” are likewise unquantified.

    Who Gains From the Full-Stack Design

    The project is NVIDIA end to end: Vera CPUs, Rubin GPUs, BlueField data processing units, Spectrum-X Ethernet, and the Nemotron, Cosmos, Isaac GR00T and NeMo software that will ship alongside Noetra’s models. For Japanese developers, that means a ready toolchain and a common starting point, which lowers the cost of building robotics applications. It also means the national physical-AI ecosystem will be built on one vendor’s architecture from the silicon upward, a trade-off between speed and supplier diversity that sovereign AI programs worldwide are weighing. Choosing Ethernet rather than InfiniBand for the cluster fabric fits a wider industry move toward standards-based networking at scale, which may make it easier to find operators and engineers who can run it.

    Background

    Japan has pursued AI computing as industrial policy for several years, building government research systems and subsidizing domestic cloud providers to expand GPU capacity. METI’s AI Robotics Strategy, released in March 2026, sharpened that focus on robotics and physical AI, where Japan’s manufacturing base gives it a large pool of real-world industrial data. FRONTia is METI’s vehicle for turning that strategy into shared foundation models.

    NVIDIA has been promoting “sovereign AI,” national AI capacity built within a country’s borders, as a major market alongside its hyperscale cloud customers. Vera Rubin is its next-generation platform after Blackwell, pairing Vera CPUs with Rubin GPUs in dense, liquid-cooled NVL72 racks, and DSX is its reference design for building complete AI data centers around that hardware.

    Sources

    Source: Japan Government, Industrial Leaders and NVIDIA Launch the World’s First National AI Infrastructure (NVIDIA Newsroom, July 16, 2026). NVIDIA and Noetra announce a 140-megawatt Vera Rubin AI factory for METI’s FRONTia physical AI project.