Today’s featured EIF speaker: Lisa Hough, who sits at the intersection of U.S. energy infrastructure, Bitcoin, and AI/HPC data center deployment. She brings institutional energy markets experience from Enron Capital & Trade, Phibro Energy, and PG&E National Energy Group, plus digital asset and regulatory expertise from Unchained and Custodia Bank, and is currently involved in structuring large-scale AI infrastructure initiatives including a roughly 2 GW campus in West Texas.

EIF Dallas (July 23-24) is a high-signal, curated conference convening leaders across energy, compute infrastructure, and capital to map the next decade of power and data center buildout.

Apply to speak or explore sponsorships for private meeting rooms, branding & more.

For much of the past halving cycles, bitcoin miners paid a steep premium to access debt markets—from high single-digit to high double-digit interest rates tied to volatile mining cash flows.

That premium appears to be now compressing.

A wave of project-backed financings tied to AI and high-performance computing (HPC) shows borrowing costs moving decisively lower. Deals that cleared above 9% in 2025 are now landing closer to the 6%–7% range in 2026, according to recent bond offerings across Core Scientific, Cipher Digital, Applied Digital, Hut 8 and more.

At the same time, bitcoin-backed financing is repricing lower too, with Riot Platforms recently cutting its effective borrowing cost to ~6% and Hut 8 refinancing its Coinbase credit with a new deal from FalconX to slash the cost from 9% to 7%.

This isn’t just better market conditions. It’s a repricing of the underlying collateral.

Then: ASICs, Volatility—and Double-Digit Risk Premiums

To see how far the market has moved, it helps to look back at the last cycle.

In 2021–2022, most mining debt was backed by Bitcoin ASIC machines—a collateral base that was highly cyclical, sensitive to bitcoin price and network difficulty, and prone to rapid depreciation.

The result: borrowing costs that routinely sat in the 12%–16% range, with some deals even higher depending on lender and structure. Lenders like Genesis, BlockFi and Celsius dominated the space, extending credit against rigs and bitcoin holdings in what was effectively a crypto-native shadow banking system. It worked—until it didn’t, and dragged big lenders underwater altogether when bitcoin prices fell and liquidity tightened in 2022.

Now: Contracted Compute Is Replacing Machine Collateral

Fast forward to today, and the collateral has fundamentally changed.

Instead of ASICs, lenders are underwriting long-term AI compute contracts, colocation leases with escalators and power-secured data center infrastructure.

That shift moves the credit profile away from commodity exposure and toward infrastructure-style cash flows.

CoreWeave’s trajectory is emblematic—moving from ~9% financing in 2025 to the low-6% range, supported by contracted revenue and investment-grade positioning.

Miners making the AI pivot are now being evaluated less like speculative operators—and more like emerging data center credits.

Miner Weekly

From MinerFi to ComputeFi: CoreWeave’s $8.5B Loan Rewrites the Playbook

Read more

a month ago · 7 likes · BlocksBridge Consulting

The Spread Tells an Interesting Story

Even as headline rates fall, dispersion remains notably—and increasingly informative.

From TheEnergyMag’s chart above:

• CoreWeave: low-6% range

• Hut 8: low-6%

• Cipher Digital: down from low-7% to low-6%

• Applied Digital: down from low-9% to high-6%

• Core Scientific: still pricing at high-7%

One of the key developments over the past year is the emergence of hyperscaler-linked credit support.

Hut 8, Cipher and TeraWulf have disclosed that Google provides financial backstop mechanisms tied to Fluidstack’s lease obligations—effectively anchoring those projects to hyperscaler demand.

That support matters and has helped pull borrowing costs into the mid- to high-single-digit range, tighter than what miners would achieve on a standalone basis.

This dynamic is clearest in the spread between Core Scientific and Hut 8. Both raised capital against AI infrastructure at roughly the same time with credible counterparties (Core Scientific with CoreWeave and Hut 8 with Fluidstack backed by Google). Yet Core Scientific continues to pay more.

AI Is Doing What Bitcoin Couldn’t

For years, miners argued they were infrastructure businesses. Markets didn’t buy it—because the collateral didn’t support the claim.

AI changed that.

With hyperscalers expected to spend roughly $700 billion on AI infrastructure this year, compute capacity has become a strategic asset. Power, land and interconnection—long the domain of bitcoin miners—are now in short supply.

Miners—who spent a decade building in exactly those constraints—are now being re-rated accordingly.

Hardware and Infrastructure News

• Riot Expands AMD Deal to 50 MW, Cuts Coinbase Loan Rate to 6.15% in AI Push

• IREN Rallies 11% on Sweetwater Energization Milestone Amid AI Power Race

• Peter Thiel Leads $140 Million Series B for Panthalassa’s Ocean-Powered AI Infrastructure

• Nscale Expands Microsoft AI Deal With 66,000 Rubin GPUs in Portugal

• Soluna Closes $53M Wind Farm Acquisition and Reports April 2026 Operational Progress

• American Bitcoin Secured Bitmain Miners at Implied 44% Discount Using BTC Collateral

Corporate News

• $70B Investor Coatue Expands Into AI Data Center Land Grab

• IREN to Acquire Mirantis in $625M Deal to Expand AI Cloud Software Stack

• Applied Digital Completes AI Cloud Spinout Into New Nasdaq Firm ChronoScale

• Nine Texas residents sue MARA, expanding Granbury bitcoin mine noise fight

• Core Scientific Reveals $232M Texas Acquisition, Sold 2,385 Bitcoin to Fund AI Expansion

• Hut 8 Surges to All-time Highs After Landing Second AI Data Center Lease Worth $9.8 Billion

Financial News

• Microsoft Committed $7 Billion Upfront in Nebius AI Deal

• Hut 8 Refinances Bitcoin-Backed Credit Facility with $200 Million FalconX Deal

• Applied Digital Secures $300 Million Bridge Loan for North Dakota AI Data Center

• Cipher Secures $200M Credit Line as Amazon AI Retrofit Cuts Bitcoin Hashrate

• Solaris Energy Prices $1.3B Notes at 6.375% as AI Power Buildout Accelerates

Email us

Today’s featured EIF speaker: Mike Alfred, Founder and Managing Partner of Alpine Fox LP, a value-equities and bitcoin investment partnership, and a board director at IREN and Bakkt. He previously co-founded Digital Assets Data and led it through its acquisition by NYDIG, bringing deep experience across bitcoin markets, public-company governance, and the energy plus compute investment landscape.

EIF is a high-signal, curated conference convening leaders across energy, compute infrastructure, and capital to map the next decade of power and data center buildout.

Apply to speak or explore sponsorships for private meeting rooms, branding & more

The AI boom is rapidly redrawing the map of U.S. data center infrastructure — but the most important signal may lie not just in where facilities exist today, but where power is being requested next.

A dataset compiled from public records by Datacenter.fyi shows there are 3,995 existing data centers across the United States, with roughly 2,900 operational. Virginia and Texas dominate that footprint, with 575 and 410 sites respectively, forming the backbone of the country’s digital infrastructure.

But the next phase of growth is already taking shape.

Texas leads the nation with 336 planned data center projects, outpacing every other state and signaling its emergence as the primary expansion frontier. Virginia follows with 203 projects, while secondary markets such as Ohio (68 planned) and Indiana (37 planned) are gaining traction as developers look beyond traditional hubs. In total, the dataset counts 932 data centers that have been planned and proposed.

At the same time, historically dense markets like California show a more limited pipeline, with just 21 projects planned — a reflection of tightening power availability and permitting constraints in legacy tech corridors.

The geographic spread points to a clear shift: proximity to users is no longer the defining factor. Power is.

That shift becomes even more apparent when looking beyond physical sites to the grid itself.

Data from Interconnect.fyi further shows active interconnection requests surged to 1,687 projects in 2025, representing a record 442 gigawatts of active requested capacity. Just five years earlier, that figure stood at roughly 100 gigawatts.

The growth has been nonlinear. From fewer than 1 gigawatt in 2012, interconnection demand has scaled into the hundreds of gigawatts, with a sharp acceleration beginning around 2021 and intensifying through 2024 and 2025.

Together, the two datasets — one mapping physical infrastructure, the other mapping power demand — reveal the full scope of the AI buildout. What is visible on the ground is only a fraction of what is queued at the grid level.

Meanwhile, the capital behind this expansion is increasingly concentrated among a handful of technology giants.

Companies including Amazon, Google, Microsoft and Meta are not only among the largest spenders on AI infrastructure, but also rank among the most prolific data center operators in the U.S., according to Datacenter.fyi’s breakdown of facility ownership.

Amazon alone accounts for 479 data centers, making it the largest operator by count, followed by CyrusOne (221), QTS (186) and Digital Realty (170). Microsoft (153), Equinix (129) and Google (79) also rank prominently, underscoring how hyperscalers and large colocation providers dominate the physical footprint of compute.

That ownership concentration is being reinforced by an unprecedented surge in capital spending.

In the first quarter alone, Amazon, Google, Microsoft and Meta collectively spent more than $130 billion on capital expenditures, largely directed toward AI data centers and supporting infrastructure. The figure marked another record and represented a sharp increase from a year earlier.

The spending trajectory shows little sign of slowing. Combined outlays from the four companies are expected to approach $700 billion this year, as each races to secure capacity for increasingly compute-intensive AI workloads.

Taken together, the data center map, the interconnection queue, and the capex surge point to the same conclusion: the current buildout is not just broad-based, but highly capital-intensive and led by a small group of firms with the balance sheets to sustain multi-year infrastructure investments.

That dynamic also helps explain the pressure building in interconnection queues. As hyperscalers commit to gigawatt-scale deployments, their demand is translating directly into large-load requests that now dominate grid pipelines across multiple regions.

The pattern feels familiar

Long before AI drove hyperscale demand, Bitcoin miners were already optimizing for the same constraint. Following the China’s Bitcoin mining ban, mining infrastructure rapidly migrated to the U.S., with Texas emerging as a global hub that effectively replaced regions like Sichuan and Xinjiang.

That migration triggered an earlier wave of power and infrastructure development — from substations to large-load interconnections — built around energy-intensive compute. In hindsight, it offered a preview of the dynamics now playing out at a much larger scale.

Industry executives have increasingly drawn that connection. At a recent panel discussion at Bitcoin 2026, featuring MARA and CleanSpark, participants described Bitcoin miners as having “plowed the road” for AI, establishing the operational and infrastructure blueprint that data center developers are now expanding upon.

The implications are becoming clearer as interconnection queues swell.

Grid access — not land or hardware — is emerging as the primary bottleneck. Projects are measured not just by capital invested, but by their position in increasingly crowded queues, where timelines can stretch years.

For the mining sector, the convergence is both validating and competitive. The same regions that once attracted hashpower are now being targeted by AI developers with deeper balance sheets and longer-term contracts, intensifying the race for power.

In that sense, mapping the U.S. data center landscape today is less about counting facilities and more about understanding the underlying energy dynamics.

The physical footprint — nearly 4,000 sites — tells the story of where computing has been. The 442 gigawatts sitting in interconnection queues tells the story of where it’s going.

Regulation News

• Maine Governor Vetoes Data Center Moratorium, Citing $550M AI Project

• China Blocks Meta’s $2 Billion Acquisition of AI Startup Manus

• Keel Wins Zoning Approval for AI Data Center Expansion at Former Bitcoin Mine

Hardware and Infrastructure News

• Central North Carolina communities pause to consider data center development

• Duke Energy Extends Robinson Nuclear Plant to 2050 Amid AI-Driven Power Surge

• Luxor, MicroBT Tie Investment to $100M Miner Deal Amid Weak Bitcoin Hashprice

• Core Scientific to Expand Pecos Campus Capacity to 1.5 GW for AI Data Center Development

• Solaris Energy Signs 600 MW Power Contract, Raises 2026 Guidance

• Galaxy Targets Full 133 MW CoreWeave Delivery at Helios by End of Q2

• Canaan Lands Tether Order for Customized Bitcoin Mining Systems

• MARA to Buy Ohio Gas Power Plant in $1.5 Billion Bet on AI Data Centers

Corporate News

• Brookfield-Backed Csquare Files Confidentially for AI-Fueled US IPO

• Tether Pushes Deeper Into Bitcoin Mining Stack With Open-Source Software Play

• Riot Amends $200M Coinbase Credit Line to Cushion Bitcoin Swings

• Galaxy Hints at Trillion-Dollar Hyperscaler Behind Helios AI Buildout

• Nscale Hires Former Oracle Data Center Lead to Drive AI Buildout

Financial News

• NextEra Energy Reports $6.7 Billion in Q1 2026 Operating Revenues

• Hut 8 Taps $3.25 Billion Debt for Fluidstack-Backed AI Data Center

• Vistra Corp. Completes $4 Billion Private Offering of Senior Notes

• NRG Energy Issues $2.6 Billion in Senior Notes and Secures $900 Million Term Loan

Feature

• For AI and Bitcoin Investors: The Grid’s ‘Secret Sauce’ Just Got Even Sweeter

• ‘We Plow the Road’: Miners Pitch Role in AI Buildout at Bitcoin 2026

• Who Owns the Stack: From Bitcoin to AI, the Race for Power Is Going Off-Grid

Email us

EIF is a high-signal, curated conference convening leaders across energy, compute infrastructure, and capital to map the next decade of power and data center buildout.

Apply to speak or explore sponsorships for private meeting rooms, branding & more.

Core Scientific grabbed headlines this week with its proposed $3.3 billion senior secured notes priced at 7.75%, the latest in a wave of high-yield debt offerings tied to the AI infrastructure boom.

While the financing itself has been widely covered, a lesser-noticed supplemental filing tied to the deal offers something more revealing: a detailed look at the actual infrastructure assets being pledged to back the debt — and, more importantly, the cash flow profile expected to service it.

At the center of the deal is Core Scientific’s long-term hosting relationship with CoreWeave. The document shows that the two parties have already funded roughly 80% of the estimated $5.5 billion in total capex for the projects, underscoring how far along the buildout is.

It also lays out how Core Scientific is packaging the CoreWeave-related data center assets into a financing vehicle with the underlying sites — and their future revenues — forming the backbone of the credit story. In doing so, it provides one of the clearest disclosures yet into how a Bitcoin miner’s AI pivot is being translated into a project-financed, debt-backed business.

A Site-by-Site Look at the Buildout

Based on the document, Core Scientific is developing six data center facilities across five campuses, having secured roughly 900 MW of grid capacity to support about 590 MW of contracted critical IT load exclusively for CoreWeave.

As of mid-March, around 350 MW has been energized, with more than 185 MW already billing — a notable milestone, as it marks the transition from construction to cash-generating operations.

The buildout, however, is not yet complete, and the timeline stretches into the second half of 2027.

The filing breaks down progress by campus, offering rare visibility into how much capacity is coming online and when revenue is expected to fully ramp up.

• Austin, TX: Fully completed in 2024, with 16.5 MW of IT capacity. This site is already generating base license fees and serves as the earliest proof point of the model.

• Denton, TX: The largest project in the portfolio at 262 MW. As of March 2026, about 132 MW — roughly half — is already billable. Full completion is expected in the second half of 2026.

• Marble, NC: A 65 MW campus that is about 57% complete, with 37 MW currently billing. It is expected to reach full completion in the first half of 2026.

• Muskogee, OK: A 70 MW campus where construction began in early 2025, with both delivery and completion expected in the first half of 2026.

• Dalton 1, GA: A smaller 30 MW site scheduled to begin delivery and reach completion in the first half of 2026.

• Dalton 4, GA: A larger 145 MW expansion that will come later in the cycle, with delivery starting in the second half of 2026 and full completion targeted for the first half of 2027.

Taken together, the portfolio shows a staggered ramp, where a meaningful portion of capacity is generating revenue even as the bulk of megawatts are still under construction.

The $10B Headline — and What Actually Drives Revenue

Core Scientific frames the exclusive long-term contract as generating roughly $10 billion in total revenue over the life of its agreements with CoreWeave, net of rent repayments tied to the upfront construction funding contributed by CoreWeave.

Notably, that headline figure includes power passthrough, which materially inflates the top line without contributing to economic margin. Under this structure, CoreWeave reimburses electricity costs paid by Core Scientific, with those payments recorded as both revenue and operating expense on Core Scientific’s balance sheet.

A more grounded way to think about the underlying revenue engine is the base license fee.

Across the portfolio, CoreWeave pays a fixed rate of $100 per kW per month for most sites and $115 per kW per month for the Austin campus, with built-in annual rent escalators of 3.5% (and 3.0% for Austin). Applying those rates to the full 588.5 MW of contracted capacity translates into approximately $59 million per month in base license fees once all sites are fully built and operational, or roughly $700 million annually. This figure is a closer proxy for the net revenue generated by the infrastructure itself, before operating costs and financing effects.

Revenue vs. Levered Free Cash Flow

The illustrative financials included in the filing further underscore that distinction.

For 2026, Core Scientific projects:

• $737 million in total revenue, including power passthrough

• $418 million in operating expenses, also including power passthrough

• That works out $319 million in net operating income

• But just $6 million in levered free cash flow

Stripping the power passthrough from both revenue and operating expenses, the net operating income reflects the actual value of the hosting agreements rather than grossed-up utility flows.

The more meaningful drop occurs below that line.

The sharp compression from net operating income to levered free cash flow is largely driven by what the filing categorizes as “Other Cash Flow Items.” While this includes maintenance capital expenditures, a substantial portion relates to CoreWeave repayment mechanics — specifically, the revenue credits that reduce the cash Core Scientific receives as reimbursement for upfront construction costs. Under the agreements, CoreWeave is entitled to credits equal to 50% of the base license fee, capped at $1.5 million per MW of deployed capacity, effectively allowing it to recoup a certain portion of the buildout costs over time through reduced cash payments.

In effect, what appears as strong contracted revenue is partially being recycled back to the customer in the early years, delaying cash realization. When combined with interest expense on the $3.3 billion debt stack, the result is a business that generates significant accounting income but very limited near-term free cash flow.

Why This Matters for the Bitcoin-to-AI Pivot

This supplemental filing may not have made headlines, but it offers a critical window into how the industry’s AI pivot is being financed.

The takeaway is nuanced. Core Scientific’s deal with CoreWeave delivers long-term, contracted revenue with built-in escalation — a stark contrast to the volatility of Bitcoin mining. But it also introduces a capital-heavy structure where cash flow is constrained in the early years and heavily shaped by financing terms.

In many ways, this mirrors what has already been seen on the other side of the relationship. Just weeks ago, CoreWeave was tapping debt markets with GPU-backed financing structures, effectively using contracted compute demand as collateral to raise capital. Here, Core Scientific is doing something similar — but with powered land and data center infrastructure instead of chips.

Both cases point to the same underlying shift: AI infrastructure is increasingly being built not just with equity and operating cash flow, but through structured, asset-backed financing models where future contracted revenue is pulled forward to fund present-day capex.

That makes this more than just a Bitcoin miner diversifying into AI. It is another example of how the entire stack — from GPUs to data centers — is converging toward a project finance model, where assets are ring-fenced, cash flows are pre-allocated, and returns are shaped as much by capital structure as by demand.

Hardware and Infrastructure News

• Soluna Eyes 98 MW Bitcoin-to-AI Conversion With $16.5M First-Step Dorothy Buyout

• Data centre delays threaten to choke AI expansion

• Alcoa Nears Sale of Idled Smelter to NYDIG as Bitcoin and AI Chase Industrial Power

• Microsoft’s $3.3B Fairwater AI Data Center Goes Live Ahead of Schedule in Wisconsin

• Soluna Expands Blockware Deal With 3.3 MW at Texas Wind-Powered Bitcoin, AI Data Center

• Keel Infrastructure Finalizes $13 Million Sale of Paraguay Site, Exits Latin America

• American Bitcoin completes installation of over 11,000 miners at Drumheller facility

• Applied Digital Secures $7.5 Billion Lease for 300 MW AI Campus

Corporate News

• Fermi Shares Sink 20% as CEO Exit Clouds AI Data Center Ambitions

• TransAlta Appoints Mike Politeski as CFO and Grant Arnold as CCO

• Amazon to invest up to another $25 billion in Anthropic as part of AI infrastructure deal

• Cango Appoints New CFO and Director Amid Strategic Shift

• Nscale, BT Plan UK Sovereign AI Data Centers With NVIDIA Stack

Financial News

• Tether Takes 8.2% Stake in Bitmain-Tied Bitcoin Mining Lender Antalpha

• Core Scientific Seeks $3.3B Bond as Bitcoin Miner Rides AI Data Center Debt Boom

• Soluna Saw 10X Volume Spike After Nasdaq Warning as Bitcoin-to-AI Pivot Gains Focus

• HIVE Announces Closing of Private Offering of US$115 Million of 0% Exchangeable Senior Notes

• Ex-Credit Suisse Team Plans $1 Billion Fund for Data Center Risk

Features

• New Gas-Powered Data Centers Could Emit More Greenhouse Gases Than Entire Nations - Wired

Email us

Today’s featured EIF speaker: Matthew Roszak, chairman and co-founder of Bloq and Hemi, and founding partner of Tally Capital, a globally recognized investor who has backed category-defining companies across the bitcoin and digital asset ecosystem, including Binance, Coinbase, Blockstream, and BitGo.
EIF is a high-signal, curated conference convening leaders across energy, compute infrastructure, and capital to map the next decade of power and data center buildout.
Register to attend today or explore sponsorships for private meeting rooms, branding & more.

Public bitcoin miners have liquidated their BTC reserves at a pace not seen since the depths of the last crypto bear market, as a prolonged slump in mining economics pushes operators into survival mode.

Several major public miners, including MARA, CleanSpark, Riot, Cango, Core Scientific, and Bitdeer, have already sold more than 32,000 BTC in the first quarter of 2026, according to data analyzed by TheEnergyMag. The dataset remains incomplete, as first-quarter earnings reports are still pending.

Even so, the figure already exceeds total net sales across all four quarters of 2025 and sets a new industry record, surpassing the roughly 20,000 BTC that public miners liquidated in the second quarter of 2022 during the market turmoil triggered by the Terra-Luna collapse.

The reversal is striking. Just over a year ago, miners were accumulating aggressively, ending 2024 with a net addition of 17,593 BTC and pushing combined reserves above 100,000 BTC.

The shift comes as hashprice — a key industry metric measuring expected mining revenue per unit of computing power — hovers in the low $30/PH/s range, near all-time lows. At those levels, margins are either compressed or outright negative, especially for operators running older, less efficient fleets or paying higher power costs.

The root of today’s pressure can be traced to the industry’s aggressive hashrate expansion following China’s mining ban in 2021 — a period that, in hindsight, fueled exponential growth at each company’s own expense.

For context, bitcoin’s current price — despite retreating from its all-time high above $120,000 — remains higher than the previous cycle peak. Yet network difficulty is now roughly 10 times higher than in 2021, and block rewards were cut in half in 2024. In effect, mining profitability has compressed by an order of magnitude, helping explain the wave of recent selling.

But the record liquidation does not tell a uniform story. Instead, it reveals an industry beginning to diverge — with some operators forced to sell into weakness, while others lean on structural advantages or capital discipline to ride out the downturn.

For many, the immediate priority is liquidity. Selling bitcoin remains the fastest way to shore up balance sheets, fund operations, and meet debt obligations in a financing environment that is both selective and expensive.

Others are taking a more measured approach. American Bitcoin (ABTC), the proprietary mining carve-out of Hut 8, has doubled down on accumulating bitcoin through both mining and market purchases. As of early April, it had built reserves of more than 7,000 BTC, up from zero a year earlier, while ramping its proprietary hashrate to 28 EH/s.

But the company is not repeating the hashrate-at-all-costs playbook of the previous cycle. Matt Prusak, president and interim CFO of ABTC, told TheEnergyMag that its focus is on quality growth under current market conditions.

“We won’t do deals that we don’t think will win. … When you see people chasing hashrate and chasing big exahash numbers, that’s never been our style,” Prusak said. “Having the biggest fleet doesn’t make a difference to me.”

Unlike many publicly traded peers, ABTC expanded its hardware fleet when demand for ASICs had already cooled. In the summer of 2025, it acquired roughly 15 EH/s of Antminer S21 series from Bitmain by pledging about 3,000 BTC — redeemable within 24 months — rather than paying cash. Such a structure would have been unlikely in prior market conditions. The value of the pledged bitcoin has since declined by about 40%, and Bitmain cannot liquidate the collateral unless ABTC chooses not to redeem it.

Based on Q4 2025 data analyzed by TheEnergyMag, ABTC’s all-in cash cost of production was around $55,000 per bitcoin, or roughly $25/PH/s — among the lowest in the public mining cohort. That allows it to accumulate newly mined bitcoin at a discount to prevailing market prices.

Even if bitcoin falls below that level, Prusak said the company retains the flexibility to allocate capital dynamically. ABTC raised $240 million through at-the-market offerings in 2025 and another $110 million in the first quarter of this year.

“We don’t have to flip into AI. We are a bitcoin allocator. If bitcoin’s expensive relative to the cost to mine, we mine. If bitcoin’s cheap relative to mining, then we buy.” he said, adding: “ At this time, we have no intent ot sell. … We are accumulating.”

But for private operators without comparable access to capital, the divergence in strategy is increasingly shaped by one of the industry’s oldest variables: power costs.

Sean McDonough, president and CEO of New West Data, a Canadian oil producer that mines bitcoin using off-grid power generated by flared natural gas from its own oil sites, said the company’s effective power cost is below $0.02 per kilowatt-hour. That is, in some cases, roughly one-third of what large-scale public miners pay.

At that level, even less efficient machines remain profitable. With hashprice around $30/PH/s, a miner paying $0.02/kWh can sustain fleet efficiencies of roughly 60 J/TH. McDonough said this enables the company to acquire older-generation equipment at lower upfront costs while maintaining margins, especially as ASIC prices have fallen alongside hashprice.

That cost advantage has allowed New West Data to expand despite the downturn. The company tripled both its oil production and bitcoin compute capacity in 2025 and expects to triple again this year. It currently operates about 15 MW of computing capacity, all powered by flared gas from its own sites.

Still, flared gas represents a niche model, requiring expertise in upstream oil production rather than traditional power procurement through utilities or long-term power purchase agreements.

Absent ultra-low-cost power, miner operators are also turning to operational optimization to preserve margins.

Luxor, a bitcoin mining pool operator, ASIC broker, and software provider, launched a fleet management tool called Commander earlier this month. The platform uses automated algorithms to evaluate hashrate and power market conditions every five minutes, dynamically adjusting power settings across a fleet based on real-time economics.

The goal is to optimize output from existing infrastructure. Luxor says internal benchmarks show an 8% to 14% improvement in profitability compared with traditional on/off curtailment strategies.

The shift toward software reflects a broader recalibration across the industry. With hashprice under pressure, upgrading to the latest generation of machines often requires capital outlays that are difficult to justify on a standalone return basis.

Instead, operators are focusing on extracting better margins from existing fleets — gaining incremental efficiency wherever possible.

Ethan Vera, Luxor’s chief operating officer, said the Commander platform has already scaled to about 5 EH/s of customer hashrate since launch. It complements LuxorOS, the company’s firmware solution introduced in 2022, which now supports roughly 45 EH/s, or about 5% of the global network.

In one recent case study, Luxor claimed that Soluna, a publicly traded bitcoin miner with colocation and proprietary mining in Texas, was able to speed up the recovery time via LuxorOS for its 1.1 EH/s fleet by 50% after curtailment events, improving uptime without additional operational expenditure.

All told, the industry is no longer moving in lockstep. What was once a relatively uniform business model defined by scale and hashrate growth is fragmenting into a range of survival and quality growth strategies shaped by power economics, balance sheet flexibility, and operational sophistication.

Regulation News

• Missouri town fires half its city council over data center deal

Hardware and Infrastructure News

• Cango Launches Commercial AI Inference Operations at 50-MW Georgia Facility

• Microsoft Secures Former OpenAI “Stargate” Site in Norway for AI Infrastructure

• Microsoft Secures 3,200 Acres for Massive Cheyenne Data Center Expansion

• Bitdeer Ramps to 70 EH/s as ‘Seasonal Factors’ Hit Bitcoin Mining Uptime

Corporate News

• CoreWeave Secures $21 Billion Meta Contract and Targets $4.7 Billion in New Debt Financing

• CoreWeave Lands Anthropic Deal, Upsizes Convertible Notes to $3.5B Amid AI Boom

• Riot Data Center Chief Departs Less Than a Year After Bitcoin Miner’s AI Push

• CoreWeave Lands $6B AI Deal With Jane Street, Secures $1B Equity Investment

Financial News

• NRG Energy Announces Multi-Tranche Debt Offerings and $900 Million Term Loan

• Jane Street in Talks to Back Fluidstack at $18 Billion Valuation

• TeraWulf raises $900 million through stock sale for Kentucky infrastructure and debt settlement

• HIVE secures $75 million through private bond issuance and seeks to list on TSX

Email us

Today’s featured EIF speaker: Congressman Earl “Buddy” Carter (R-GA), a member of the House Energy and Commerce Committee and the House Budget Committee, where energy reliability, infrastructure investment, and technology policy collide.
EIF is a high-signal, curated conference convening leaders across energy, compute infrastructure, and capital to map the next decade of power and data center buildout.
Register to attend today or explore sponsorships.

If 2021 was the year of ASIC-backed loans, 2026 may go down as the moment compute financing grew up.

CoreWeave’s $8.5 billion delayed draw term loan—secured primarily by GPU infrastructure—has already made headlines as the first and largest deal of its kind. The headline numbers alone are notable: the facility allows CoreWeave to draw capital through June 2027, with final maturity in March 2032. It is split between floating-rate debt priced at roughly SOFR + 2.25% and fixed-rate tranches around the high-5% range, putting the overall cost of capital in the mid-single digits—remarkably low for an asset class that didn’t meaningfully exist a few years ago.

But buried inside the credit agreement is something far more intricate: a blueprint for how Wall Street now wants to finance compute. This isn’t just “GPU-backed debt.” It’s compute treated as infrastructure.

And in many ways, it’s what MinerFi tried—and failed—to become.

The lesson from 2021: hardware alone isn’t enough

Back in 2021, bitcoin miners tapped a wave of ASIC-backed loans to expand aggressively. The premise was simple: machines generate bitcoin, bitcoin repays debt.

What that model underestimated was how volatile and fragile both sides of the equation were.

When bitcoin’s price fell in 2022, the hashrate went up relentlessly as a result of the MinerFi boom, deteriorating the hashprice even further. At the same time, the resale value of ASIC machines declined. So lenders were left with collateral that was worth less and generating less cash.

That double exposure—asset value risk and cash flow risk—was fatal.

The real collateral isn’t GPUs

CoreWeave’s credit agreement emphasizes the Data Center (DC) Funding Conditions, which are really the spine of the whole deal.

Simply put, CoreWeave cannot simply point to a purchase order for GPUs and borrow against it. The financing is tied to whether those machines are actually becoming usable compute inside a functioning data center, for a customer contract with the customer’s acceptance and satisfaction.

That matters even more because the advance rate is high.

The agreement defines the borrowing base for a funding date as the “Funding Date GPU Amount,” which equals 90% of Funding Date Capital Expenditures, plus certain fees and expenses tied to the transaction. Those capital expenditures include the cost of the infrastructure itself and installation costs.

In practical terms, lenders are saying they are willing to fund almost the entire build cost of the GPU deployment. But they only do that because the deal does not treat a GPU as collateral the moment it is bought.

Instead, the agreement requires the financed infrastructure to satisfy the DC Funding Conditions. Those conditions tie funding to real-world deployment: the equipment has to be physically associated with the relevant data center site, prior loans must already have been used properly, and for the applicable infrastructure, the DC Funding Conditions must be satisfied at the time of the borrowing. The contract also includes language elsewhere that accepted GPU clusters count toward Funding Date Capital Expenditures only when the customer has accepted them in writing and the borrower certifies testing has been completed for the relevant data halls.

So the key idea is this:

CoreWeave is getting very high leverage — roughly 90 cents of debt for every dollar of qualifying deployment cost — but only after proving that the machines are not just purchased, but installed, tested, accepted, and on the path to generating contracted revenue.

That is a notable difference from the MinerFi era.

In 2021, ASIC-backed lenders often financed miners largely on the assumption that once the machines were delivered, the economics would take care of themselves. Here, the lenders are willing to go bigger precisely because they are financing not a box of chips, but a controlled, trackable, revenue-linked compute system. That is what makes this feel less like equipment lending and more like infrastructure finance.

Following the machines down to the rack level

Once you start from that premise—only funding “ready-to-earn” assets—the rest of the structure builds around visibility and control.

Every GPU financed under the facility is tracked:

• It must be located at an approved site

• It must be tied to a specific deployment

• Lenders must have contractual access to the facility

This is very different from early ASIC lending, where collateral could be loosely defined and, in some cases, difficult to locate or control in distress scenarios.

Here, lenders know not just what they are financing, but where it is and how it is being used.

Cash doesn’t flow freely—it follows a strict order

Once those GPU clusters start generating revenue, the cash doesn’t simply go to CoreWeave.

It flows through a controlled system—known as a cash waterfall.

Here’s the idea:

Every dollar that comes in is a pre-assigned job.

1. First, pay the costs to keep the system running (power, operations)

2. Then, pay lenders (interest and principal)

3. Only after that can any remaining cash go back to the company

Why it matters: it ensures that debt repayment is prioritized automatically, rather than relying on management discretion.

In MinerFi, that level of control was often opaque.

And the agreement goes one step further with what’s called a cash trap.

If the project starts underperforming—for example, if it’s not generating enough cushion above its debt obligations—then excess cash stops flowing back to the company.

Instead, it stays within the system to prevent situations where value leaks out of the structure just as risks are rising.

Even the electric bill is part of the credit story

One of the more telling details is how seriously the agreement treats power costs.

Running GPUs at scale is energy-intensive, just like running bitcoin miners, and electricity is one of the largest—and most volatile—expenses.

The structure requires mechanisms to manage that risk, including hedging or other arrangements to stabilize power pricing.

In simple terms: the loan isn’t just underwriting compute—it’s underwriting the cost of the electricity that makes that compute possible. And that variable is actively controlled rather than left exposed.

Borrowing only happens when the math works

All of this feeds into a final layer of discipline: the project must prove it can pay for itself.

That’s where the debt service coverage ratio (DSCR) comes in. DSCR compares how much cash a project brings in versus how much it needs to pay lenders (principal+interest).

• If a project generates $120 and owes $100 in debt payments, its DSCR is 1.2x

• That extra $20 is the cushion—it allows for things to go slightly wrong without immediately causing stress

In CoreWeave’s case:

• To take on new debt, the projected DSCR needs to be around 1.20x

• To stay in compliance, it needs to maintain roughly 1.15x

That buffer is what gives lenders confidence that even if things don’t go perfectly, the system can still hold.

The bigger shift: from machines to systems

Put all of this together, and the contrast with 2021 becomes clear.

MinerFi was built around machines: Finance the hardware and expect the revenue to follow.

This deal is built around systems:

• Verify the infrastructure is live

• Confirm the cash flow is real with customer contracts

• Control how that cash is used

It’s a fundamentally different way of risk management.

At $8.5 billion, backed by major banks and structured to investment-grade standards, this isn’t just a large deal. It’s a signal.

Compute—specifically AI compute—is being redefined as infrastructure that can be financed with the same discipline as power plants or toll roads.

Bitcoin mining showed that energy can be turned into digital output at scale.

CoreWeave’s financing shows how that output can be turned into something lenders are willing to underwrite at massive scale.

Of course, structure can’t eliminate risk. Both MinerFi and this new wave of compute financing ultimately depend on demand holding up.

The difference is where the risk sits.

• In mining, it was concentrated in network and market-driven revenue (hashprice)

• Here, it’s embedded in contract durability and utilization

Whether that shift makes the system more resilient—or simply changes how stress emerges—remains to be seen.

But one thing is already clear: Compute financing has entered a new phase—and this time, it looks a lot more like Wall Street than crypto.

Hardware and Infrastructure News

• Soluna Buys Texas 150MW Wind Farm for $53M to Power AI, Bitcoin Data Centers

• Bitdeer Unveils 9.45 J/TH Miner as Bitcoin Hashprice Remain Under Pressure

Corporate News

• DeepSeek’s V4 model will run on Huawei chips, The Information reports

• CleanSpark Shutters 25MW Tennessee Bitcoin Mine Tied to GRIID Noise Dispute

Financial News

• Riot Sells 3,778 Bitcoin in Q1, Posts More Collateral as BTC Slides

• Cango Sells 2,000 Bitcoin in March After 4,616 BTC February Exit to Cut Debt

• CoreWeave Signs $21B Meta AI Cloud Deal as It Seeks Another $4B Debt Financing

Email us

Introducing EIF: A new conference in Dallas (July 23-24, 2026) for executives, investors, and operators across mining, AI/HPC, power, and infrastructure. Full details in today's press release.

The AI infrastructure boom has no shortage of capital.

It has a shortage of conversion.

Across the industry, billions of dollars are being raised, allocated, and announced at a pace that would have been unthinkable just a few years ago. But turning that capital into functioning data centers, energized megawatts, and ultimately revenue-generating compute is proving far more complex — and far slower — than the headlines suggest.

That gap is beginning to define the next phase of the AI supercycle.

Capital is flooding in

By TheEnergyMag’s analysis, nine publicly listed bitcoin mining data center operators with HPC/AI initiatives collectively raised $17 billion in net financing inflows in 2025, up from $10.9 billion in 2024. Over the same period, their net spending on property, plant, and equipment reached $6.3 billion, compared with $4.5 billion a year earlier.

The direction is clear: more capital is entering the system, and more is being deployed.

But the gap between capital raised and capital converted into productive infrastructure remains wide.

Zooming out, the same pattern is now playing out across the broader AI ecosystem.

Major technology companies, including Microsoft, Amazon, Alphabet and Meta Platforms, are expected to spend as much as $635 billion on AI infrastructure in 2026, up sharply from $383 billion in 2025.

The capital is there.

The question is how quickly — and how efficiently — it can be turned into revenue.

The conversion problem

At NVIDIA’s GTC 2026, the tone from infrastructure investors and operators was less about fundraising and more about execution. The Luxor team summarized their key takeaways from attending the event in a recent newsletter post.

One point stood out: money is no longer the scarcest input.

Instead, the constraints are increasingly physical and operational.

A single gigawatt-scale AI campus can require more than 9,000 workers across dozens of states. Equipment supply chains — from transformers to cooling systems — remain tight. Interconnection queues are clogged with projects that may never materialize. And rapidly evolving compute architectures are forcing redesigns even after construction has begun.

In other words, the bottlenecks are no longer just about securing capital or even power in isolation.

They sit at the intersection of:

• labor availability

• grid access and transmission

• equipment manufacturing

• site design and cooling architecture

• tenant readiness and utilization

Each layer introduces friction between capex committed and cash flow realized.

The grid is no longer a passive input

Nowhere is this friction more visible than in power markets.

In the Electric Reliability Council of Texas (ERCOT), as of February 2026, the Large Load Interconnection queue contained approximately 400 individual requests totaling 239,000 MW. This volume is roughly 2.8 times the current ERCOT record for system peak demand.

The headline number suggests overwhelming demand.

But it also highlights a deeper issue: not all megawatts are equally real.

In PJM Interconnection, regulators and grid operators are already moving to separate speculative load from bankable demand. PJM’s latest long-term forecast places greater emphasis on “firm” commitments — projects backed by actual construction progress or binding service obligations — rather than early-stage requests.

Meanwhile, the Federal Energy Regulatory Commission has stepped in to push for clearer rules governing co-located large loads such as data centers, reflecting mounting concern over how these facilities interact with existing grid infrastructure.

The shift is subtle but important.

The market is moving from who can secure power on paper to who can actually deliver power in practice.

From gigawatts to monetization

Even once a site is financed and energized, the conversion process is not complete.

Revenue depends on:

• securing long-term customers

• achieving high utilization rates

• aligning infrastructure with evolving compute requirements

• maintaining uptime and service guarantees

This is particularly relevant for bitcoin miners pivoting into HPC and AI colocation.

Their advantage lies in power sourcing, site development, and operational expertise. But participating in the AI infrastructure stack requires adapting to a different economic model — one that is contract-driven, utilization-sensitive, and increasingly dependent on creditworthy counterparties.

The result is a growing divergence between announced capacity and monetized capacity

The deeper story is that the AI boom is no longer confined to software or semiconductors.

It is now fully entangled with the physical economy.

Grid operators are rewriting interconnection frameworks. Utilities are reassessing cost allocation. Local communities are pushing back against large-scale developments. Labor markets are tightening around specialized construction and engineering roles.

Taken together, these dynamics point to a broader conclusion: The AI infrastructure cycle has shifted from capital scarcity to execution complexity.

Raising billions is no longer the defining challenge. Aligning capital, power, infrastructure, labor, and demand into a functioning, revenue-generating system is.

That alignment does not happen automatically. It requires coordination across stakeholders that have historically operated in silos — investors, developers, utilities, grid operators, equipment suppliers, and end users.

Where mining, AI/HPC, energy, and capital markets converge

As the gap between capital raised and revenue realized becomes more visible, the industry is entering a phase where execution, not ambition, will separate winners from the rest.

That is why we are launching the Energy Investor Forum (EIF).

EIF will bring together the companies and investors shaping the next phase of AI and power-intensive infrastructure, not just for on-stage discussion but for curated meetings and real business development across power, sites, interconnection, financing, customer relationships, and the partnerships needed to convert announced capacity into operating, monetized infrastructure. To get involved, readers can explore sponsorship opportunities, apply to speak, or request an investor attendee pass.

Because the next phase of the AI boom will not be defined by who can raise the most money. It will be defined by who can put it to work.

Regulation News

• Chip Security Bill Advances in House After Super Micro Case

• Proposed GOP Bill Targets Foreign Crypto Mining Hardware

Hardware and Infrastructure News

• Meta boosts investment in West Texas AI data center by over sixfold to $10 billion

• TSMC’s 3nm Chip Capacity Has Become So Constrained That Only “Long-Term, Loyal” Customers Are Getting Priority

• Bitdeer Contracts DCI for 180 MW AI Data Center Development in Norway

• Nebius Plans 310 MW AI Data Center in Finland

• Applied Digital Modifies Ellendale Data Center Leases Following CoreWeave Credit Upgrade

Corporate News

• Bitfarms Nears Keel Rebrand, Plans to Sell $161M in Bitcoin to Fund AI Push

• Duke Energy concludes $2.48 billion divestment of Tennessee gas assets to Spire

• Cango Faces NYSE Delisting Risk Over Low Share Price

• Luxor Launches Commander, a Fleet Management and Profitability Optimization Software for Bitcoin Mining Operations

Financial News

• Google Backs $5B Texas Data Center for Anthropic, Deepening AI Infrastructure Push

• Mistral AI Secures $830 Million Debt Deal for France AI Data Center

• CoreWeave Lands $8.5 Billion GPU-Backed Financing in Latest AI Deal

• NVIDIA Invests $2 Billion in Marvell to Expand AI Infrastructure Ecosystem

• OpenAI Closes $122B Raise at $852B Value as AI Capex Surge Accelerates

Feature

• Microsoft is picking up a Texas data center project OpenAI didn’t want, in a telling sign of how far they’ve drifted apart

• The Same Engine Behind Bitcoin and AI - TheEnergyMag

Email us

Across rural America, the AI infrastructure boom is increasingly taking shape behind closed doors.

For much of the past year, some of the biggest names in AI infrastructure have been hunting for land in rural America the same way they hunt for power: quietly, quickly and often before the public knows what is happening.

The playbook has become increasingly familiar. Developers approach local officials under nondisclosure agreements (NDAs), use intermediaries or shell entities to mask the ultimate buyer, and begin assembling hundreds of acres before residents can organize around what a project might mean for farmland, water use, noise and electricity demand. That secrecy may help companies lock up strategic sites in a fiercely competitive market for AI infrastructure, but it is also becoming a political liability.

Now the backlash is getting harder to ignore.

In Wisconsin, at least five communities have now been linked to NDAs tied to proposed data center developments, according to Wisconsin Watch, which found secrecy arrangements in Beaver Dam, Kenosha, Janesville, Menomonie and most recently Beloit. In Beaver Dam, Meta used shell companies in the early stages of a 520-acre, roughly $1 billion campus. Across the state, seven major data center projects are pending with a combined value of more than $57 billion, making Wisconsin one of the clearest examples of how the AI buildout is colliding with demands for public disclosure.

The pushback has become forceful enough that Microsoft — one of the most aggressive builders of AI data center capacity globally — said this month it would stop using NDAs with local governments for data center development. The company said the old practice had been meant to protect commercially sensitive information, but that transparency with communities had become "paramount." Microsoft added that it is working to identify and terminate active NDAs with local governments, even as it reserved the right to keep using confidentiality protections for certain trade secrets and land acquisitions.

That policy shift says as much about politics as it does about process. Microsoft President Brad Smith said this week that winning over local communities has become essential as towns across the U.S. increasingly protest data center projects. Reuters reported that opposition in parts of the Midwest and Northeast has already contributed to canceled developments over concerns ranging from power prices and water usage to pollution from related energy infrastructure.

What is changing is not just the scale of the projects, but the scale of what communities feel they are being asked to absorb before anyone has had an honest public debate.

In Mason County, Kentucky, for instance, officials have backed a proposed hyperscale campus that would span about 2,080 acres of agricultural land outside Maysville. The plan envisions six data center buildings, multiple substations, stormwater ponds and phased construction beginning in summer 2027 if approvals move forward. County officials have pitched the project as an economic engine that could bring 400 full-time jobs and thousands of construction jobs, while also keeping the identity of the company behind it undisclosed unless the development is ultimately approved.

That secrecy has helped turn the project into a local flashpoint. Realtor.com and local broadcaster WKRC reported that several farming families have rejected lucrative offers to sell. One mother and daughter turned down more than $26 million for part of their land, while another family previously refused nearly $8 million. Officials say at least 20 residents were approached and 18 signed conditional sale agreements, but the holdouts have become symbols of a broader resistance: not necessarily to data centers as such, but to the idea that farmland can be quietly assembled for an unnamed AI customer and only debated in earnest after the map is largely drawn.

A similar drama is playing out in Pennsylvania, where 86-year-old farmer Mervin Raudabaugh rejected more than $15 million from a data center developer for his 261 acres in Cumberland County. Instead, he sold the development rights for about $2 million to preserve the land for agriculture. Realtor.com reported that the developer’s outreach had been persistent, while local preservation advocates said deep-pocketed data center buyers are increasingly targeting large tracts of open farmland that conservation groups cannot match on price.

Taken together, the cases point to a deeper tension in the AI infrastructure boom. For hyperscalers and their development partners, secrecy is often treated as a practical necessity. Land is scarce, power is scarcer, and once word gets out, prices can jump, competitors can swoop in and local opposition can organize before a project is mature enough to present in full. Microsoft said as much in explaining why NDAs became common in the first place.

But the communities being targeted are increasingly viewing that same secrecy as a warning sign.

In Wisconsin, backlash over confidential negotiations has fueled legislative efforts to restrict local governments from signing NDAs with data center developers. Critics say the issue is bigger than whether companies are entitled to protect trade secrets. It is about whether elected officials should be negotiating projects that can reshape land use, tax policy, infrastructure planning and local utility systems without telling the people who will live next to them.

That tension is likely to intensify as the AI buildout moves beyond established data center corridors into rural counties with abundant acreage and relatively easier paths to large-scale development. The industry still has money, urgency and political support on its side. But the farmers turning down eight-figure offers in Kentucky and Pennsylvania are a reminder that not every acre is just another site-control exercise.

In the AI race, land is becoming as strategic as chips and megawatts. The more quietly developers try to secure it, the more likely rural communities are to ask what else they are not being told.

Regulation News

• Pennsylvania House Passes AI Data Center Bill, Setting Up Grid Cost Showdown

Hardware and Infrastructure News

• Illinois City Approves $20B AI Data Center After Marathon Public Hearing

• SoftBank, DOE Back 10GW Ohio AI Data Center and Power Buildout

• Arm Enters Merchant Silicon Market with AI-Focused “AGI CPU”

• Cipher Digital Lands Third AI Data Center Lease, Secures $200M Credit Facility

• Iowa Bitcoin Miner Pushes Expansion as Cedar Falls Weighs Gas Plant

• Bell and HIVE Launch Merritt AI Facility, Formalizing 16.6 MW Canadian Expansio

Corporate News

• Bitfarms Shareholders Approve US Relocation and Rebranding to Keel Infrastructure

• BitFuFu Sues Over Alleged Bitcoin Mining Hosting Fraud in Mississippi

• NVIDIA and Emerald AI collaborate with U.S. energy firms to develop adaptable AI infrastructure for grid management

• Hut 8 leans into modular ‘LEGO block’ model to switch between AI and bitcoin mining

• Auradine Rebrands as Velaura AI, Shifts Teraflux Inventory to In-House Bitcoin Mining

• Crusoe and Form Energy Ink 12 GWh Iron-Air Battery Deal for AI Infrastructure

Financial News

• NextEra Energy Issues $600M in Junior Subordinated Debentures

• Core Scientific increases credit line to $1 billion with J.P. Morgan commitment

• US to pay total $1bn to switch from wind to oil and gas development

• WhiteFiber Reports 61% Revenue Growth in Q4 2025 and Secures $865 Million Nscale Contract

• MARA Sells $1.1 Billion in Bitcoin to Retire $1 Billion in Debt

Email us

The most important power market for U.S. data centers is rewriting the rules.

PJM Interconnection — the grid operator spanning from Illinois to Virginia — sits at the epicenter of the global AI infrastructure buildout. The region hosts Northern Virginia, the world’s largest data center hub, and already accounts for roughly 40% of U.S. data center capacity.

Late last month, PJM filed a proposal with federal regulators to overhaul how large electricity users — particularly data centers colocated with power generation — connect to and pay for the grid. The filing was made in response to a December directive from the Federal Energy Regulatory Commission (FERC), which concluded that PJM’s existing framework no longer reflects the realities of large-scale AI-driven demand.

The timing is not accidental. The proposal comes as the White House recently emphasized a “Ratepayer Protection” stance, signaling that households and traditional businesses should not bear the cost of infrastructure built to serve energy-intensive AI and data center growth. That message closely aligns with regulators’ concerns that some large users are benefiting from the grid without paying a proportional share.

The initial public comment period on that proposal closed on March 16, drawing formal objections from industrial power users and setting the stage for a regulatory decision in the months ahead.

PJM’s filing is more than a procedural update. It represents a structural shift in how the grid treats the fastest-growing source of electricity demand in the U.S.

At stake is not just a tariff adjustment, but the economics of where — and how — the next generation of AI compute gets built.

To understand why this matters, it helps to simplify how things used to work.

From “Self-Powered” to Grid-Dependent

For years, PJM allowed something called “behind-the-meter” generation.

In simple terms, this means a company produces electricity right next to where it uses it — for example, a factory with its own gas turbine, or more recently, a data center built next to a power plant.

Under the old rules, these users could “net” their electricity usage. Think of it like this:

• If a data center uses 100 MW of power

• And produces 70 MW on-site

• The grid only counts it as using 30 MW

That accounting matters because many of the biggest grid charges are based on how much demand you place on the system.

In PJM, large customers typically pay for several things tied to their load:

• Transmission costs (paying for high-voltage lines that deliver power across the region)

• Capacity costs (paying for the grid to have enough generation available during peak demand)

• Certain system and reliability charges

If your load is counted as 30 instead of 100, those charges are calculated on the smaller number.

That made sense when onsite generation truly reduced dependence on the system.

But the problem is that large facilities — especially data centers — still rely heavily on the grid in ways that aren’t captured by that simple math:

• They need backup power when onsite generation fails

• They rely on the grid for stability and frequency control

• The grid must still be sized to serve them at full demand

In other words, even if they “self-generate” most of their power, the grid still has to be there — fully built — just in case.

That’s where regulators see a mismatch.

The 50 MW Line

PJM’s proposal introduces a clear dividing line:

• Below 50 MW → treated like traditional, smaller onsite users

• Above 50 MW → treated as large grid users, regardless of onsite generation

In practical terms, this means large data centers can no longer use accounting to significantly reduce their grid costs.

Instead, they must choose:

• Pay for firm access to the grid (guaranteed power, higher cost)

• Or take non-firm service (cheaper, but can be curtailed)

The key shift is conceptual: You are no longer “off-grid” just because you have power next door.

Why This Is Happening Now

The rule shift only makes sense when you look at what PJM is seeing internally.

The operator is facing a surge in demand unlike anything in its history — driven almost entirely by data centers.

• PJM estimates up to ~30 GW of new data center demand by 2030

• It already represents ~40% of U.S. data center capacity

Even more striking:

• Total proposed large-load requests can exceed 100 GW before filtering

To put that in perspective, that’s comparable to adding multiple large states’ worth of electricity demand onto the grid.

PJM has had to heavily discount speculative projects and apply conservative assumptions just to make the numbers usable.

Still, the conclusion is unavoidable: this is not incremental growth — it’s a structural demand shock.

The Hidden Constraint: The Grid Can’t Keep Up

Another key issue is timing.

• Data centers can be built in 1–2 years

• Grid infrastructure often takes 3–7 years

PJM’s proposed new service options reflect this mismatch. They effectively say:

• You can start operating early

• But you may be interrupted until the grid is fully upgraded

This introduces a tradeoff the industry hasn’t had to face at scale before: speed versus certainty.

The Pushback: Not From Big Tech (Yet)

Interestingly, though, the strongest opposition so far appears to be coming from industrial users, not hyperscalers just yet.

Manufacturers and combined heat & power operators argue:

• The 50 MW threshold is too low

• The rules are being rewritten for data centers

• But applied to everyone

Their concern is that longstanding onsite generation models — which genuinely reduce grid demand — could be penalized alongside large AI facilities.

Why This Matters for Bitcoin Miners

At its core, this is about cost allocation.

FERC’s position is increasingly clear:

If the grid must be built to serve you, you should pay for it — even if you don’t use it all the time.

That challenges a key strategy used by data centers: colocating with power to minimize grid costs while still relying on it as a safety net.

For Bitcoin miners, this dynamic is familiar because mining has long operated on a different model:

• Flexible demand

• Willingness to curtail

• Optimization around power pricing

In many ways, PJM’s framework formalizes that distinction:

• Firm power → expensive but reliable

• Flexible power → cheaper but interruptible

The difference is that AI workloads are far less tolerant of disruption. You can shut down ASIC miners any time – bitcoin doesn’t care. But you can’t turn off AI compute hardware without jeopardizing long-term service revenues from customers.

The Bottom Line

PJM is not just tweaking rules — it is redefining what it means to be a large power user and implementing a regulator-driven reset of how large loads interact with the grid.

The old idea was:

“If I generate my own power, I pay less for the grid.”

The new reality is:

“If the grid needs to exist for you, you pay for it.”

The first round of public comments has already been filed, and FERC is now deciding whether PJM’s proposal meets the mandate it laid out in December. Even if approved, the rule would be phased in over several years, with legacy users getting some protection and newer large-load projects facing a much stricter framework almost immediately.

That means the fight is no longer about whether the old model survives. It is now about how quickly the new one takes hold — and how expensive it becomes for the next wave of AI infrastructure to secure reliable power.

Hardware and Infrastructure News

• Nscale in Talks to Acquire 8GW West Virginia AI Data Center Site: Report

• DOE Targets AI-Driven Power Demand With $1.9B Grid Upgrade Initiative

• Bitdeer Launches Scrypt Miner as Bitcoin Mining Hardware Market Tightens

• DOE Issues Second Emergency Order to Keep Centralia Coal Plant Online

• HIVE Launches GPU Cloud Cluster in Paraguay Supporting Columbia University AI Research

• Bitcoin Difficulty Set for ~8% Drop, Exceeding 2022 Capitulation

• Fluidstack Abandons $11B France AI Data Center Projects, Pivoting to US Expansion

Corporate News

• AMC Robotics Taps HIVE GPU Cloud to Power AI Robotics Development

• HIVE Expands AI Data Centers in Canada as Sweden Bitcoin Hashrate Winds Down

• Bitcoin Miner Creditors Claim $13.6M in BlockFills Bankruptcy

• NVIDIA releases new open models to support autonomous and healthcare AI application

• Cango Reports $452.8 Million Net Loss for 2025 Amid AI Infrastructure Pivot

• Constellation agrees to divest PJM assets to LS Power amid regulatory settlement

Financial News

• Bitdeer Reaches 68 EH/s Hashrate and Completes $375M Note Offering

• Terawulf Adds $500M Bridge Loan for Kentucky AI Data Center as Debt-Fueled HPC Pivot Deepens

• Nebius Signs $27 Billion AI Infrastructure Deal with Meta

• Southern Company Prices $1.3 Billion Junior Subordinated Note Offering

• Nebius Prices Upsized $4 Billion Convertible Debt Offering

Feature

• How Eric Trump Became an Ally of One of China’s Biggest Crypto Companies

Email us

Before we dig in: Our team has advised mining and data center companies since 2014, from ASIC launches on Bitcointalk.org and the great scaling wars to public-market debuts in 2020, environmental opposition in blue states, and today’s pivot from mining to AI/HPC. We bring that hard-won context to crisis communications, rebrands, product launches, and investor-facing messaging. If your organization could benefit from strategic communications or PR support, email us.

For years, nuclear power in the United States was widely seen as a sunset industry. Aging reactors were closing, utilities struggled to justify operating costs, and cheap natural gas combined with subsidized renewables eroded the economics of baseload generation.

Now the trend is reversing — and the catalyst isn’t government policy alone. It’s the explosive electricity demand from AI data centers.

Across recent annual filings, major utilities and power producers describe what amounts to a nuclear renaissance, fueled by long-term power contracts with hyperscalers such as Microsoft, Amazon and Meta. These companies are no longer simply purchasing renewable energy credits to offset emissions. They are underwriting entire nuclear facilities to secure carbon-free power that runs around the clock.

The reason is simple: AI infrastructure doesn’t sleep. Training clusters and high-performance computing facilities operate continuously, requiring massive quantities of reliable electricity that intermittent sources like wind and solar cannot guarantee on their own.

Nuclear plants, once considered uneconomic relics, suddenly look like strategic infrastructure.

Hyperscalers Step In as Nuclear Backers

The clearest signal of the shift is the emergence of long-term nuclear power purchase agreements (PPAs) with technology companies acting as anchor customers.

Microsoft signed a 20-year deal with Constellation Energy to support the restart of the former Three Mile Island Unit 1 reactor in Pennsylvania. The plant, retired in 2019 for economic reasons, is being revived as the Crane Clean Energy Center and will deliver roughly 835 megawatts of carbon-free power once operational.

Amazon’s cloud unit AWS struck a similar 20-year agreement with Vistra Corp for electricity from the Comanche Peak nuclear plant in Texas, securing about 1,200 megawatts of generation.

Meta has gone even further. The company signed a 20-year agreement with Vistra covering more than 2,600 megawatts of nuclear output across several plants in the PJM grid. It also reached a separate deal with Constellation Energy to purchase the output from the Clinton Clean Energy Center in Illinois.

These agreements effectively turn hyperscalers into long-term underwriters of nuclear generation — guaranteeing revenue streams that allow operators to maintain, upgrade or even restart reactors that might otherwise shut down.

An Early Signal From Bitcoin Mining

Before hyperscalers began underwriting nuclear plants for AI infrastructure, Bitcoin miners had already begun experimenting with colocating compute directly next to large power assets.

One of the clearest examples is in Pennsylvania, where Talen Energy operates the Susquehanna nuclear power plant. In 2021, Bitcoin miner TeraWulf formed a joint venture with Talen called Nautilus Cryptomine to build a large-scale mining facility directly adjacent to the plant.

The project allowed the mining operation to draw electricity directly from the nuclear facility behind the meter, bypassing the wholesale market and securing extremely low-cost power. TeraWulf described Nautilus as the first U.S. Bitcoin mining operation powered entirely by nuclear energy.

The arrangement demonstrated a concept that has since become central to the AI data-center buildout: colocating energy-intensive computing loads directly with large baseload generation assets.

That experiment later evolved into something much larger. In 2024, Talen repurchased TeraWulf’s stake in the Nautilus venture and shifted its strategy toward supplying power to hyperscale data centers. The company sold its adjacent Cumulus data center campus to Amazon Web Services for $650 million, enabling AWS to develop a large campus connected to the Susquehanna plant.

The progression illustrates how prime power-adjacent sites are being repriced in the AI era. Locations once used for Bitcoin mining are now becoming anchor points for hyperscaler data infrastructure.

Retired Reactors Return to the Grid

The new economics are already reversing retirement decisions.

Constellation’s restart of the Crane Clean Energy Center marks one of the most visible examples. The project is supported by a 20-year Microsoft contract and backed by a U.S. Department of Energy loan guarantee of up to $1 billion from the Federal Financing Bank. The loan carries a 30-year term maturing in 2055 with interest set just 37.5 basis points above U.S. Treasury rates.

Another example is the Duane Arnold Energy Center in Iowa. NextEra Energy Resources is pursuing the recommissioning of the plant, which shut down in 2020. The company has secured a 25-year power purchase agreement for the facility’s full capacity and submitted an application to the Nuclear Regulatory Commission to reinstate the operating license.

If approved, the reactor could return to service as early as 2029.

Even as companies pursue restarts, they are also extending the lifespan of existing facilities. Dominion Energy has received approvals to operate its Surry and North Anna reactors for up to 80 years, while Vistra secured a license extension for its Perry plant through 2046.

Turning Old Reactors Into Bigger Ones

Utilities are also squeezing more capacity out of existing reactors through what the industry calls “uprates.”

These are engineering upgrades that increase a plant’s electricity output without building an entirely new reactor.

Vistra plans to add about 433 megawatts of additional capacity across three plants — Perry, Davis-Besse and Beaver Valley — largely to meet commitments under its long-term agreement with Meta.

Constellation is pursuing a smaller expansion at the Clinton facility, adding about 30 megawatts of additional output as part of its supply arrangement with the social media company.

While incremental, these upgrades highlight a broader shift: existing nuclear assets are increasingly being treated as expandable infrastructure for the data economy.

The Grid Problem AI Created

The surge in nuclear investment ultimately reflects a deeper structural challenge facing power grids.

Utilities across the U.S. are reporting unprecedented electricity demand growth driven largely by data centers.

Dominion Energy, for example, disclosed that data centers now account for roughly 28% of electricity sales in its PJM Dominion zone. Similar demand spikes are emerging in markets such as Texas and the Midwest.

Because hyperscalers increasingly target hourly carbon-free energy — matching clean electricity supply with consumption every hour — nuclear power provides a uniquely valuable resource: reliable, emissions-free generation that operates continuously.

In other words, nuclear power is becoming the backbone of the AI era’s electricity system.

And if the current wave of data center development continues, the industry that once struggled to justify its survival may soon find itself at the center of the world’s fastest-growing energy market.

Hardware and Infrastructure News

• Northern Data Touts 85% GPU Allocation as AI Infrastructure Revenue Jumps

• Cango Unplugs 30% of Bitcoin Hashrate as Hashprice Slump Bites

• Foundry Expands into Zcash Mining Pool as Bitcoin Hashprice Slides

Corporate News

• Greenidge Generation Reports 2025 Preliminary Results, Obtains 100MW Power Approval for Datacenter Expansion

• Cathedra to Merge With Sphere 3D in US-Listed Bitcoin Mining Deal

• Bitfarms Expands Leadership Bench to Support AI HPC Build-Out, US Domiciliation

• OpenAI to Acquire AI Security Startup Promptfoo to Bolster Frontier Platform

• Nebius Names Dan Lawrence to Lead Expansion in the US as Senior Vice President and General Manager for the Americas

• Cadillac Formula 1® Team Joins Forces with Core Scientific as Official Data Center Partner

Financial News

• Duke Energy Announces $6 Billion Common Stock Offering Program

• Duke Energy Prices Upsized $1.3 Billion Convertible Notes Offering

• Nscale Raises $2B for AI Infrastructure Expansion With Nvidia, Jane Street Backing

• Andreessen Horowitz Leads $500 Million Series B for Nexthop AI

• NVIDIA Invests $2 Billion in Nebius to Scale 5-Gigawatt AI Cloud Infrastructure

Feature

• Why this fired OpenAI employee is betting $1bn on Bitcoin miners

• The 20 Million Milestone: Bitcoin Mining Is, and Always Will Be, an Energy Business

Email us

It’s not news that public mining hodlers have been selling coins for months. Riot liquidated four times its production in December; Cango sold 4,451 BTC in February, or 60% of its reserves; and Bitdeer cleared out its bitcoin reserves last month.

But the latest round of annual filings this week suggests the pace may be accelerating — and could get even more turbulent ahead.

Fresh disclosures from several mining companies indicate that treasury liquidations picked up sharply after Bitcoin’s post-October slide squeezed margins and pushed mining economics below breakeven. Together, they sold more than 15,000 BTC in five months.

Over the past week, Core Scientific revealed in its latest annual report that it plans to sell all of its roughly 2,500 BTC holdings in the first quarter, with 1,900 BTC already sold in January.

Riot noted that bitcoin’s downward trend "may necessitate the sale of a greater volume" of its bitcoin than previously anticipated, to maintain the liquidity required for ongoing operations and working capital

Even MARA — historically one of the industry’s most aggressive Bitcoin accumulators — has softened its stance. While the company still held more than 53,000 BTC at the end of 2025, it has adjusted its treasury policy to allow sales beyond just newly mined coins.

In other words: the HODL era is quietly loosening.

Miners Are Selling Into a Profit Squeeze

The selling pressure shouldn’t come as a surprise.

Since October, Bitcoin’s average price has slid steadily, pushing mining economics into uncomfortable territory.

Hashprice — the key profitability metric for miners — recently fell below $30/PH/s, a level below the total cash-based hashcost for most public miners, according to TheEnergyMag’s analysis of the latest quarterly filings.

In simple terms: many miners are currently operating at or below breakeven.

The gap between hashprice and hashcost has historically been one of the strongest triggers for treasury liquidation.

And the numbers are starting to show it.

Debt Is Rising — and Bitcoin Is the Collateral

Unlike previous downturns, many public miners entered 2025 with significantly more leverage.

Across the industry, companies tapped credit facilities, Bitcoin-backed loans, and senior secured bonds to fund both operating costs and the massive buildout of AI and high-performance computing (HPC) data centers.

That leverage often comes with Bitcoin collateral requirements.

As of Dec. 31, three major miners — Hut 8, MARA and Riot Platforms — had pledged more than 14,500 BTC combined against loans.

The catch: as Bitcoin prices fall, loan-to-value ratios rise.

With Bitcoin dropping sharply after the new year, these collateral ratios likely increased — meaning additional Bitcoin may already have been pledged to maintain lending thresholds, which are typically set at around 60%.

A Classic Miner Capitulation Signal?

Historically, large-scale miner selling has often coincided with late-stage bear market conditions.

When margins compress and treasuries shrink, weaker miners exit and stronger operators consolidate — setting the stage for the next recovery.

Whether this cycle follows the same pattern remains uncertain.

Bitcoin has already rebounded to around $73,000, lifting hashprice back above $30/PH/s.

But that recovery hasn’t fully restored mining profitability. For most public operators, cashflow remains tight.

Which leaves the industry in an unusual position:

Miners are almost simultaneously selling Bitcoin to survive the present while borrowing aggressively to build AI infrastructure for the future.

If history is any guide, miner capitulation has often marked the end of painful market phases.

But for now, the industry is still very much in the middle of the squeeze.

Regulation News

• Russian Draft Notices Issued to Chinese Digital Asset Workers

• U.S. Nuclear Regulatory Commission Publishes Proposed Rule and Guidance for Fusion Regulatory Framework

Hardware and Infrastructure News

• American Bitcoin to Add 3 EH/s With New S21XP Purchase, Reactivating Drumheller Site

• IREN Expands AI Cloud Capacity to 150,000 GPUs

Corporate News

• MARA and Starwood Form Strategic Partnership to Develop 2.5 GW AI Infrastructure Platform

• TeraWulf CEO Sets Plan to Sell 1.65M Shares Amid $3.2B AI Pivot

• Core Scientific Sold 1,900 BTC in January; to Sell Remaining Bitcoin in Q1

• MARA Expands BTC Sales Flexibility After Mixed 2025 Bitcoin Treasury Results

• Canaan Announces Open Market Share Purchases by CEO NG Zhang and CFO

Financial News

• OpenAI Secures $110 Billion in New Funding to Fuel Global AI Scaling

• Blackstone to Launch Public Entity for AI Data Center Investments

• Applied Digital Targets $2.15B Bond Sale for North Dakota AI Data Center

• Core Scientific Lands $500M Loan at ~7.8% as AI Data Center Financing Boom Continues

Email us

If you want to understand the AI and data center boom in 2026, don’t just look at GPUs, megawatts, or backlog announcements.

Look at the bond market.

Over the past 12 months, more than $33 billion of long-term senior notes have been issued by just a short list of bitcoin mining/AI infrastructure companies, utilities and power producers — and that’s excluding convertible notes. This isn’t equity dilution. It’s hard debt: Fixed coupons; Real maturities; Real interest expense.

And the spread between who pays 4% and who pays 9% tells you almost everything about how the market is underwriting the data center arms race.

The 9% Club: AI and Bitcoin Infrastructure

At the high-yield end of the spectrum, the capital isn’t cheap.

CoreWeave printed:

• $2.0B at 9.25% (May 2025)

• $1.75B at 9.00% (July 2025)

Applied Digital:

• $2.35B at 9.25% (Nov 2025)

TeraWulf:

• $3.2B at 7.75% (Oct 2025)

Cipher Mining:

• $1.4B at 7.125% (Nov 2025)

• $2.0B at 6.125% (Feb 2026)

Cipher’s February deal is interesting. In just three months, its pricing improved by a full percentage point even as it doubled down with a $2 billion issuance. That suggests there’s still appetite for “compute-backed” credit, especially when colocation leases and power contracts are in place.

But zoom out and compare this to regulated utilities and power producers.

The 4–5% World: Incumbent Energy Giants

Now look at the other side of the ledger.

Dominion Energy:

• Multiple tranches between 4.6% and 5.65%

NRG Energy:

• Mostly 4.7%–6.0%

Vistra Corp.:

• January 2026: $2.25B at 4.70% and 5.35%

The Southern Company:

• Several issuances clustered around 4%–5.5%

Constellation Energy:

• $2.75B in January 2026, multi-tranche, largely sub-5% depending on maturity

Same macro environment. Same Treasury curve. Different credit pricing.

The message from lenders is clear: regulated load and contracted generation still get treated as infrastructure. AI and bitcoin, even when attached to long-term offtake agreements, are still treated as growth credit.

The Spread Story is a Credit-Rating Story

If you line these issuers up by coupon, you get a rough risk ladder:

• 4%–5%: Regulated utilities and diversified power producers

• 5%–6%: Stronger independent generators

• 6%–9%: Bitcoin miners and AI infrastructure builders

Regulated utilities and established power producers tend to sit in the investment-grade universe, with long operating histories, predictable (often regulated) cash flows, and deep institutional demand for their paper.

On the other side, the newer “compute” names — especially the ones still scaling, still building, or still proving the durability of their customer base — are typically borrowing as high-yield / speculative-grade credits. Even when they have real contracts, the market still prices in execution risk, refinancing risk, and the reality that capex eats cash before it creates cash.

Why So Much Debt, So Fast?

The common thread isn’t crypto cycles. It’s data center demand.

Utilities are openly revising capital plans upward. Southern now expects a $78.1 billion investment plan through 2030, with $15.9 billion in 2026 alone — explicitly citing projected load growth from data centers. Dominion similarly flagged billions in anticipated long-term debt issuance (between $6B and $9.5B in 2026) to support infrastructure expansion driven by large new data center customers.

On the AI side, the logic is simpler: secure power first, figure out monetization later.

For miners transitioning into HPC, the debt stack is becoming the bridge between legacy bitcoin cash flows and future AI tenancy – assuming there’s still cash flow from bitcoin mining. For AI players like CoreWeave, it’s about scaling ahead of revenue realization under hyperscaler contracts.

Is This a Bubble — or Just a Capex Supercycle?

There’s a bigger question hanging over all of this.

If AI demand holds, these coupons may look entirely rational. Debt gets refinanced lower. Assets appreciate. Power scarcity becomes the bottleneck.

But if AI demand cools — or hyperscaler buildouts lose momentum — the 7%–9% debt stack tied to merchant-exposed compute assets could become burdensome fast, particularly with bitcoin mining economics providing little buffer.

Remember: most of these maturities cluster around 2030–2036. That’s not far away in infrastructure time. This isn’t just a power story anymore. It’s a balance sheet story.

Regulation News

• Trump says he will raise US global tariff rate from 10% to 15%

• Trump to announce data center energy deals during State of the Union

Hardware and Infrastructure News

• Bitcoin Difficulty Jumps 15% to Push Hashprice Below $30/PH/s

• Canaan Acquires Cipher’s Stake in Texas Bitcoin Mines in $40M Stock Deal

• Bitfarms Wins Local Approval to Advance Pennsylvania AI Data Center Project

• Fire in Wenatchee, Washington Tied to Bitcoin Mining Activity

Corporate News

• Blue Owl Struggles to Line Up Debt for $4B CoreWeave AI Data Center

• Bitdeer Clears Out Bitcoin Reserves While Refinancing Higher-Cost Convertibles

• Cipher Mining Rebrands as Cipher Digital to Double Down on AI Data Center Leases

• Cipher CEO Sets New 10b5-1 Plan Covering Up to 1.5M CIFR Shares

• NextEra Energy to Raise $2 Billion in Equity Units to Fund Power Projects

Financial News

• Tether Reloads on Bitdeer in $42M Buys After Selling Near 2025 Peak

• CoreWeave Seeks $8.5B Loan as AI Infrastructure Debt Piles Up

• Nvidia Revenue Tops $216B as Blackwell Ramp Lifts AI Data Center Sales

• Hut 8 Pledges 4,533 Bitcoin for $200M Coinbase Loan as BTC Slides

• American Bitcoin Tops 6,000 BTC Holdings After Q4 Production, ATM Buys

Email us

On paper, 14 public bitcoin miners are about to build a small country’s worth of power infrastructure.

In reality, they have roughly 11 gigawatts online today — and about 30 gigawatts more sitting in “pipeline” decks, interconnection queues, and early-stage development plans, according to the chart compiled by TheEnergyMag.

That three-to-one gap is the clearest signal yet that the mining industry’s next cycle won’t be fought only with cheaper or more efficient ASICs or higher uptime, but with something far more contested: who can secure power, finance it, and deliver data-center-grade capacity fast enough to matter.

This is the megawatt arms race of the AI boom—assuming the boom lasts long enough, and demand stays deep enough, for these companies to actually monetize the power they deliver in the first place.

But before we get carried away, it’s worth acknowledging why the bubble question is unavoidable. Every bull market comes with a new narrative that promises to lift the whole sector. In 2017, it was blockchain. In 2021, it was the hashrate arms race. In 2025, it became “AI data centers.” The story changes, the capital follows, and the industry relabels itself — often faster than the underlying business model actually changes.

Riot is a good example of that identity churn. Before 2017, it was a biotech play; then it became Riot Blockchain, rebranded again as Riot Platforms, and now faces pressure from an activist investor to accelerate its pivot away from bitcoin mining toward AI data center colocation. The names and narratives have evolved with the market cycle. The harder part is whether the assets and execution can keep up.

From GPU to ASIC and back to GPU!

The AI boom has turned “pipeline megawatts” into a valuation talking point. But “pipeline” can mean everything from a signed lease and an active construction site to a project in feasibility studies, a grid application, or simply a target a company is marketing. Some of those megawatts will become energized, revenue-generating capacity. Some will slip by years. Some will never materialize, especially if AI demand cools or financing conditions change.

Even if power delivery does catch up, there’s a second issue that doesn’t get enough attention: it’s genuinely difficult to transition from monetizing ASIC hashrate to monetizing data center infrastructure. It’s also striking to watch the industry’s arc—from GPUs to ASICs, and now back to GPUs again, but this time in a completely different business.

Capex aside, the operational and commercial risk is hard to ignore. Unlike bitcoin mining, monetizing GPUs for AI companies or enterprise customers depends heavily on product fit, sales execution, and service delivery. In bitcoin mining, the formula is simple: power plus ASICs produces bitcoin. In AI infrastructure, you can secure the power and still fail to deliver a marketable product—or fail to sell it on terms that meet customers’ expectations. That’s what makes “pipeline megawatts” a more fragile metric.

The 11 GW operational figure is part of the story that already exists. The 30 GW pipeline figure is the story miners are trying to make true. The AI pivot is a real strategic response to hashprice pressure and capital-market incentives, but it also carries the familiar scent of bull-market hype.

If the AI boom sustains, the winners won’t be the companies that announce the biggest pipelines. They’ll be the ones that can close the infrastructure gap—turning planned megawatts into energized capacity, energized capacity into critical IT load, and critical IT load into contracted revenue—without stretching their balance sheets or overpromising timelines. None of this makes the pivot impossible; it just makes it harder, and slower than the narrative often assumes.

Hardware and Infrastructure News

• HIVE Signs $30M GPU Cloud Contracts to Expand Canada AI Data Center - TheEnergyMag

• Caturus Secures 20-Year LNG Offtake Agreement with Aramco Trading for Louisiana Export Facility - TheEnergyMag

• Mon Power and Potomac Edison Announce New Natural Gas Power Plant in West Virginia - TheEnergyMag

• Transource Energy and FirstEnergy Transmission Joint Venture Approved for Major Electric Transmission Project in Central Ohio - TheEnergyMag

• Soluna Expands Blockware Partnership, Adding 6 MW Bitcoin Mining Capacity at Project Dorothy 1 - TheEnergyMag

• Meta to Deploy Millions of NVIDIA GPUs as AI Buildout Pushes CapEx Higher - TheEnergyMag

• Yotta to Build $2 Billion AI Hub in India Using Nvidia Blackwell Chips - TheEnergyMag

• Tata Group, OpenAI Plan AI Compute Hub in India - TheEnergyMag

Corporate News

• OpenClaw founder Steinberger joins OpenAI, open-source bot becomes foundation - Reuters

• IREN Appoints John Gross as Chief Innovation Officer to Enhance Data Center Engineering - TheEnergyMag

• Portland General Electric Acquires Washington Utility Operations from PacifiCorp - TheEnergyMag

• 1606 Corp. Signs Term Sheet to Acquire 55 MW Texas Power Generation Facility and 50,000 Sq. Ft. Data Center-Ready Infrastructure Site - TheEnergyMag

• Activist Starboard Pushes Riot to Speed Up Shift From Bitcoin Mining to AI Data Centers - TheEnergyMag

• Core Scientific Board Set for Overhaul Under Two Seas Pact - TheEnergyMag

Financial News

• Constellation Energy Surges 10% This Week on Data Center Deals and Analyst Upgrades - TheEnergyMag

• Blackstone Commits Up to $1.2 Billion to Neysa for India’s Domestic AI Infrastructure Build-Out - TheEnergyMag

• HIVE Digital Technologies Reports Record Q3 Revenue Driven by Bitcoin Hashrate Growth - TheEnergyMag

• Bitdeer Plans New $300M Convertible Bonds Aimed at Deleveraging - TheEnergyMag

Email us

A quick note before we dive in: TheMinerMag is now TheEnergyMag.

What began as focused coverage of Bitcoin mining has expanded into something broader. AI-driven data center demand is reshaping utilities, capital markets and long-term infrastructure investment. The center of gravity has shifted from pure hashrate to power.

TheEnergyMag reflects that evolution. Read more here.

The scale of capital now flowing into AI infrastructure is starting to look less like a typical tech upcycle and more like a generational buildout — one that stretches from Bitcoin miners to regulated utilities to Big Tech itself.

Over the past few weeks alone, we’ve seen a rare 100-year bond, a Bitcoin mining giant deploy more capital in one year than it spent in the three years after its IPO — and two of the largest U.S. utilities line up tens of billions in regulated investment and fresh debt to support rising data center load.

This isn’t incremental expansion. It’s a balance-sheet shift.

IREN’s spending curve bends sharply upward

Start with IREN.

In its latest quarter, net spending on property, plant and equipment topped $800 million — significantly above prior quarters. But the annual view is even more telling: IREN’s total net PP&E spending in 2025 alone exceeded the combined total from 2022 through 2024.

In practical terms, the company deployed more capital in a single year, building AI data center infrastructure and procuring GPU hardware than it spent across three years expanding its Bitcoin mining fleet post-IPO, according to TheEnergyMag’s analysis.

That shift highlights how different the AI buildout looks from the last mining cycle. High-density GPU clusters, substation upgrades, advanced cooling and long-duration interconnections resemble utility-scale infrastructure more than modular ASIC deployments. The capex profile follows accordingly.

Duke’s $103B plan — and the load behind it

On the utility side, Duke Energy is openly framing this as a multi-year expansion cycle.

The company outlined a $103 billion five-year capital plan, which CEO Harry Sideris described during the earnings call as “the largest fully-regulated capital plan in the industry focused on critical energy infrastructure investments that strengthen the system and serve increasing load.”

That “increasing load” is tied in part to AI and advanced manufacturing demand.

Sideris also stressed that as investment needs accelerate, “the cost of energy has been and will remain a key focus for Duke Energy.” In a news release accompanying the results, he said the company is entering the year “with incredible momentum,” adding that the fundamentals of the business “have never been stronger” and pointing to operations in some of the most attractive jurisdictions in the country.

In 2025 alone, Duke broke ground on 5 gigawatts of new dispatchable generation. Management extended its long-term adjusted EPS growth target of 5% to 7% through 2030, supported in part by contracted demand.

The quarterly results also reflected the other side of heavy capital spending: higher depreciation and interest expense tied to a growing asset base.

NextEra Energy has been equally active in capital markets.

On Feb. 7, it announced raising $1.3 billion in new debt across two tranches. On Feb. 10, its subsidiary issued another €1.3 billion in euro-denominated debentures. The structure — blending intermediate and long-dated maturities — mirrors how large utilities fund long-lived generation and grid assets.

While proceeds were not specifically itemized, the activity comes as NextEra expands its partnership with Google Cloud to co-develop multiple gigawatt-scale data center campuses in the U.S., including land, interconnection and dedicated generation resources.

Google sells a century

At the other end of the stack, Alphabet Inc. — parent of Google — sold a rare 100-year bond as part of a $31.51 billion global debt raise.

The century tranche raised 1 billion pounds at a 6.125% coupon and reportedly drew demand nearly ten times the amount offered. The broader transaction included sterling, Swiss franc and U.S. dollar bonds, with maturities stretching from three years out to 2066.

Century bonds are typically associated with governments or regulated utilities. Analysts noted that ultra-long-term issuance of this kind signals a pivot from asset-light models toward long-term infrastructure.

Alphabet’s deal stands out not just for its maturity but for its scale. Alphabet, Microsoft, Amazon and Meta are collectively expected to spend at least $630 billion this year, with most of that directed toward data centers and AI chips.

In that context, a 100-year bond begins to look less like a novelty and more like a financing tool for infrastructure measured in decades.

Viewed together, the signals are clear. Compute operators are spending at levels that eclipse their prior mining cycles. Utilities are formalizing record capital plans and tapping debt markets to expand generation and grid infrastructure. Earnings calls now routinely reference AI-driven load growth alongside traditional drivers.

The AI boom is not just reshaping revenue mix — it is redefining the scale and duration of capital deployment across the energy-compute stack.

Hardware and Infrastructure News

• Bitcoin Miners Could Face Crisis After BTC Price Falls 50% From Peak - Decrypt

• Bitcoin Mining Difficulty Saw Biggest Drop Since 2021 China Ban - TheEnergyMag

• Soluna Begins Energization of Project Kati 1, adding 83 MWs to Total Capacity - Link

• Bell Canada Advances AI Fabric Expansion With Proposed 160-Acre Campus South of Regina - TheEnergyMag

Corporate News

• Bitfarms Deleverages, Unveils Keel Rebrand as Bitcoin Mining Margins Sink Below $30/PH/s - TheEnergyMag

• Bitcoin Miner NFN8 Files Chapter 11 After Fire, Lease Strain as Hashprice Hits Record Low - TheEnergyMag

• BlockFills Suspends Client Withdrawals Amid Market Volatility - TheEnergyMag

• Cipher Welcomes Thomas Duda to the Company’s Board of Directors - Link

• Anthropic says it will pay 100% of the grid upgrade costs tied to its AI data centers - Business Insider

Financial News

• NextEra Raises $1.3B in Debt as AI Data Center Power Demand Accelerates - TheEnergyMag

• Cango Sells 4,451 BTC, Cuts Bitcoin Reserves by 60% to Repay Loan - TheEnergyMag

• Canaan revenue more than doubles in Q4 as bitcoin miner grows treasury to record levels - The Block

• Cipher Mining and TeraWulf are buys, MARA a sell, as Morgan Stanley begins bitcoin miner coverage - CoinDesk

Email us

If it wasn’t already obvious, ERCOT’s message this week was pretty blunt: the way Texas has been approving massive new power users no longer works.

At a Feb. 3 workshop that drew more than 700 participants, the Texas grid operator rolled out a draft plan to overhaul how large loads — think AI data centers, Bitcoin miners, and other hyperscale compute — connect to the Texas grid. The backdrop is a staggering backlog: more than 250 gigawatts of large-load requests, roughly three times ERCOT’s current peak demand.

For years, large loads have been studied one at a time by transmission providers. That model breaks down when hundreds of gigawatts pile up behind it. New projects change assumptions, old studies get invalidated, and developers get stuck in restudy hell. ERCOT officials made it clear they’re close to the point where they’d have to go back to projects that already “passed” and tell them those approvals might need to be reopened.

That’s what this proposal is trying to avoid.

The core idea is simple: stop studying projects in isolation and start studying them in batches. Under ERCOT’s plan, large loads would be grouped into system-wide batch studies run every six months. ERCOT would look at their combined impact, figure out what the grid can actually support, and then allocate power to each project — potentially less than requested, at least at first.

To kick things off, ERCOT is creating a transitional phase called “Batch Zero,” which is meant to deal with the most advanced projects already in the queue. Even here, though, not everyone gets a free pass. ERCOT said around 8.2 gigawatts of load could be subject to reassessment, depending on how far along projects really are and whether they meet new readiness and commitment standards tied to Senate Bill 6.

Jeff Billo, ERCOT’s vice president of interconnection and grid analysis, summed up the motivation pretty cleanly during the workshop. ERCOT wants to get to “one study to rule them all,” rather than constantly revisiting old assumptions as new requests flood in.

A big shift embedded in the proposal is ERCOT’s open embrace of partial and ramped service. Instead of promising full power on day one, ERCOT would tell projects how much load they can count on each year, with allocations increasing as transmission upgrades come online. ERCOT staff argued that most developers would rather have a clear, constrained answer than an open-ended maybe.

Not everyone was fully convinced. Meta’s energy program manager, Katie Bell, pointed out that some projects have been in the interconnection process for 18 months and still wouldn’t qualify for Batch Zero, raising concerns about even more delays. Others echoed worries about transparency — how allocation decisions are made, how assumptions change, and how developers are supposed to plan around them.

ERCOT acknowledged those concerns and said predictability is the main goal. Fixed timelines, explicit go/no-go decision points, and clearer communication are supposed to replace the current fog of restudies and shifting goalposts. The tradeoff is that not every project gets what it asked for, at least initially.

One area that drew noticeable interest was how ERCOT plans to treat flexible loads. Officials repeatedly referenced strong stakeholder support for recognizing controllable loads and co-located generation — concepts that matter a lot for Bitcoin miners and hybrid compute sites. The rules here aren’t fully baked yet, and ERCOT admitted much of this won’t be finalized until later in 2026, but the direction is clear: flexibility is starting to count for something.

ERCOT plans to hold another workshop and then take the proposal to state regulators later this month. Until the Public Utility Commission signs off, the existing interconnection process stays in place.

For miners, the subtext of the workshop was hard to miss. In Texas, access to power is becoming less about who showed up first and more about who’s real, who’s flexible, and who can live with ramping instead of instant scale. That’s a big shift — and one that could end up favoring loads that already know how to dance with the grid.

Regulation News

• British Columbia Opens 400MW Power Competition for AI Data Center Projects - TheMinerMag

• Texas Approves 7.65 GW Air Permit for Pacifico Energy’s AI Power Campus - TheMinerMag

• Pennsylvania Approves 15 On-Site Gas Generators Backing AI Data Center in Clearfield - TheMinerMag

Hardware and Infrastructure News

• Bitcoin miners get an open-source alternative as Tether launches MiningOS - CoinDesk

• Terawulf Adds 1.5 GW to AI Power Portfolio With Kentucky, Maryland Site Acquisitions - TheMinerMag

• Western Kentucky Aluminum Plant Sold for Redevelopment into Data Center - WDRB

• Bitcoin Hashprice Hits Record Low as BTC Falls Below $70,000 - TheMinerMag

Corporate News

• BitRiver Insolvency Looms as Bitcoin Miner Founder Placed Under House Arrest - TheMinerMag

• Cango Breaks Bitcoin HODL Stance, Sells 550 BTC Amid Mining Pressure - TheMinerMag

Financial News

• Cipher Seeks $2B in New Secured Notes to Fund AI Data Center Build in Texas - TheMinerMag

• Cipher Bond Tied to AWS Data Centers Gets $13 Billion of Orders - Bloomberg

Email us

Fresh off a $2 billion equity investment from Nvidia, CoreWeave is back at the center of the AI infrastructure narrative.

The Nvidia check — announced this week alongside a broader expansion of the two companies’ partnership — reinforced CoreWeave’s position as one of the most important pure-play GPU infrastructure operators in the market. Shares jumped more than 10% on the news and closed Wednesday at $106. That was well below a mid-June peak near $184, though it is still up 180% since its market debut.

But while the strategic endorsement grabbed headlines, SEC filings tell a longer, quieter story unfolding since the company’s March 2025 IPO.

Based on Form 144 filings reviewed by TheMinerMag, CoreWeave insiders have sold roughly 17.3 million shares for about $1.69 billion in gross proceeds since going public.

That selling has been dominated by the company’s founders and senior leadership. CEO and co-founder Michael Intrator, chief strategy officer and co-founder Brian Venturo, and chief development officer and co-founder Brannin McBee together sold about 10.9 million shares for roughly $1.09 billion — close to two-thirds of total insider proceeds in the dataset.

McBee has been the largest seller among the group, unloading about 5.7 million shares for roughly $558 million. Venturo follows with about 3.9 million shares sold for around $399 million, while Intrator sold roughly 1.35 million shares for about $136 million across dozens of transactions from last fall through January. Jack Cogen, a board director since 2017 and a private investor of CoreWeave, sold about 5.35 million shares for roughly $497 million, including several large summer blocks.

Other executives also appear in the filings, though at smaller scales. CFO Nitin Agrawal, SVP of engineering Chen Goldberg and general counsel Kristen McVeety all recorded recurring sales as well. One notable absence: CTO and co-founder Peter Salanki appears to be the only co-founder among the four who has not sold shares via Form 144 filings, based on the data reviewed.

None of the insider selling is unusual on its own. Liquidity events after lockups expire are standard for newly public companies, particularly ones that have gone from crypto mining roots to tens of billions in market capitalization in under a decade.

The bulk of the founders’ share sales were executed under pre-arranged Rule 10b5-1 trading plans adopted in advance. That structure points to planned liquidity rather than reactive selling tied to day-to-day price moves or news flow.

In fact, Bloomberg reported on Oct. 10 that CoreWeave insiders had already sold roughly $1 billion worth of shares following the IPO. Based on the Form 144 filings since then, another roughly $600 million in gross proceeds has been realized in the months that followed, suggesting the selling has not slowed materially even as the stock retraced from its summer highs and Nvidia stepped in with fresh capital.

What makes this worth lingering on is where CoreWeave came from, and there’s a counterfactual lurking beneath CoreWeave’s trajectory that’s hard to ignore: Ethereum’s 2022 switch from proof-of-work to proof-of-stake, which abruptly sidelined vast amounts of GPU mining capacity.

That transition didn’t just strand hardware — it forced operators with GPU fleets, power access and operational know-how to rethink their business models. CoreWeave, which began as a GPU-focused crypto miner in 2017, was among the earliest to repurpose that infrastructure toward rented compute just as demand for AI workloads began to accelerate.

In that sense, CoreWeave’s success sits at the intersection of two structural shifts: Ethereum exiting proof-of-work and AI entering its scale-up phase. Had Ethereum continued to rely on GPU mining, it’s not hard to imagine a very different timeline — one where GPUs remained tied up securing blockchains rather than training models, and where CoreWeave’s pivot may have been slower, smaller, or unnecessary altogether. Instead, the post-merge vacuum helped turn surplus GPUs into a launchpad, setting the stage for what ultimately became one of the market’s most valuable AI infrastructure platforms.

In that sense, CoreWeave’s story will sound familiar to anyone who has watched bitcoin miners evolve over the past few cycles: power, hardware access and operational muscle mattered first, and narratives followed later. CoreWeave made the leap earlier — and with GPUs rather than ASICs.

Hardware and Infrastructure News

• Bitcoin Miners Unplug 110+ EH/s to Ease Grid Strain During Arctic Blast - TheMinerMag

• Bitcoin miners brace for major ice storm across southern US - Cointelegraph

Corporate News

• Bitcoin miners HIVE, Bitfarm and Bitdeer downgraded as analyst warns on AI shift - CoinDesk

• CleanSpark shares slide 10% as snowstorm risk, $45 million CEO pay disclosure spook investors - The Block

• Amazon to Cut 16,000 Jobs in Latest Round of Layoffs - The New York Times

• Bit Digital Seeks to Ease Overhang as AI Spin-Off WhiteFiber Lock-Up Expires - TheMinerMag

Financial News

• Vistra Raises $2.25B for Cogentrix Acquisition Amid Growing AI Data Center Load - TheMinerMag

• Nvidia Invests $2 Billion in CoreWeave as AI Data Center Buildout Accelerates - TheMinerMag

Features

• Inside The Trade That Made Mike Novogratz An AI Kingpin—And Could Net Him $20 Billion - Forbes

Email us

For much of the past decade, bitcoin miners expanding across the United States learned that access to cheap power and industrial land did not guarantee social license. After China’s 2021 mining ban pushed activity stateside, projects in New York, Texas, Arkansas and Kentucky were met with complaints over noise, power prices and environmental impact—often after local residents realized how little say they had in the decision-making process.

Now, the U.S. AI compute boom is running into a familiar obstacle.

As hyperscalers and AI developers race to secure power-dense data center capacity, community resistance is emerging as a material constraint, echoing the pushback that once slowed, reshaped or outright stopped bitcoin mining projects.

Across the country, local governments and residents are no longer waiting passively for assurances that AI infrastructure will be different.

From crypto backlash to AI scrutiny

The migration of bitcoin mining to the U.S. revealed a recurring pattern: large, energy-intensive facilities promised jobs and tax revenue, but do not always deliver long-term employment while introducing new stresses on local grids and land use.

AI data centers—despite being quieter and more politically palatable—are now triggering many of the same concerns.

In Texas, Illinois, Georgia and Mississippi, local news outlets have reported packed hearings, zoning disputes and calls for moratoriums as residents question water consumption, backup generation, transmission upgrades and whether rising electricity costs will ultimately be passed on to households.

This week, commissioners in Thomas County, Georgia, voted to temporarily halt new AI data center developments while officials study the long-term impacts on infrastructure and public services. The move reflects a growing trend: communities are slowing approvals not to reject AI outright, but to avoid being locked into consequences they do not yet fully understand.

Industry trackers estimate that $64 billion in U.S. data center projects have already been delayed or blocked by local opposition—an increasingly hard figure for investors to ignore.

Big Tech responds: “paying its own way”

The industry response is starting to change.

Earlier this month, Microsoft rolled out its “Community-First AI Infrastructure” framework, pledging that its data centers would cover the full cost of new generation, transmission and grid upgrades rather than shifting those expenses onto residential ratepayers. The company also committed to water replenishment, transparency and workforce investment in host communities.

Now, OpenAI has taken a similar stance.

OpenAI has committed to “paying its own way” for energy costs tied to its AI data center expansion, signaling that the company recognizes community acceptance and power-market credibility as strategic priorities—not afterthoughts. The move aligns OpenAI more closely with utilities and regulators wary of socializing infrastructure costs driven by private AI demand.

For veterans of the bitcoin mining sector, the language is familiar. Miners that survived local pushback often did so only after renegotiating power contracts, investing in mitigation measures or agreeing to clearer community benefit structures—frequently after costly delays.

There are, however, important operational differences. Bitcoin miners, by design, can curtail power usage or shut down entirely during periods of peak demand or extreme weather, allowing utilities to rebalance load in real time. In several U.S. markets, this flexibility has been used as a grid-management tool, with miners participating in demand-response programs that reduce strain during emergencies. Proponents argue this has helped offset infrastructure costs by supporting grid expansion while lowering energy prices for residential customers during peak periods. AI data centers, by contrast, are built to serve continuous computing workloads and are generally less amenable to rapid curtailment, limiting their usefulness as flexible load in times of grid stress.

Policymakers draw firmer lines

State governments are also recalibrating.

New York Governor Kathy Hochul has proposed stricter safeguards to ensure large data centers pay higher charges tied to grid upgrades and reliability. While framed as a consumer-protection measure, the policy is inseparable from surging AI-driven electricity demand.

New York’s posture is shaped by experience. The state spent years navigating backlash against bitcoin mining facilities, particularly those tied to fossil-fuel generation. AI data centers may wear a different label, but from a grid-planning perspective, they pose many of the same challenges: large, inflexible loads seeking rapid interconnection.

For bitcoin miners pivoting toward AI or HPC colocation, the implications are significant.

Capital markets have largely rewarded AI data center narratives with higher multiples and cheaper capital, often assuming smoother permitting and stronger political support than crypto mining ever enjoyed. Community resistance complicates that thesis.

The AI compute boom is real. The power demand behind it is even more real. But local consent—long treated as a secondary consideration—is reasserting itself as a gating factor.

Bitcoin mining learned this lesson the hard way after 2021. AI infrastructure builders are now discovering that, even in the age of trillion-dollar valuations, local communities may still not buy it.

Regulation News

• Thomas County Board of Commissioners temporarily halt AI data centers coming to the area - WALB

• Pa. lawmakers hear warnings on data centers, rising electric costs - Post-Gazette

Hardware and Infrastructure News

• Bitcoin Hashrate Slides 14% From October Peak as Difficulty Set for Moderate Pullback - TheMinerMag

• Two solo bitcoin miners hit rare $300,000 jackpots in the same week - CoinDesk

Corporate News

• OpenAI Joins Microsoft in Pledging Data Centers Will Not Shift Energy Costs to Residents - TheMinerMag

• BCE’s Data Center Rollout Moving Faster Than Expected, CEO Says - Bloomberg

Financial News

• Riot Stock Jumps 12% as AMD Deal Anchors Texas AI Data Center Expansion - TheMinerMag

• Saudi’s Humain secures up to $1.2 bln to expand AI infrastructure - Reuters

Email us

While most attention in recent months has focused on publicly traded miners racing to add hashrate or pivot toward AI, one of the fastest-growing bitcoin mining operations has expanded largely outside the spotlight.

NIP Group (NASDAQ: NIPG), the $130 million parent company of esports brand Ninjas in Pyjamas, disclosed on Thursday that its bitcoin mining operations produced 151.4 BTC during an initial operating period from September through November 2025. At current prices, that output represents roughly $14.2 million in revenue.

The company also said its installed mining capacity has reached 9.66 EH/s following the partial closing of a previously announced transaction, with another 1.64 EH/s expected to come online later this month. Once fully deployed, NIPG expects total capacity of about 11.3 EH/s—placing it among the top 20 publicly traded bitcoin miners by disclosed hashrate and the largest by reported capacity in the Middle East and North Africa.

From esports to hashrate

NIPG’s move into bitcoin mining began in July, when the company announced it would acquire 3.11 EH/s of on-rack mining capacity from Fortune Peak and Apex Cyber Capital in exchange for newly issued Class A ordinary shares. The transaction closed in September, marking the launch of a dedicated digital computing division within the company.

In November, NIPG expanded the strategy significantly, agreeing to acquire an additional 8.19 EH/s of on-rack capacity from Apex Cyber Capital, Prosperity Oak Holdings and Noveau Jumpstar. That deal, also settled largely through equity issuance and convertible instruments, lifted the company’s stated long-term target to 11.3 EH/s and signaled that bitcoin mining had become a material second business line alongside gaming and entertainment.

Ownership links point back to Antalpha

What makes NIPG’s mining push particularly notable is not just its speed, but who is on the other side of the transactions.

Following the November expansion, Apex Cyber Capital held about 31.2% of NIPG, while Prosperity Oak Holdings controlled roughly 29.4%. Fortune Peak, the original seller of on-rack capacity, shares common ownership with Prosperity Oak through Chiu Chang-Wei.

Chiu is also a director at Antalpha, the financing and treasury arm of Bitmain. In parallel, Chiu has been heading Cango, which has emerged as one of the largest proprietary bitcoin miners following its own pivot into mining. Fortune Peak—the same entity that sold on-rack capacity to NIPG—previously sold mining capacity to Cango during its transformation into a large-scale prop miner. Additionally, Simon Ming Yeung Tang, the chief investment officer of Cango, was appointed to the board of NIPG after the July purchase.

Taken together, the transactions suggest a familiar structure: Bitmain-aligned entities transferring energized hashrate into publicly listed vehicles, financed through equity issuance rather than traditional capex-heavy self-builds.

A growing proprietary mining bloc

The scale of this activity is becoming difficult to ignore.

Cango has disclosed roughly 50 EH/s of mining capacity, while NIPG is approaching 11 EH/s, implying that known Bitmain-aligned proprietary mining operations now total more than 60 EH/s. That figure alone would make the “Bitmain camp” one of the largest bitcoin miners globally, even before accounting for undisclosed capacity or third-party arrangements.

The timing is also notable. In December, Bitdeer sharply increased realized hashrate through the deployment of its own SEALMINER machines, reinforcing a broader industry trend: the largest proprietary miners are increasingly those with access to proprietary hardware.

As public miners grapple with compressed hashprice, rising financing costs and shareholder dilution, vertically integrated players—whether through hardware ownership, financing arms or affiliated public shells—appear to be consolidating influence over a growing share of network hashrate.

Regulation News

• New York Weighs Data Center Energy Charges as AI Strains the Grid - TheMinerMag

Hardware and Infrastructure News

• Vistra, Meta Strike 20-Year Nuclear Power Deal Amid Rising AI Demand - TheMinerMag

• CleanSpark Land Deal in Texas Eyes 600 MW AI Data Center Build - TheMinerMag

• CoreWeave Deploys 16,000 GPUs at Delayed Texas Data Center for OpenAI - TheMinerMag

Corporate News

• Bitdeer Challenges MARA for Top Spot After December Bitcoin Production Jump - TheMinerMag

• Microsoft Vows to Pay Higher Power Rates to Shield Locals From AI Data Center Costs - TheMinerMag

• Bitfarms Announces Board Chair Transition in Anticipation of U.S. Redomiciliation - Link

• Applied Digital Appoints Co-Founder Jason Zhang as President - Link

• HIVE Expands into Paraguay, Launching One of the First Purpose-Built AI BUZZ Cloud Platforms - Link

Feature

• America’s Grip on Bitcoin Mining Slips, Despite Trump’s Ambitions for Dominance - Decrypt

• Texas Gets Tough on Data Center Power – Who’s Next? - Data Center Knowledge

Email us

As 2025 came to a close, the stories that dominated bitcoin mining over the past 12 months point to an industry that quietly but decisively crossed a structural threshold. The year was not defined by a single capitulation event or regulatory shock, but by a steady re-ranking of what actually matters in mining—and who still benefits from it.

The headline number still looks impressive on paper. The network’s monthly average hashrate grew by roughly 34%, rising from 778 EH/s in December 2024 to about 1,048 EH/s by December 2025. Yet beneath that growth sits a more telling shift. North American pools steadily lost ground in terms of block share. At the start of the year, Foundry USA, MARA Pool and Luxor together mined more than 40% of all bitcoin blocks. By December, their combined share had slipped to around 35%, despite continued hashrate additions across the globe.

That divergence—rising hashrate, falling block share—captures the defining narrative of 2025: bitcoin mining is no longer the singular focus for many of the industry’s largest players. For some, mining became one leg of a broader infrastructure strategy. For others, it became a legacy business gradually giving way to AI and HPC data centers, or, in a few cases, something to exit altogether.

Price action offered little relief. Bitcoin did set new all-time highs in October, briefly trading above $125,000. But the relentless growth in network hashrate absorbed most of the upside. By year-end, as prices retraced, hashprice fell to historically low levels below $40/PH/s. Even during a bull year for bitcoin, mining economics failed to meaningfully improve.

On the hardware front, the technology story continued to advance. Bitmain, MicroBT, Canaan, Bitdeer and Auradine all pushed new machines into the low-teens joules-per-terahash range. Yet efficiency gains ran headfirst into collapsing revenue per unit of compute. The result was a sharp shift in market dynamics: for the first time in a bull cycle, the ASIC market tilted decisively in favor of buyers. Public miners began phasing out S19-series machines early in the year, and as the hashprice squeeze intensified, that replacement cycle only accelerated. By year-end, even Bitmain was cutting ASIC prices, signaling that manufacturers may need to focus more on the secondary and replacement market than on pure growth demand.

Investor behavior reinforced the pivot. 2025 will be remembered as the year mining equities decoupled from bitcoin production narratives and re-rated around AI and data center exposure. The strongest stock performers—IREN, Cipher, Applied Digital, Hut 8 and Terawulf—were all tied to major HPC colocation or cloud servicing announcements. Even long-time bitcoin-only operators began repositioning, emphasizing power capacity, land and grid access over incremental exahash growth.

The financing backdrop also looked very different from prior cycles. Mining debt reached new highs in 2025, but the structure had changed. Instead of high-interest, equipment-backed loans like those seen in 2021, much of the capital came via zero- or near-zero-coupon convertible bonds. The proceeds were largely earmarked for data center buildouts and power infrastructure, not additional bitcoin miners.

That evolution leads to a broader conclusion. In 2021, hashrate was the hardest asset in the system. In 2025, and likely beyond, energy became the real currency. Bitcoin miners, by virtue of their power expertise and grid positioning, simply happened to arrive early.

Across seven major public miners, announced power capacity in development now totals roughly 15 gigawatts, a pipeline that increasingly overlaps with prospective AI and HPC data center demand. Much of this capacity is still early-stage—interconnection studies, land control, or preliminary agreements—and there is no guarantee it ultimately reaches full energization. But the sheer scale of the pipeline highlights what has replaced exahash growth as the industry’s preferred signaling mechanism. That dynamic extends beyond miner press releases. ERCOT has received an unprecedented wave of large-load interconnection requests over the past year, driven primarily by data centers rather than bitcoin mining alone.

This shift also helps explain a quieter but symbolically important change: the retreat from monthly disclosures. Public miners once used monthly production and hashrate updates as a way to showcase dominance and growth. In an HPC-first world, those metrics matter far less. For many large operators, there are fewer bitcoin mining muscles left to flex.

Taken together, 2025 was not the year bitcoin mining collapsed. It was the year mining stopped being the center of gravity.

Hardware and Infrastructure News

• Goldman Helps Lead Financing for 5-Gigawatt Texas AI Power Sites - Bloomberg

• Bitfarms Exits LatAm With Paraguay Sale as Bitcoin Miner Deepens Shift to HPC, AI - TheMinerMag

Corporate News

• Riot Revamps Executive Pay as CFO Transition and Data Center Incentives Take Shape - TheMinerMag

• Cipher Appoints Lee Bratcher, Drew Armstrong to Lead Policy and HPC Strategy - TheMinerMag

• Canaan’s new initiative plans to help grow tomatoes with heat from Bitcoin mining - The Block

• Riot’s December Bitcoin Sales Mark Largest Monthly Liquidation as Hashprice Sinks - TheMinerMag

• Calgary company eyes power generation for bitcoin mining as stranded oil and gas well solution - The Globe and Mail

Financial News

• Riot Resets ATM With New $500M Raise as Bitcoin Mining Cash Flow Remains Thin - TheMinerMag

• Hut 8 Expands Bitcoin-Backed Credit Facility With Coinbase to $200M - TheMinerMag

• Vistra to Add 5.5 GW of Gas Capacity in $4B Cogentrix Deal as AI Load Rises - TheMinerMag

Feature

• Meta Acquisition Puts Spotlight on ManusAI Founder’s Early Bitcoin Roots - TheMinerMag

Email us

For years, peaker plants were treated like that dusty backup generator in the basement — something you hope never has to turn on, and definitely something you plan to replace someday. Lately, though, the grid has been reaching for that switch a lot more often.

A recent Reuters deep dive put numbers around what many power markets are now quietly acknowledging: retirement plans are being torn up. In the footprint of PJM Interconnection, about 60% of oil, gas and coal plants that were supposed to shut down have delayed or canceled those plans this year. Most of them are peakers — older units designed to run only during brief spikes in demand.

What changed? Data centers. As AI-driven load piled into PJM, power prices during peak periods jumped, and suddenly these rarely used plants looked less like liabilities and more like insurance policies that actually pay. Reuters highlighted Chicago’s Fisk plant, where NRG Energy reversed course on a planned shutdown after concluding there was now an economic case to keep its eight oil-fired units running. Across PJM, 11 of the 13 plants that avoided retirement this year were peakers.

This lines up neatly with what’s happening in Texas. ERCOT is staring at an avalanche of large load requests, overwhelmingly from data centers chasing AI growth. The sheer scale of those applications dwarfs what the grid can serve in the near term, forcing operators to squeeze more out of existing infrastructure while new generation and transmission crawl through multi-year timelines.

Miner Weekly

164 GW: AI Power Requests Flood Texas Grid

Read more

5 months ago · 2 likes

That same power-market gravity is pulling bitcoin miners beyond Texas. Cipher Mining recently made its first move outside ERCOT with a 200-megawatt site in Ohio, plugging directly into PJM. Bitfarms is already there as well, following its acquisition of Stronghold’s power and infrastructure assets in Pennsylvania — a deal that gave Bitfarms a foothold in PJM alongside generation capacity, not just grid access.

Step back, and the message from Reuters’ analysis is pretty blunt. The grid is keeping the old stuff around because it has to. Peaker plants may only generate a small slice of U.S. electricity in a typical year, but they represent a big pool of capacity that can be switched on when things get tight — and right now, things are very tight.

None of this comes without trade-offs. These units are older, dirtier, and often located in communities that have already spent decades pushing back against industrial pollution. But until new power, transmission, and storage actually show up, grid operators are signaling that every available megawatt matters.

For bitcoin miners and data center operators alike, that’s the backdrop for today’s power scramble: a system leaning harder on its “emergency” buttons, and discovering they’re no longer just for emergencies.

Regulation News

• US Energy Regulator Directs PJM to Clarify Grid Rules for Co-Located AI Data Centers - TheMinerMag

• Georgia Power Wins Approval to Add 10 GW as AI Data Center Demand Surges - TheMinerMag

Hardware and Infrastructure News

• WhiteFiber signs 10-year, 40 MW colocation deal with Nscale valued at about $865 million - CoinDesk

• Bitdeer Leases Nevada Facility to Scale US Bitcoin Miner Production - TheMinerMag

• Cipher Enters PJM With 200 MW Ohio Site Aimed at HPC Workloads - TheMinerMag

• Bitcoin Difficulty Holds Firm, Pointing to Limited Miner Capitulation - TheMinerMag

• Bitmain Slashes Bitcoin ASIC Prices as Weak Hashprice Pressures Hardware Market - TheMinerMag

Corporate News

• Tether-Linked Entities Reportedly Bought Northern Data’s Bitcoin Mining Unit - TheMinerMag

• Bitmain Proxy Boosts Control at Bitcoin Miner Cango with $10.5M Stake - TheMinerMag

Financial News

• Declining mining activity flashes ‘bullish signal’ for bitcoin price, VanEck says - The Block

• The AI Boom Is Making Real-Estate Investors Rich—and Exposing Them to Risk - WSJ

• Bitcoin mining in 2025: IREN claims the crown as Bitdeer’s stock trails the pack - CoinDesk

Feature

• Bitcoin miners will struggle with the allure of AI in 2026, experts say - DL News

Email us

The latest round of speculation around a major bitcoin mining crackdown in China appears to have been amplified by social media, with network data suggesting the impact may be far smaller than some early claims implied.

The narrative gained traction after Jack Jianping Kong, a former co-chair of Canaan who now runs chipmaker Nano Labs, posted on X on Dec. 13 that bitcoin mining activity in Xinjiang was facing renewed scrutiny. Two days later, Kong said the network hashrate had fallen by roughly 100 EH/s — about 8% day over day — and estimated that at least 400,000 miners had been shut down in China. The post quickly spread and was cited by several media outlets as evidence of a large-scale Xinjiang mining shutdown.

There was, in fact, a sharp hashrate dip. But a closer look at pool-level, real-time reported hashrate suggests the situation was more complex and likely conflated with unrelated events outside China.

Based on snapshots of reported pool hashrate, the steepest declines during the Monday drawdown came from North American–facing pools. Foundry USA alone saw its reported hashrate fall by roughly 180 EH/s within hours, while Luxor also recorded a sharp drop. Combined, the two pools accounted for about 200 EH/s of the total decline — a figure that closely coincides with power curtailments across parts of the U.S. over the weekend due to a cold snap.

Other large pools with Chinese origins — including Antpool, F2Pool, ViaBTC, SpiderPool and Binance Pool — collectively showed declines of around 100 EH/s. That number roughly matches the figure cited on X, but attributing the entire drop to Xinjiang would be a stretch. These pools operate globally (ViaBTC is also Russia-heavy, for instance), and some of their connected hashrate — including Bitmain-linked operations in the U.S. that connect to Antpool — could also have been affected by power curtailments outside China.

Just as important, the recovery profile does not support the idea of a whopping 100 EH/s shutdown. By Dec. 17, most major pools were reporting higher real-time hashrate levels, with the combined total sitting only about 20 EH/s below where it was before the Monday dip. That points to a largely temporary disruption rather than a sustained, region-specific shutdown.

Updated pool-level data through Dec. 18 further weakens the case for a broad, lasting shutdown tied to Xinjiang. Most major pools have now fully or nearly recovered to pre-dip hashrate levels. FoundryUSA and Luxor have rebounded above their Dec. 14 readings, reinforcing the view that U.S. power curtailments were a major driver of the initial drawdown.

Among China-origin pools, Antpool and f2pool remain down by only a few exahashes compared with earlier in the week — a gap that appears marginal in the context of normal operational fluctuations. SpiderPool stands out as the main exception. As of Dec. 18, its reported hashrate remains roughly 20 EH/s below levels seen before the dip, making it the only major pool showing a sustained decline.

That divergence may offer a more realistic signal of Xinjiang-related disruptions. SpiderPool’s continued weakness suggests it may have had a relatively higher concentration of miners operating in the region who were forced offline and have yet to return. Even so, the scale implied by the data points to low tens of exahashes rather than the roughly 100 EH/s cited in early social media posts.

With that being said, it is plausible that some miners in Xinjiang may have voluntarily powered down over the weekend to stay under the radar during inspections, then quietly came back online as scrutiny eased. If so, pool-level recovery would mask short-lived regional shutdowns that never showed up as a lasting network deficit.

The timing made the episode particularly noisy. U.S. cold-weather curtailments and Xinjiang inspections unfolded almost simultaneously, creating a perfect storm for over-attribution in social media narratives.

What does stand out more clearly is that enforcement activity in Xinjiang may be hitting non-bitcoin mining harder. Litecoin’s hashrate, for example, fell sharply on Monday, dropping from about 3.38 PH/s to 2.65 PH/s — a roughly 20% decline — and has yet to meaningfully recover. That suggests the resurgence of mining in the region has not been limited to bitcoin, and that other proof-of-work networks may be more exposed.

The bottom line is that there is little doubt some level of shutdown activity occurred in Xinjiang, but claims that as much as 100 EH/s of bitcoin hashrate disappeared there appear overstated when viewed through pool-level data. That said, this does not mean inspections in Xinjiang will not intensify in the near term.

Recent activity suggests that mining in remote parts of Xinjiang has been tolerated as long as operators kept a low profile. Public bragging about those operations on Chinese social media, however, appears to have drawn unwanted attention — and may have ultimately triggered a broader response.

Regulation News

• Arizona city unanimously rejects AI data center after residents’ outcry - Fox Business - Fox Business

• Bitcoin mining ban welcomed by power-depleted Russian regions - DLnews

Hardware and Infrastructure News

• Lucky solo bitcoin miner beats 1-in-82-year odds to win $285,000 block reward - The Block

• Sangha Energizes 20 MW Texas Solar Bitcoin Mine Amid Record-Low Hashprice Pressure - TheMinerMag

Corporate News

• Bitdeer Lifts November Bitcoin Output as Mining Sites Begin AI Data Center Conversions - TheMinerMag

• Canaan Inc. Renews Its US$30 Million Share Buyback Program - Link

• Weather Disruptions Delay CoreWeave, Core Scientific AI Data Center in Texas: Report - TheMinerMag

• Hut 8 Shares Jump 20% After $7B Fluidstack AI Data Center Lease - TheMinerMag

Feature

• We mapped the world’s hottest data centers - Rest of World

Email us