Nvidia just printed $81.62 billion in quarterly revenue, an 85% jump from a year ago, and Bitcoin miners with AI ambitions caught a bid on the news. Core Scientific and Cipher Mining both ticked higher in after-hours trading Wednesday, while IREN gave back an early gain to finish down about 1%.
The earnings beat itself was substantial. Wall Street had penciled in $78.9 billion; Nvidia delivered nearly $3 billion more. Adjusted earnings landed at $1.87 per share against expectations of $1.76. And guidance for the current quarter, at roughly $91 billion, came in above consensus again. Yet Nvidia shares slipped 1.5% in extended trading, a reminder that even blowout numbers get discounted when a stock is already priced for perfection.
For the subset of Bitcoin miners that have been quietly repositioning themselves as AI infrastructure plays, the details inside Nvidia’s report mattered more than the headline print. Data Center revenue now accounts for more than 90% of the chipmaker’s total, a concentration so pronounced that Nvidia reorganized its reporting into two segments: Data Center and Edge Computing. That structural shift underscores where the dollars are flowing.
Hyperscalers Drive Half of Data Center Demand
Nvidia’s CFO Colette Kress broke down the $75 billion Data Center haul on the earnings call. Hyperscalers, the cloud giants like Amazon Web Services, Microsoft Azure, and Google Cloud, generated roughly $38 billion of that figure, up 12% from the prior quarter. The remaining $37 billion came from a segment Nvidia now calls ACIE, covering AI cloud providers, industrial customers, and enterprise markets.
The ACIE bucket is where Bitcoin miners with data center ambitions fit. AI cloud revenue inside that segment more than tripled year-over-year, Kress said, as Nvidia helped rapidly expand computing capacity across more than 80 data centers with individual capacities exceeding 10 megawatts. That 10-megawatt threshold is notable: many Bitcoin mining facilities operate in the tens or hundreds of megawatts, meaning their existing power infrastructure can be repurposed for AI workloads without starting from scratch.
CEO Jensen Huang framed the spending surge in grand terms. “The buildout of AI factories, the largest infrastructure expansion in human history, is accelerating at extraordinary speed,” he said in a statement. “Agentic AI has arrived, doing productive work, generating real value and scaling rapidly across companies and industries.”
The company also disclosed that its outlook excludes any Data Center compute revenue from China, where U.S. export restrictions have curtailed sales of advanced AI chips. That geographic carve-out means the $91 billion guidance is essentially a Western-market number, which could actually benefit domestic infrastructure providers competing for the same hyperscaler contracts.
Bitcoin Miners as Data Center Landlords
The thesis connecting Bitcoin miners to Nvidia’s fortunes rests on a simple observation: miners have spent years acquiring what AI companies now desperately need. Cheap land in energy-abundant regions. Direct relationships with utilities. Substations and transformers already permitted and built. Cooling systems designed for heat-intensive computing. And, critically, power purchase agreements that lock in electricity at rates most enterprise data centers cannot match.
Core Scientific, which emerged from bankruptcy in early 2024, has been the most aggressive in pivoting toward AI hosting. The company signed a 12-year, 200-megawatt deal with CoreWeave, an AI cloud provider backed by Nvidia itself, in mid-2024. That single contract, worth an estimated $3.5 billion over its life, repositioned Core Scientific from a pure Bitcoin miner into a hybrid infrastructure play. Its stock rose modestly in after-hours trading Wednesday.
Cipher Mining has pursued a similar strategy, though with a smaller footprint. The company has emphasized its Texas facilities’ proximity to ERCOT’s grid, where curtailment programs allow miners to monetize flexibility during peak demand. That same flexibility makes the sites attractive for AI workloads that can tolerate variable power allocation.
IREN, formerly known as Iris Energy, has positioned itself as a renewable-powered computing company rather than a Bitcoin miner per se. Its facilities in Texas and Canada draw heavily from solar and hydro sources, a selling point as AI hyperscalers face pressure to meet sustainability targets. The stock’s 1% decline Wednesday, after an initial pop, may reflect profit-taking rather than any fundamental concern. IREN has been one of the better performers in the mining sector this year.
Investors tracking this space should keep an eye on our derivatives dashboard for funding rate signals that often precede sharp moves in mining equities, which tend to trade as leveraged bets on Bitcoin itself during risk-on periods.
The Revenue Math for Diversified Miners
Here is where the numbers get interesting. Bitcoin mining revenue depends on two variables: the price of Bitcoin and the network’s mining difficulty. Both are outside any individual miner’s control. AI hosting revenue, by contrast, locks in fixed-rate contracts denominated in dollars, often with inflation escalators and multi-year terms.
Consider a simplified comparison. A 100-megawatt Bitcoin mining facility running current-generation ASICs might generate $15 million to $20 million per month in Bitcoin at today’s prices and difficulty, assuming competitive power costs around $0.03 per kilowatt-hour. The same 100 megawatts allocated to AI hosting under a CoreWeave-style contract might yield $8 million to $12 million per month in guaranteed revenue, regardless of Bitcoin’s price action.
The Bitcoin operation has higher upside if prices rally, but it also carries wipeout risk if difficulty spikes or prices crash. The AI contract offers lower peak returns but near-certain cash flow that banks will lend against. For miners that spent 2022 and 2023 facing liquidity crises and margin calls, the appeal of predictable revenue is not abstract. Core Scientific’s bankruptcy, after all, stemmed from exactly the kind of cash-flow mismatch that AI hosting contracts are designed to prevent.
Nvidia’s disclosure that AI cloud revenue more than tripled year-over-year suggests the demand curve for this infrastructure is still steepening, not flattening. If hyperscalers continue expanding capacity at the pace Kress described, miners with ready-to-go megawatts could find themselves in a seller’s market for the next several years.
For broader context on how corporate treasury strategies interact with Bitcoin price volatility, our recent coverage of Strategy’s pause on weekly BTC purchases highlights the accounting pressures that can affect even the largest institutional holders.
Nvidia’s Stock Dip and What It Signals
Nvidia shares falling 1.5% on a quarter this strong might seem counterintuitive, but it reflects a market already priced for excellence. The stock has more than tripled over the past two years, and its valuation assumes continued growth at rates that would be extraordinary for any company, let alone one already generating $80 billion per quarter.
The China exclusion in guidance is one concern. U.S. export restrictions have effectively locked Nvidia out of the world’s second-largest AI market, and there is no sign those restrictions will ease. Competitors like AMD and Intel are angling for the same hyperscaler contracts domestically, while Chinese chipmakers like Huawei are developing alternatives for their home market.
Another concern is the sustainability of capital expenditure at current levels. AI infrastructure spending has been running hot for three years now, and at some point CFOs will demand returns on those investments. If the payback period for AI computing stretches longer than expected, capex budgets could tighten, and Nvidia’s growth would slow accordingly.
For Bitcoin miners, though, the relevant question is not whether Nvidia can sustain 85% revenue growth indefinitely. It is whether AI infrastructure demand remains strong enough to absorb the capacity miners can supply. On that score, Wednesday’s report was unambiguously positive. Eighty data centers with 10-plus megawatt capacities brought online in a single quarter. AI cloud revenue tripling. Hyperscaler spend up 12% sequentially. These are not the numbers of a market about to saturate.
The $80 billion stock buyback and dividend increase (from $0.01 to $0.25 per share, a 25x jump) signal that Nvidia’s board sees more cash generation ahead than it needs for reinvestment. That is a confidence indicator, even if the stock market shrugged.
Calculating the Miner Upside
Pulling the pieces together: Nvidia’s Data Center segment generated $75 billion in the quarter. If ACIE (the non-hyperscaler portion) represented roughly half, that is $37 billion flowing to AI cloud providers, enterprise customers, and industrial users. Bitcoin miners with AI hosting capacity sit somewhere in that ACIE bucket, competing for contracts alongside dedicated data center REITs and cloud startups.
The total addressable market for AI infrastructure spending is notoriously hard to pin down, but Nvidia’s own numbers imply annualized Data Center revenue north of $300 billion if current trends hold. Even capturing a fraction of a percent of that figure would represent material diversification for public Bitcoin miners, most of which have market capitalizations measured in single-digit billions.
Core Scientific’s market cap sits around $3 billion. Cipher Mining is closer to $1 billion. IREN is in the $2 billion range. For companies of this size, a $3.5 billion contract like Core Scientific’s CoreWeave deal is transformative. It does not make Bitcoin mining irrelevant, but it does provide a revenue floor that changes the risk profile entirely.
Investors weighing exposure to these names should monitor our Bitcoin treasury tracker to see how corporate Bitcoin holdings compare to AI infrastructure valuations across the sector.
The Catch: Competition Is Coming
None of this means Bitcoin miners are guaranteed winners in the AI infrastructure buildout. The same dynamics that make their existing facilities valuable also attract well-capitalized entrants. Digital Realty, Equinix, and other data center REITs have been building out AI-ready capacity for years. Amazon, Google, and Microsoft are all constructing proprietary data centers at scale. And a wave of AI-focused startups, many with venture backing and no legacy operations to maintain, are competing for the same hyperscaler contracts.
What miners have is speed. Permitting a new data center from scratch can take two to three years in most U.S. jurisdictions. Repurposing an existing mining facility, with its already-permitted power connections and cooling infrastructure, can be done in months. That time advantage matters in a market where AI demand is outrunning supply, but it is not permanent. Once the new builds come online, miners will face pricing pressure from purpose-built competitors with newer equipment and more efficient layouts.
The smart play for miners is to monetize the current supply-demand imbalance while it lasts, locking in long-term contracts that provide revenue certainty even after the infrastructure gap closes. Core Scientific’s 12-year CoreWeave deal is a template. Miners that wait too long to pivot may find the window has closed.
What Happens If Bitcoin Rallies
There is a scenario where the AI pivot looks premature in hindsight. If Bitcoin breaks convincingly above $80,000 and sustains a new trading range, the opportunity cost of converting mining capacity to AI hosting becomes substantial. A 100-megawatt facility dedicated to Bitcoin mining could generate significantly more revenue than the same capacity under a fixed-rate AI contract, assuming favorable difficulty adjustments.
Our earlier coverage of Bitcoin hitting $77,400 on big tech earnings noted that $80,000 resistance remains intact, with ETF outflows and energy price volatility keeping traders cautious. If that resistance breaks, miners who over-committed to AI hosting might face uncomfortable questions from shareholders about leaving money on the table.
The counterargument is that Bitcoin’s volatility is precisely why diversification makes sense. A miner with 50% of capacity dedicated to AI hosting and 50% to Bitcoin captures upside from price rallies while maintaining cash flow stability during drawdowns. The optimal allocation depends on each company’s risk tolerance, debt load, and shareholder base, but the days of pure-play Bitcoin mining as the default strategy may be ending.
Bottom Line
Nvidia’s Q1 2026 report confirmed what Bitcoin miners pivoting to AI have been betting on: infrastructure demand for artificial intelligence is not slowing down. Data Center revenue hit $75 billion in a single quarter, hyperscaler spending rose 12% sequentially, and the company guided to $91 billion for the current period. For miners with megawatts to spare and power agreements in place, those numbers translate into contract opportunities that did not exist three years ago.
Core Scientific, Cipher Mining, and IREN all carry exposure to this theme, though their individual strategies and risk profiles differ. The modest after-hours moves suggest the market is not yet pricing in a full rerating of these names as AI infrastructure plays. Whether that rerating comes depends on contract wins, execution, and whether Bitcoin itself cooperates by staying range-bound long enough for the diversification thesis to prove out.
