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

Tag: data center power

  • Vectris Claims Up to 73% More AI Throughput From GPUs Already Deployed

    Vectris Claims Up to 73% More AI Throughput From GPUs Already Deployed

    Vectris Labs, a Birmingham, Alabama startup incubated by Thumos Capital, announced on August 20, 2026 that its Waveform software — a “control plane” that sits between AI serving infrastructure and the GPU — recovered substantial unused capacity from GPUs already in production racks. In company-run tests of Mistral inference workloads on RunPod-hosted NVIDIA hardware, Vectris measured 30–73% higher throughput, 51–56% lower energy consumption, and 22–42% faster job completion, with no model retraining, weight changes, or GPU-kernel modifications.

    Waveform launches October 1, 2026 to a limited set of design partners. The results are Vectris-measured and, by the company’s own disclosure, have not yet been independently reproduced in customer production.

    Executive Summary

    The announcement reframes the AI capacity crunch — the industry-wide shortage of GPUs, data-center space, and grid power — as partly a software-efficiency problem. Vectris claims to have found “deterministic structural patterns” in AI inference (the process of running a trained model to answer queries) that reveal where deployed GPUs are wasting cycles, and to have built software that captures that waste as productive output. The company brands the resulting metric Compute Yield™: how much quality-equivalent, accepted AI output an operator gets from infrastructure already in place.

    If the numbers hold up outside Vectris’ own testing, the implications are significant. At even the conservative +30% end of its measured range, the company illustrates that a 10,000-GPU fleet would produce output comparable to 13,000 GPUs — capacity gained without new hardware, new power contracts, or new construction. Vectris is explicit that this is an extrapolation, not a measured deployment.

    The caveats matter as much as the headline. The figures come from one model family (Mistral), one hosting environment (RunPod), and one measuring party (Vectris itself). The release is unusually candid about those limits, which is to its credit — but it also means the claim currently rests entirely on vendor-run benchmarks awaiting independent reproduction.

    Efficiency Is the New Front in the AI Capacity War

    For three years, the dominant response to surging AI demand has been construction: more GPUs, more data centers, more megawatts. But power availability, capital intensity, and build timelines have become structural constraints — a data center can take years to energize, while inference demand compounds monthly. That makes software that extracts more work from installed hardware strategically interesting regardless of which vendor ultimately delivers it. Vectris’ framing — that the binding economic question is shifting from “how many GPUs can you deploy?” to “how much useful output can deployed GPUs produce?” — is a fair description of where operator economics are heading, and it explains why the company says it has engaged a data-center advisory network representing roughly 300 MW of capacity.

    The energy numbers may be the most consequential part of the claim for infrastructure operators. A 51–56% reduction in energy per unit of inference work, if reproducible, would ease the single tightest constraint in the industry — grid power — and change the calculus on every pending interconnection queue. That is precisely why the figure deserves the most scrutiny before anyone builds plans around it.

    What’s Substantiated — and What Isn’t

    The release is more disciplined than most in this category. It names the hardware (H100, H200, B200 on third-party RunPod infrastructure), the workload (Mistral inference), publishes per-GPU figures rather than a single cherry-picked number, labels the 10,000-GPU example as illustrative, and states plainly that results “have not yet been independently reproduced in customer production.” On Intel silicon, Vectris cites 67% energy savings and 32% faster time-to-result using MLPerf LoadGen, a recognized benchmark harness. AMD hardware has been “tested,” but no numbers are given.

    What remains unsubstantiated is the core of the claim. The release does not describe the baseline configuration Waveform was compared against — a critical omission, because inference throughput varies enormously with batching strategy, serving stack, and tuning. A 73% gain over a poorly tuned baseline is a very different achievement than 73% over a well-optimized production stack. Vectris says Waveform targets waste “that remains after conventional optimization,” but offers no detail on what conventional optimization was applied. Nor does it explain the mechanism: “deterministic structural patterns” is evocative but not technical, and “quality-equivalent accepted output” — the foundation of the Compute Yield metric — is not defined in measurable terms. None of this means the claims are wrong; it means they are, for now, claims.

    Winners, Losers, and the Demand Question

    If Waveform performs as described, the clearest winners are inference-heavy operators who are power- or capital-constrained: neoclouds, enterprise AI platforms, and colocation tenants who could defer hardware purchases while serving more demand. Data-center operators face a more nuanced picture — efficiency software could modestly slow demand for new capacity, but historically, cheaper compute has expanded consumption rather than shrinking footprints, a dynamic economists call the Jevons effect. GPU vendors face the same ambiguity: software that makes an H100 do 30–73% more work makes existing fleets more valuable even as it potentially trims marginal unit demand.

    Vectris also enters a genuinely crowded field. Inference optimization is one of the most active areas in AI infrastructure — serving frameworks, compilers, schedulers, and quantization techniques all chase the same waste. Vectris positions Waveform as complementary, a layer above the optimized stack rather than a replacement for it. Whether meaningful recoverable capacity really persists after state-of-the-art serving optimizations is exactly the question independent testing needs to answer.

    From Benchmark to Business

    The commercial plan is early-stage: an October 1, 2026 launch limited to design partners, technical demonstrations with unnamed “AI-infrastructure and channel leaders,” and no disclosed pricing, customers, or funding. The team’s stated pedigree — backgrounds spanning AMD, Graphcore, Oracle Cloud Infrastructure, ByteDance, the U.S. Department of Energy, and Oak Ridge National Laboratory — is relevant to credibility on low-level GPU behavior, but pedigree is not production validation. The supporting quote from Innovate Alabama Chairman Bill Poole speaks to regional economic-development enthusiasm rather than technical endorsement, and the release’s own disclosure notes that third-party names do not imply endorsement. The sensible read: a credible team making a large, testable claim that the market should now test.

    Background

    Vectris Labs is a newly announced entrant in AI infrastructure software, based in Birmingham, Alabama and incubated by venture firm Thumos Capital — a notable geography in an industry concentrated in traditional tech hubs, and one the release leans into with a supporting quote from Innovate Alabama Chairman Bill Poole. The company says it has completed technical demonstrations with AI-infrastructure and channel leaders and engaged a data-center advisory network representing roughly 300 MW of capacity.

    The market context is the defining tension of the current AI buildout: inference — serving trained models to end users — is becoming the dominant AI workload, while power availability and capital costs constrain how fast new GPU capacity can come online. That squeeze has pushed the industry’s attention toward yield: getting more accepted output per deployed GPU, per megawatt, and per dollar, which is precisely the territory Vectris is staking out.

    Source: Vectris Discovers Recoverable AI Compute Capacity Inside Deployed GPUs, Demonstrating Up to 73% More Productive Capacity — Vectris Labs press release via PR Newswire, August 20, 2026, announcing the Waveform control plane and company-measured GPU efficiency results.

  • Nvidia and Mitsubishi Heavy Reportedly Weigh AI Data Center Cooling and Power Tie-Up

    Nvidia and Mitsubishi Heavy Reportedly Weigh AI Data Center Cooling and Power Tie-Up

    Nvidia and Japan’s Mitsubishi Heavy Industries are reportedly in early discussions about collaborating on cooling and power infrastructure for AI data centers, according to a July 13, 2026 Seeking Alpha report surfaced via Google News.

    The item is a brief report of external reporting; no formal announcement, deal value, timeline, or product scope has been confirmed by either company.

    Executive Summary

    The reported talks would pair the dominant supplier of AI accelerators with one of the world’s largest heavy-engineering conglomerates, whose portfolio spans gas turbines, HVAC systems, and industrial cooling. On paper, the fit is obvious: AI clusters built around Nvidia’s highest-end GPUs are pushing rack densities and heat loads well past what conventional air-cooled data centers were designed to handle, and grid interconnection queues in key markets are measured in years rather than months.

    What matters for readers is less the headline than the pattern. Chipmakers are increasingly reaching upstream into the physical plant — power generation, thermal management, on-site energy — because compute deployment is now gated by megawatts and cooling capacity, not silicon supply. Whether this specific pairing produces a concrete product, a joint venture, or nothing at all remains unclear from the available reporting.

    Why a Chip Company Cares About Chillers

    Modern AI training racks can dissipate 100 kilowatts or more — an order of magnitude above traditional enterprise servers — and next-generation GPU platforms are pushing hotter still. At those densities, air cooling stops being economical and liquid cooling, whether direct-to-chip cold plates or full immersion, becomes mandatory. Mitsubishi Heavy’s industrial thermal and HVAC businesses are the kind of scaled manufacturing base that a chip vendor would want aligned with its reference designs, so that when a customer buys a rack, the cooling loop is engineered, warrantied, and shippable at the same cadence as the servers.

    The power side of the reported discussion is equally telling. Mitsubishi Heavy builds gas turbines and is active in nuclear and hydrogen-adjacent equipment. Data center developers in the United States, Japan, and Europe are increasingly signing behind-the-meter or on-site generation deals because utility interconnection timelines cannot keep pace with hyperscaler expansion plans. A relationship with a turbine manufacturer is one way to shorten that critical path.

    Strategic Logic, With Caveats

    For Nvidia, the strategic prize is deployment velocity: every month a customer waits for power or cooling is a month of deferred GPU revenue and a window for a rival platform. For Mitsubishi Heavy, aligning with the dominant AI compute vendor could pull its industrial equipment into a growth market with unusually inelastic demand. Both narratives are plausible, and both have been used to explain similar chatter around other equipment makers over the past 18 months.

    The measured read, however, is that a report of exploratory talks is not a partnership. Neither company has published terms, and Seeking Alpha itself is aggregating reporting rather than breaking primary news. Readers should treat the item as a signal of direction — chip vendors seeking closer ties to power and thermal OEMs — rather than as a confirmed commercial arrangement.

    Winners, Losers, and the Middle of the Stack

    If a formal collaboration materializes and yields co-engineered reference designs, the pressure would land squarely on independent liquid-cooling specialists and on power-equipment competitors that lack a chip-vendor relationship. Colocation operators would likely welcome a validated, warrantied stack because it reduces integration risk on the largest deals. Hyperscalers, who tend to prefer multi-sourcing and their own custom designs, may care less at the design level but still benefit from a deeper supplier bench.

    The counter-scenario is that talks fizzle, or produce only a narrow marketing arrangement. That outcome would be consistent with how many announced infrastructure partnerships have played out — press coverage first, meaningful shipments much later, if at all. Either way, the underlying constraint is real: AI infrastructure is now a power-and-cooling problem as much as a semiconductor one.

    Background

    Nvidia is the dominant supplier of graphics processing units used for AI training and inference, and its data center segment has become the fastest-growing business in enterprise compute. Its accelerators are the reference platform for most large-model training clusters, which has made the physical constraints of deploying them — power, cooling, real estate — the industry’s binding bottleneck.

    Mitsubishi Heavy Industries is a diversified Japanese engineering conglomerate with more than a century of history in power generation, thermal equipment, aerospace, and industrial machinery. Its portfolio includes gas turbines, HVAC systems, and nuclear-related equipment, giving it multiple potential entry points into the data center power-and-cooling stack.

    Source: Nvidia, Mitsubishi Heavy mull team up for AI data center cooling, power: report – Seeking Alpha — brief report of exploratory discussions between the two companies on AI data center infrastructure, aggregated via Google News.

  • SLB and Liberty Energy Ally to Power Data Center Buildout

    SLB and Liberty Energy Ally to Power Data Center Buildout

    SLB, the global oilfield services company, and Liberty Energy, a North American oilfield services and power provider, announced on July 13, 2026 that they are forming a strategic alliance focused on data center infrastructure and power. The two firms plan to combine capabilities to serve the fast-growing compute build-out with integrated energy and site solutions.

    Executive Summary

    The alliance pairs SLB, one of the largest energy technology companies in the world, with Liberty Energy, a Denver-based firm best known for hydraulic fracturing services and, more recently, distributed power generation. Together they intend to address data center customers who need both physical infrastructure and reliable electricity at sites where grid capacity is constrained.

    The announcement matters because it is another concrete signal that the oil and gas services industry sees data center power — particularly behind-the-meter and gas-fired generation — as a durable adjacent market. For hyperscalers and colocation operators facing multi-year interconnection queues, packaged offerings from experienced heavy-industrial contractors could shorten the path from land to live megawatts.

    Oilfield Services Pivots Toward the Compute Grid

    Both SLB and Liberty Energy come from the upstream oil and gas world, where they routinely mobilize large mechanical, electrical and civil crews to remote sites on tight schedules. That skill set — moving turbines, engines, fuel systems and instrumentation to greenfield locations quickly — maps unusually well to the current data center bottleneck, which is less about chips and more about getting power to the meter. Framing the alliance as “infrastructure and power” (rather than a single-product play) suggests the partners want to sell a bundle: site engineering, generation equipment, fuel logistics and operations.

    The commercial logic is straightforward. Utility interconnection timelines in many U.S. markets now stretch beyond the useful life of a GPU generation, pushing operators to consider on-site or “behind-the-meter” power. Companies that already own the supply chain for gas turbines, reciprocating engines and fuel handling can, in principle, stand up hundreds of megawatts faster than a regulated utility can expand a substation. The release does not, however, quantify what capacity SLB and Liberty intend to deliver, or on what timeline.

    Winners, Losers and the Questions That Follow

    If the alliance executes, the most obvious beneficiaries are AI-focused developers who value speed-to-power over the lowest possible energy cost, and hyperscalers seeking a single accountable counterparty for hybrid on-site generation. Traditional EPC (engineering, procurement and construction) firms and independent power producers should read this as competitive pressure at the top of the market, particularly for gas-fired projects co-located with compute campuses.

    The harder questions concern durability and emissions. Behind-the-meter gas generation is faster to build than grid transmission, but it locks customers into fossil fuel exposure at a time when several hyperscale buyers have publicly committed to carbon reduction targets. The release itself makes no environmental claims, which is worth noting in both directions: the partners are not overselling a green story, but they are also not addressing how the offering would fit customers’ existing sustainability commitments.

    What the Announcement Substantiates — and What It Doesn’t

    Read narrowly, the July 13 release confirms a strategic alliance and a stated market focus. It does not, based on the material available, disclose a joint venture structure, capital commitments, named anchor customers, target geographies, project pipeline or specific technology partners for turbines, fuel cells or grid interconnection. Announcements of this form frequently precede more detailed deal structures; they can equally remain framework agreements that generate limited near-term revenue. Buyers evaluating the alliance should treat the current disclosure as an intent signal rather than a contracted capability.

    Background

    Data center power has become the binding constraint on AI infrastructure growth. Utility interconnection queues in major U.S. markets now routinely stretch several years, and hyperscalers have publicly explored gas turbines, small modular reactors and on-site renewables to get megawatts online sooner. This backdrop has drawn industrial and energy firms — including OEMs, EPC contractors and, increasingly, oilfield services companies — into the data center supply chain.

    SLB (formerly Schlumberger) is a global energy technology company with a long history in drilling, reservoir and production services. Liberty Energy, founded in 2011 and headquartered in Denver, built its business in North American hydraulic fracturing and has expanded into distributed power generation. Both companies bring project execution capabilities honed in remote, capital-intensive oilfield environments to a data center market that increasingly values speed of deployment.

    Source: SLB, Liberty Energy to Form Strategic Alliance for Data Center Infrastructure and Power — joint announcement from SLB describing a strategic alliance to supply integrated infrastructure and power to data center customers.

  • Utilities Scramble for Transformers as Data Center Demand Strains the Grid Supply Chain

    Utilities Scramble for Transformers as Data Center Demand Strains the Grid Supply Chain

    Reuters reported on July 8, 2026 that US power companies are scrambling to secure electrical equipment — the transformers, switchgear, and related grid hardware that move electricity from generators to customers — as surging demand from data centers strains available supplies. The report frames a nationwide procurement crunch: utilities that once ordered this equipment on routine replacement cycles are now competing for constrained manufacturing capacity against a wave of new large-load projects.

    Executive Summary

    The headline is not about a single deal or data center campus; it is about the industrial base underneath all of them. Transformers step electrical voltage up for long-distance transmission and back down for delivery, and switchgear is the apparatus that switches, protects, and isolates circuits. Neither is optional: every new data center interconnection, substation upgrade, and grid expansion needs both. Reuters’ reporting indicates that US utilities can no longer take timely delivery of this equipment for granted.

    Why it matters: for the first time in decades, US electricity demand is growing meaningfully, and data centers — particularly AI-driven facilities — are a leading cause. When the equipment supply chain becomes the pacing item, it stops being a utility procurement problem and becomes a constraint on data center delivery schedules, grid reliability investment, and ultimately on how fast the AI buildout can proceed. Power availability has already emerged as the industry’s defining bottleneck; this report locates part of that bottleneck one layer deeper, in the factories that make grid components.

    Why Transformers Became the Grid’s Chokepoint

    Large power transformers are among the least glamorous and most consequential machines in the economy. They are heavy, highly engineered, often custom-built to a specific substation’s requirements, and produced by a relatively small number of manufacturers worldwide. Capacity to build them cannot be added quickly: it requires specialized factories, scarce materials such as grain-oriented electrical steel, and skilled workers who take years to train.

    The US grid spent roughly two decades with flat electricity demand, and the supply chain sized itself accordingly — tuned for steady replacement of aging units, not for a demand shock. When data center load growth, electrification, and grid-hardening programs all began pulling on that thin manufacturing base at once, order backlogs stretched and utilities found themselves queuing for hardware. The scramble Reuters describes is the predictable result of a just-in-time supply chain meeting a step change in demand.

    When Equipment Lead Times Set the Data Center Schedule

    For data center developers, this crunch changes what “time to power” means. A site can have land, fiber, permits, and even a utility willing to serve it, and still wait on a transformer delivery slot. Interconnection — the process of physically and contractually tying a new load into the grid — increasingly depends less on paperwork and more on whether the required substation equipment physically exists.

    That reality is reshaping behavior on both sides of the meter. Utilities are reported to be securing equipment earlier and more aggressively, which effectively shifts them from reactive procurement to strategic stockpiling. Large data center operators, for their part, have strong incentives to lock in capacity years ahead, pre-order long-lead equipment themselves, or favor sites where grid infrastructure already exists — one reason established carrier hotels and campuses with existing substation capacity have gained strategic value relative to greenfield sites.

    The Economics of Scarcity: Who Absorbs the Cost

    Scarcity moves pricing power toward manufacturers. Electrical-equipment makers with transformer and switchgear capacity are in an unusually strong position, and the open question is how much they will invest in expansion — factories are decade-scale bets, and executives remember the last long stretch of flat demand. Utilities, meanwhile, typically recover equipment costs through regulated rates, which means sustained price inflation in grid hardware eventually reaches ratepayers and invites regulatory scrutiny over how much of the buildout data center customers should fund directly.

    Among data center players, scarcity favors scale and incumbency. Hyperscale operators can pre-purchase equipment, sign long-term supply agreements, and absorb schedule risk in ways smaller developers cannot. If the crunch persists, expect it to act as a filter: well-capitalized projects with early equipment commitments proceed, while speculative projects — announced capacity without secured power and hardware — quietly slip or die. That could rationalize an overheated development pipeline, but it also raises barriers to entry across the industry.

    What Could Break the Bottleneck

    Several paths out exist, none fast. Manufacturers can and do add capacity, but new production lines take years to reach output. Standardizing transformer designs — reducing the custom engineering in each order — could raise effective throughput. Utilities can extend the life of existing units, share spares, and prioritize deployments. On the demand side, data centers that bring their own generation or agree to flexible operation reduce the immediate grid equipment burden.

    The honest assessment is that this is a multi-year imbalance. Equipment supply is a lagging system responding to a leading demand signal, and the gap between them is where project delays, price escalation, and strategic maneuvering will play out. For infrastructure operators, the practical takeaway is that secured power and in-hand electrical equipment are now assets in their own right, worth nearly as much as the buildings around them.

    Background

    For most of the 2000s and 2010s, US electricity demand barely grew, thanks to efficiency gains offsetting economic expansion. That era ended as data centers — driven most recently by AI training and inference workloads — joined manufacturing reshoring and electrification as major new sources of load. Utilities, regulators, and grid operators have spent the past several years revising demand forecasts upward and confronting the fact that generation, transmission, and the equipment supply chain were all sized for a slower world.

    Concerns about transformer supply predate the AI boom — the aging of the US transformer fleet and the concentration of manufacturing capacity have been discussed in grid-security circles for years — but data center growth has converted a slow-burning replacement problem into an acute procurement race. The July 2026 Reuters report captures that shift from the utilities’ side of the table.

    Source: US power companies scramble to secure equipment as surging data center demand strains supplies — Reuters reporting, July 8, 2026, on utilities competing for transformers and switchgear amid data-center-driven load growth.

  • Oregon Approves PGE’s 29.7% Data Center Rate Hike Under Landmark POWER Act

    Oregon Approves PGE’s 29.7% Data Center Rate Hike Under Landmark POWER Act

    Oregon regulators have approved a 29.7% electricity rate increase for data centers served by Portland General Electric (PGE), the state’s largest utility, as reported by Oregon Public Broadcasting on July 6, 2026. The decision is the first major rate action taken under Oregon’s landmark POWER Act, a 2025 law that directed regulators to place large energy users such as data centers into their own rate class so that the costs of serving them are not spread across households and small businesses.

    Executive Summary

    The approval makes Oregon one of the first states to move from debating data-center cost allocation to actually pricing it. Under the POWER Act — passed in 2025 amid rapid data-center load growth and rising residential bills — utilities must charge very large customers rates that reflect the full cost of serving them, including the new generation and transmission their demand triggers. The 29.7% figure now approved for PGE’s data-center class is the concrete output of that mandate.

    Why it matters: electricity has become the gating resource for AI and cloud expansion, and the question of who funds grid upgrades — the data centers driving demand, or all ratepayers — is now the central fight in utility regulation. Oregon has produced a working template, with a specific number attached, that commissions and legislatures in Virginia, Georgia, Ohio, Texas and elsewhere are likely to study closely.

    Who Pays for the AI Buildout Just Got a Concrete Answer

    For most of the past century, utilities spread the cost of new infrastructure across all customers on the theory that everyone benefits from a stronger grid. Data centers broke that logic: a single hyperscale campus can demand as much power as a small city, arriving faster than utilities can build generation and wires. When those costs land in general rates, households effectively subsidize some of the world’s largest companies. Oregon’s POWER Act rejected that outcome by mandating a separate rate class — a distinct pricing category with its own cost-based rates — for large energy users.

    The 29.7% increase is the first hard number to emerge from that framework. It represents a regulator’s judgment, tested through a formal rate proceeding, of what cost-causation pricing for data centers actually looks like at PGE. Whether one views the number as fair depends on the underlying cost studies, which the reporting summarized here does not detail — but the structural shift is unambiguous: growth-driven costs are being assigned to the customers driving the growth.

    A Template Other States Will Study — and Contest

    Regulators across the country are wrestling with the same problem, mostly through case-by-case special contracts with individual data-center customers. Oregon instead wrote the principle into statute and applied it class-wide, which offers predictability but less flexibility. Expect both sides of the national debate to cite this decision: consumer advocates as proof that ratepayer protection is achievable, and data-center developers as evidence of rising regulatory risk in some markets.

    The competitive question is real. Oregon, particularly the Portland-Hillsboro area that PGE serves, built a significant data-center cluster on the strength of relatively inexpensive Northwest power and long-standing tax incentives. A nearly 30% jump in the power line-item — often the largest operating cost of a modern facility — changes site-selection math. States hungry for data-center investment may market themselves against Oregon’s approach; states worried about residential bills may copy it. Either way, the era of uniform, geography-blind data-center power pricing is ending.

    The Economics Cut Both Ways

    For utilities, a dedicated large-load class is double-edged. It insulates existing customers and reduces political backlash against growth, but it also raises the price of the very load that funds new investment. If data-center operators respond by self-supplying — building on-site generation, contracting directly with power producers, or siting behind other utilities — PGE could face slower load growth than planned, and the fixed costs of any already-committed infrastructure would need a home.

    For operators, the decision reinforces a trend already visible across the industry: power strategy is now a first-order business function, not a facilities detail. Companies that locked in long-term supply arrangements, invested in efficiency, or diversified their geographic footprint are better positioned than those that assumed grid power would stay cheap and socialized. The Oregon decision does not end data-center growth in the state — but it prices that growth honestly, and honest prices change behavior.

    Background

    Oregon became a data-center destination over the past two decades thanks to relatively inexpensive Pacific Northwest power, a mild climate, strong fiber routes, and generous local tax incentives — attracting major cloud and internet companies to clusters around Hillsboro in PGE territory and along the Columbia River. As AI workloads accelerated demand in the 2020s, utilities projected unprecedented load growth while residential electric bills climbed, fueling a political backlash over who should fund grid expansion.

    The POWER Act, passed in 2025, was Oregon’s answer: separate very large energy users into their own rate class and charge them the full cost of serving them. The rate decision reported here is the first major application of that law, moving the cost-allocation debate from principle to an approved price.

    Source: Oregon approves PGE’s 29.7% rate hike for data centers under landmark law — Oregon Public Broadcasting report on the first major rate decision under Oregon’s POWER Act, published July 6, 2026.

  • Galaxy’s Helios Phase I Delivers 133 MW of AI Capacity to CoreWeave

    Galaxy’s Helios Phase I Delivers 133 MW of AI Capacity to CoreWeave

    Galaxy announced on July 5, 2026 that it has completed Phase I of its Helios data center campus in West Texas, delivering 133 megawatts (MW) of critical IT load to CoreWeave, the AI-focused cloud provider. Critical IT load refers to the power available to the computing equipment itself — servers and GPUs — as distinct from the total power a facility draws for cooling and other overhead.

    The completion converts a site that began life as a Bitcoin mining campus into dedicated AI infrastructure under Galaxy’s long-term lease arrangement with CoreWeave, one of the most prominent examples of the crypto-to-AI conversion trend reshaping the data center market.

    Executive Summary

    Galaxy, the digital assets and data center infrastructure firm, has finished the first phase of its Helios campus buildout and handed over 133 MW of critical IT load to its anchor tenant CoreWeave. Phase I completion moves the project from promise to delivery: Helios is now an operating revenue-generating AI data center rather than a conversion story on a slide deck.

    The milestone matters beyond Galaxy. Helios is the flagship test case for whether former cryptocurrency mining sites — which come with grid interconnections and power contracts already in place — can be economically retrofitted to the far more demanding standards of AI training and inference infrastructure. Delivering a first phase at this scale suggests the model can work, at least for sites with strong power positions.

    For CoreWeave, the delivery adds substantial contracted capacity at a time when access to powered land and energized shells — not GPUs — is widely seen as the binding constraint on AI cloud growth.

    Why Crypto Sites Became AI Real Estate

    The most valuable asset in data center development today is not land or buildings but secured power: a grid interconnection agreement and the megawatts behind it. Bitcoin mining operators spent the late 2010s and early 2020s locking up exactly that, often in low-cost power markets like West Texas. When AI demand exploded, those interconnections became worth far more serving GPUs than mining rigs, because AI tenants sign long-term leases at data center economics rather than riding volatile crypto margins.

    Galaxy’s Helios campus, acquired from a Bitcoin mining operator, is the highest-profile execution of that arbitrage. The conversion is not trivial — AI facilities require far denser power delivery, liquid or advanced air cooling, and enterprise-grade redundancy that mining sites never needed — but the timeline still beats greenfield development, where new grid interconnection requests can queue for years.

    What 133 MW Actually Buys

    133 MW of critical IT load is a substantial block of capacity by any historical standard — a few years ago it would have ranked among the larger single-tenant deployments in the world. In the AI era it is best understood as a first tranche: large frontier training clusters are increasingly specified in the hundreds of megawatts, and operators including Galaxy have discussed multi-phase expansion at Helios well beyond Phase I.

    Because the load is contracted to a single tenant, the economics resemble a triple-net real estate deal more than a retail colocation business: predictable lease revenue over a long term, with Galaxy carrying development and delivery risk and CoreWeave carrying utilization risk. That structure has become the dominant template for AI data center finance because lenders can underwrite the lease.

    Winners, Losers, and the Competitive Field

    The clearest winners are holders of energized or near-energized power positions — converted mining sites, utilities with spare interconnection capacity, and developers who queued early. CoreWeave benefits by adding capacity faster than greenfield timelines would allow, supporting its competition with hyperscale clouds for AI workloads. The pressure lands on developers still waiting in interconnection queues, and on regions whose grids cannot absorb gigawatt-class requests.

    The open competitive question is durability. Conversion sites tend to sit in remote, power-rich locations, which suits training workloads that tolerate latency. If the market shifts toward inference — which favors proximity to users — the value of remote megawatts could be repriced. Phase I’s completion answers the execution question; it does not settle the location question.

    Background

    Helios began as one of the larger Bitcoin mining campuses in the United States before Galaxy acquired the site and redirected it toward AI and high-performance computing. Galaxy subsequently signed long-term lease agreements making CoreWeave the campus’s anchor tenant, with capacity to be delivered in phases — Phase I, now complete, being the first.

    The conversion sits inside a broader industry shift: as demand for AI compute outran the pace of new grid connections, sites with existing power infrastructure — many of them crypto mining facilities in Texas and the Mountain West — became prime targets for repurposing. Helios is widely watched as the leading proof point for whether that playbook delivers at scale.

    Source: Galaxy Completes Phase I of Its Helios Data Center Campus, Delivering 133 Megawatts of Critical IT Load to CoreWeave — PR Newswire press release, July 5, 2026, announcing Phase I completion at Galaxy’s West Texas AI campus.

  • National Grid’s $1.75B Joulent Deal: When Interconnect Delays Force Utilities to Buy

    National Grid’s $1.75B Joulent Deal: When Interconnect Delays Force Utilities to Buy

    National Grid has struck a $1.75 billion deal with Joulent, according to a Data Center Knowledge report published July 1, 2026. The report frames the transaction as a response to mounting interconnection delays driven by AI data center demand — utilities, unable to connect new load fast enough through conventional build-out, are increasingly spending to acquire capacity and capability rather than queue for it.

    Executive Summary

    The reported transaction pairs one of the world’s largest electricity transmission and distribution operators with Joulent in a deal valued at $1.75 billion. The headline framing is the important part: the deal is attributed not to routine portfolio strategy but to AI interconnect delays — the growing backlog of requests to connect large new loads and generation to the grid, a process that in many regions now takes years.

    Why it matters: if the reporting’s framing holds, this is a data point in a broader shift. Utilities have historically grown connection capacity by building — new substations, transformers, transmission lines — on regulated timelines. When AI-driven demand outruns those timelines, acquisition becomes the faster path. A $1.75 billion commitment suggests National Grid sees the capacity crunch as durable, not a passing spike. That said, the available source is a single news headline; the deal’s structure, scope, and closing conditions are not detailed in the material we can verify, and readers should treat specifics beyond the reported figure and parties with appropriate caution.

    Why Buying Beats Building When the Queue Is the Bottleneck

    Interconnection — the engineering and regulatory process of physically wiring a new data center, factory, or power plant into the grid — has become one of the defining constraints of the AI build-out. Studies, permitting, equipment procurement, and construction stack into multi-year waits in many markets, and lead times for critical hardware such as large power transformers and high-voltage switchgear have stretched dramatically since the early 2020s. In that environment, anything that already exists — installed equipment, an established delivery capability, a workforce, a manufacturing slot — carries a scarcity premium.

    A utility that spends $1.75 billion to acquire capacity or capability it would otherwise wait years to build is making a straightforward time-for-money trade. The economics can work because the cost of delay is now enormous on both sides of the meter: hyperscale customers measure the cost of a stranded, unpowered data center shell in the millions per month, and utilities that cannot connect large customers forgo years of revenue from their fastest-growing load class.

    National Grid’s Position in the AI Load Story

    National Grid sits at the center of this dynamic in two major markets. It operates the high-voltage transmission network in England and Wales — where grid connection queues became a widely acknowledged national bottleneck and the subject of regulatory reform efforts — and it owns large regulated electricity and gas utilities in New York and Massachusetts, in the demand path of the US Northeast’s data center and electrification growth. Few companies feel interconnection pressure from as many directions at once.

    That context makes the reported deal legible even without full details: a transmission-heavy utility facing connection backlogs on two continents has clear motives to secure capacity, equipment supply, or delivery capability by acquisition. It also carries risk. Large deals struck during a scarcity cycle can look expensive if the cycle turns — if AI load forecasts moderate or supply chains normalize, capacity bought at peak-crunch prices may earn a thinner return than capacity built patiently through the regulated process.

    What $1.75 Billion Signals — and What It Doesn’t

    The figure itself is the strongest signal in the reporting. Utilities are conservative, regulated businesses; a commitment of this size typically requires board conviction that the underlying driver — here, sustained AI-driven demand outpacing conventional grid expansion — will persist long enough to pay back the investment. In that sense the deal is a vote of confidence in continued data center growth, made by a party with unusually good visibility into actual connection requests rather than press-release pipelines.

    What the number does not tell us is the mechanism. “Buying your way to capacity” can mean acquiring a company outright, purchasing assets, locking up equipment manufacturing capacity, or securing services under a long-term contract — and each has very different implications for competitors, regulators, and customers. The single-source material available does not specify which of these the National Grid–Joulent transaction is, what Joulent brings to the arrangement, or how the spend will be recovered. Those distinctions matter: an acquisition that removes a supplier or contractor from the open market can tighten conditions for every other utility shopping in it, while a capacity contract merely reallocates near-term supply.

    Background

    National Grid built its position over decades as the operator of Great Britain’s electricity transmission backbone before expanding into the US Northeast, where it serves millions of electricity and gas customers in New York and Massachusetts. In both markets it entered the mid-2020s facing an unprecedented problem: connection requests from data centers, electrified transport, and new generation arriving faster than networks could be studied, permitted, and built, prompting queue-reform efforts by regulators on both sides of the Atlantic.

    The AI boom sharpened that squeeze into a defining industry constraint. Transformer and switchgear lead times stretched, hyperscale campuses began requesting connections measured in hundreds of megawatts, and ‘time to power’ displaced real estate as the data center industry’s scarcest resource — the backdrop against which a utility paying $1.75 billion to shortcut the queue becomes a rational, if notable, move.

    Source: AI Interconnect Delays Spur $1.75B National Grid-Joulent Deal — Data Center Knowledge report, July 1, 2026, on National Grid’s $1.75 billion deal with Joulent amid AI-driven grid interconnection backlogs.

  • PJM Moves to Manage Data Center Demand: A Turning Point for AI Power

    PJM Moves to Manage Data Center Demand: A Turning Point for AI Power

    Reuters reported on June 30, 2026 that PJM Interconnection — the largest power grid operator in the United States, coordinating electricity across 13 states and the District of Columbia for roughly 65 million people — is moving toward actively managing data center demand on its system. The report signals a shift from treating data centers as ordinary customers whose consumption must simply be served, toward a framework in which the grid operator can shape when and how much power the largest new loads draw.

    Details of the mechanism, timeline, and scope were not spelled out in the headline announcement, but the direction alone is consequential: PJM’s territory includes Northern Virginia’s “Data Center Alley,” the densest concentration of data centers in the world, and the region at the center of the AI-driven surge in U.S. electricity demand.

    Executive Summary

    According to Reuters, PJM is taking steps toward managing data center demand rather than passively absorbing it. For decades, U.S. grid planning worked on a simple premise: customers decide how much electricity they need, and the grid builds to serve it. AI data centers — single facilities that can draw hundreds of megawatts, comparable to a small city — have broken that premise. Interconnection queues are backed up, capacity prices in PJM’s markets have surged, and the gap between how fast data centers can be built (one to two years) and how fast power plants and transmission can be built (five to ten years) keeps widening.

    Moving to “manage” that demand means the operator of America’s biggest wholesale power market is preparing tools — potentially ranging from voluntary demand-response participation to conditions on new large-load interconnections to curtailment provisions, though the report does not specify which — to control the timing and firmness of data center consumption. That matters far beyond PJM’s footprint: as the largest grid and the home of the world’s biggest data center cluster, PJM’s rules tend to become the template other regions study.

    For the data center industry, the message is that access to the grid is no longer an unconditional entitlement. Flexibility — the ability to shift, shed, or self-supply load — is becoming a bargaining chip in getting connected at all.

    From Passive Host to Active Manager

    Grid operators like PJM are regional transmission organizations (RTOs): nonprofit entities that run the wholesale electricity market and the high-voltage network across their territory, under rules approved by federal regulators. Historically, their job was to forecast demand and make sure supply met it. Demand itself was treated as a given. A move toward managing data center demand inverts that relationship for the first time at this scale — the grid operator would have a say in how the largest customers consume, not just how generators produce.

    The trigger is arithmetic. Load growth in PJM was essentially flat for nearly two decades; AI data centers ended that era abruptly. When a single campus can request as much power as a steel mill or a small utility’s entire service territory, and dozens of such requests arrive at once, the traditional “build to serve” model produces either reliability risk or enormous costs socialized across all ratepayers. Managing demand is the third option: make the new load itself part of the reliability solution.

    The Economics of Curtailable Compute

    The core idea behind demand management is that not every megawatt-hour of computing is equally urgent. AI training runs can, in principle, pause or shift in time; some workloads can migrate between facilities in different regions. If data centers agree to reduce consumption during the few dozen hours a year when the grid is most stressed, the system needs less peak capacity — which is exactly the product whose price has been surging in PJM’s capacity auctions, the market where power plants are paid to be available.

    The unresolved tension is that most data center operators sell their customers uninterrupted uptime, and inference workloads serving live users are far harder to pause than training. Whether flexibility is genuinely available at scale — and at what price data center operators would sell it — is the open economic question. If PJM’s framework rewards flexible loads with faster interconnection or lower costs, it effectively creates a market price for interruptibility, and data center designs will adapt to capture it: more batteries, more on-site generation, more workload-orchestration software.

    Winners, Losers, and the Ratepayer Question

    Developers with flexible-by-design facilities, on-site generation, or storage stand to gain priority in a demand-managed regime. Operators marketing strict 24/7 firmness with no curtailment tolerance may face slower interconnection or higher costs. Utilities and generators face a subtler effect: managed demand blunts the extreme scarcity that has driven capacity prices up, which helps consumers but trims the windfall that scarcity was delivering to existing power plants.

    For households and businesses in PJM’s 13-state footprint, the stakes are direct. Capacity costs flow into retail electricity bills, and the politics of ordinary ratepayers subsidizing infrastructure for the world’s wealthiest technology companies have grown sharp. A credible demand-management framework is partly a political instrument: it lets PJM tell states and consumers that data centers are being asked to carry reliability risk, not just impose it. Whether the framework has real teeth — mandatory obligations versus voluntary programs — will determine whether that assurance holds up.

    A Template Other Grids Will Study

    PJM is not acting in a vacuum. Texas’s ERCOT grid, the other major destination for large flexible loads, has been developing its own approach to interconnecting and, when necessary, curtailing very large customers. When the two biggest data center markets in the country both condition grid access on demand flexibility, it stops being an experiment and becomes the emerging national norm. Data center site selection, financing models, and colocation contracts will all have to price in a world where the grid can ask the largest computers on Earth to throttle down.

    Background

    PJM Interconnection, headquartered in Pennsylvania, grew from a 1927 power pool into the largest regional transmission organization in the United States, dispatching generation and running wholesale power markets across a footprint from Illinois to the mid-Atlantic. Its territory includes Northern Virginia, where decades of fiber density and proximity to federal and enterprise customers created “Data Center Alley” — the largest data center market in the world.

    The generative-AI boom that accelerated from 2023 onward transformed data centers from a steady, modest slice of electricity demand into the dominant driver of U.S. load growth, ending a long era of flat consumption. PJM’s capacity auctions delivered record-high prices as demand forecasts jumped, interconnection requests piled up, and state officials began questioning who should bear the cost. The June 2026 move toward managing data center demand is the institutional response to that collision between AI’s growth curve and the grid’s construction timelines.

    Source: Biggest US power grid PJM moves towards managing data center demand — Reuters report, June 30, 2026, on PJM Interconnection’s move toward actively managing data center electricity demand.

  • DOE Emergency Order for PJM Ahead of Heatwave Signals a Grid Under Strain

    DOE Emergency Order for PJM Ahead of Heatwave Signals a Grid Under Strain

    The US government has issued an emergency order covering PJM Interconnection — the largest electric grid operator in the United States — ahead of a heatwave expected to drive electricity demand toward the edge of available supply, Reuters reported on June 30, 2026. Emergency orders of this kind allow the Department of Energy to temporarily relax normal operating constraints so that generators can run at maximum output to keep the lights on.

    Executive Summary

    According to the Reuters report, federal authorities acted preemptively: the order was issued as the heatwave loomed, not after the grid had already buckled. That timing matters. Emergency authority — typically exercised under Section 202(c) of the Federal Power Act, which lets the Energy Secretary direct generators to operate notwithstanding permits or other limits — was historically reserved for rare, acute crises such as hurricanes or sudden plant failures.

    That such an intervention now precedes a forecastable summer weather event suggests the buffer between peak demand and available generation in PJM’s territory has grown uncomfortably thin. PJM coordinates power for roughly 65 million people across 13 states and the District of Columbia — including Northern Virginia, the densest data-center market on Earth — so an emergency footing on this grid is a material signal for the entire digital-infrastructure industry.

    When Emergency Powers Become Routine Tools

    An emergency order is, by design, an extraordinary instrument. It can authorize power plants to exceed environmental or operational limits, keep units scheduled for retirement running, and compel generation that market signals alone would not produce. Using it in anticipation of hot weather — one of the most predictable stresses a grid faces — indicates that ordinary market and reliability mechanisms are no longer producing enough headroom on their own. Similar orders were issued for PJM and other regions during heat events in prior summers, so the June 2026 action fits an emerging pattern rather than standing as a one-off.

    The pattern is the story. Each individual order is defensible as prudent risk management; a sequence of them amounts to the federal government repeatedly bridging a structural gap between demand growth and supply additions. That gap has causes on both sides of the ledger: large thermal plants retiring faster than replacement capacity comes online, interconnection queues that delay new generation for years, and demand rising after two decades of near-flat load.

    AI Load Growth Meets a Tightening Grid

    PJM sits at the center of the demand-growth debate because its footprint includes Northern Virginia’s ‘Data Center Alley,’ along with fast-growing campuses in Ohio, Pennsylvania, and Maryland. Grid planners across the country have sharply raised load forecasts, driven in large part by AI-oriented data centers, electrification, and new manufacturing. PJM’s own capacity auctions — the market that pays generators to be available during peaks — have cleared at record-high prices in recent cycles, a direct financial symptom of scarcity.

    A heatwave is where these abstractions become physical. Air-conditioning load peaks at exactly the moment thermal plants lose efficiency in the heat, and data-center cooling demand rises in parallel. When the margin for error narrows, operators lean on emergency tools. For the industry we cover, the lesson is blunt: electricity availability, not land or fiber, is now the binding constraint on digital-infrastructure growth in America’s largest power market.

    What It Means for Data-Center Operators and Their Customers

    For operators, recurring grid emergencies raise both operational and reputational stakes. Operationally, facilities in PJM territory should expect more frequent conservation appeals, demand-response calls, and scrutiny of backup-generation readiness during peak season. Reputationally, data centers are increasingly cast as the face of load growth; every emergency order sharpens public and regulatory questions about who pays for grid stress and whether large loads should be required to be curtailable or bring their own generation.

    The likely winners in this environment are firms that treat power as a first-class engineering problem: those with flexible-load capability, on-site or contracted generation, long-dated capacity positions, and sites in regions with genuine surplus. The exposed parties are speculative projects counting on grid interconnection timelines and power prices that no longer reflect reality. Utilities and generators in PJM, meanwhile, gain leverage — scarcity is lucrative for whoever owns dispatchable megawatts.

    Background

    PJM Interconnection, founded as a utility power pool in 1927, evolved into the largest competitive wholesale electricity market in the United States, coordinating generation and transmission across the Mid-Atlantic and parts of the Midwest. Its footprint includes Northern Virginia’s data-center corridor, which has made PJM the frontline grid for AI-era load growth. Section 202(c) of the Federal Power Act gives the Department of Energy authority to order emergency generation during grid crises — a power used sparingly for decades but invoked more frequently in recent years as plant retirements, slow interconnection of new resources, and surging demand forecasts have narrowed the system’s reserve margins.

    Source: US issues emergency order for PJM Interconnection as heatwave looms — Reuters report, June 30, 2026, on federal emergency action to shore up the largest US grid ahead of extreme heat.

  • Brookfield, Bloom Energy Expand AI Power Partnership to $25 Billion

    Brookfield, Bloom Energy Expand AI Power Partnership to $25 Billion

    Brookfield and Bloom Energy announced on June 29, 2026 that they are expanding their AI infrastructure partnership to $25 billion — a fivefold increase over the original framework — to build and finance rapid power deployment for AI data centers. The expanded arrangement pairs Bloom’s solid oxide fuel-cell technology with Brookfield’s infrastructure capital.

    Executive Summary

    Bloom Energy, the fuel-cell manufacturer, and Brookfield, one of the world’s largest infrastructure investors, have scaled their partnership from an original framework — implied by the announcement’s “fivefold” language to have been on the order of $5 billion — to $25 billion. The stated purpose is to build and finance “rapid power” for AI infrastructure: on-site electricity generation that can be deployed faster than utility grid connections.

    The announcement matters because electricity availability, not chips or land, has become the binding constraint on AI data-center construction. A $25 billion commitment of this shape signals that major infrastructure capital now treats on-site fuel-cell generation as a bankable asset class rather than a niche backup option. That said, the release as reported gives a headline dollar figure without megawatt targets, named customers, or deployment timelines — so the scale of actual near-term power delivery remains to be demonstrated.

    Why Fuel Cells Are Jumping the Grid Queue

    The core problem this partnership targets is speed. In many major data-center markets, a new facility requesting a large grid connection can wait years for utilities to build the transmission and generation needed to serve it — a delay measured in lost AI product cycles. On-site generation sidesteps that queue. Bloom’s solid oxide fuel cells convert fuel, typically natural gas, into electricity through an electrochemical reaction rather than combustion, and they arrive as factory-built modules that can be installed in months rather than the multi-year timelines of large power plants or grid upgrades.

    That “speed-to-power” pitch has become the dominant selling point across the AI power market — gas turbines, batteries, and behind-the-meter deals all compete on the same axis. Fuel cells’ specific claim is modularity and siting flexibility: they are quiet, produce no combustion emissions like NOx at the point of generation, and can be permitted in places where a turbine plant could not. The trade-off is cost per megawatt-hour and dependence on fuel supply, which is why financing structure matters as much as technology.

    The Capital Stack Behind the Megawatts

    The division of labor is the interesting part. Bloom manufactures and services the equipment; Brookfield brings the balance sheet. In a typical arrangement of this kind, the infrastructure investor owns the generating assets and sells power or capacity to data-center operators under long-term contracts, so the data-center customer avoids a large upfront capital outlay. For Bloom, a deep-pocketed financing partner converts its technology into an offering that can compete for hyperscale-sized deals it could never finance from its own balance sheet.

    For Brookfield, fuel-cell fleets serving AI campuses look like classic infrastructure: long-lived assets, contracted revenue, and a customer base — AI compute operators — currently willing to pay a premium for firm power delivered quickly. Growing the framework fivefold within roughly a year of the original announcement suggests the partners believe demand from AI builders exceeds what the initial commitment could serve. It is a strong demand signal, though announced frameworks and deployed megawatts are different things.

    What a Fivefold Scale-Up Signals — and What It Doesn’t

    A $25 billion figure invites careful reading. Partnership frameworks of this kind typically describe a ceiling — capital the partners intend to deploy if projects materialize — rather than contracted orders. The announcement as reported does not specify how much is committed versus targeted, how much power it represents, or over what period. Until customer contracts and megawatt figures are disclosed, the number is best understood as a statement of ambition backed by a credible financier, not a backlog.

    Competitively, the deal sharpens the contest to power AI. Utilities and grid operators risk losing their largest new customers to behind-the-meter generation; gas-turbine suppliers, battery vendors, and small modular reactor developers are chasing the same load. For data-center operators, more credible power options mean more negotiating leverage — and for the industry’s critics, more scrutiny of what fuels that power. Fuel cells running on natural gas still emit carbon dioxide, so the climate profile of this buildout will depend on fuel sourcing choices the announcement does not detail.

    Background

    Bloom Energy, founded in 2001 and headquartered in California, went public in 2018 and built its business selling solid oxide fuel-cell “Energy Servers” to commercial, industrial, and utility customers seeking reliable on-site power. Brookfield is a global asset manager with hundreds of billions of dollars across infrastructure, renewable power, and real estate, and has been among the most aggressive institutional investors in AI-related infrastructure. The two first announced an AI-focused partnership in late 2025, part of a wider industry wave in which data-center developers turned to behind-the-meter generation — fuel cells, gas turbines, and eventually nuclear — as utility interconnection queues stretched to multiple years in key markets.

    Source: Brookfield and Bloom Energy Expand AI Infrastructure Partnership to $25 Billion — Bloom Energy announcement, June 29, 2026, reporting a fivefold expansion of the companies’ AI power partnership.