AI Is Rewriting the Economics of Bitcoin Mining
Bitcoin miners have traditionally competed over two scarce resources: efficient hardware and cheap electricity. AI is changing that equation by giving miners another way to monetize the power infrastructure they spent years assembling.
Summary:
- AI is creating a higher-value alternative use for miners’ existing power.
- AI-linked miners command substantially higher market valuations.
- Most contracted capacity still has to be converted into operating revenue.
- Capital efficiency will determine which AI pivots ultimately justify their premium.
CoinShares’ Q2 2026 mining report shows the shift becoming economically significant. Weak Bitcoin mining margins are encouraging listed operators to redirect sites toward AI and high-performance computing, while equity markets are assigning considerably higher valuations to companies that have already secured AI customers.
The emerging question is no longer simply whether a miner can produce Bitcoin profitably. It is whether mining remains the best use of each available megawatt.
Every Mining Megawatt Now Has an Opportunity Cost
CoinShares estimates that AI compute can currently generate around $1.5 million in annualized profit per MW, compared with approximately $500,000 for Bitcoin mining.
That threefold difference introduces an opportunity cost that barely existed when many large mining campuses were developed.
A miner controlling 100 MW is no longer choosing only between running current ASICs, upgrading them or temporarily curtailing operations. It can potentially redevelop that capacity for a workload offering substantially higher returns.
The calculation becomes particularly important when mining margins are weak. Listed miners’ weighted average ex-tax cash production cost reached approximately $75,500 per BTC in Q2, while Bitcoin ended June at $58,400. June hashprice also fell to a record monthly average of $27.70 per PH/s/day.
Some companies have responded by reducing their commitment to hashrate rather than waiting for mining economics to recover.
Core Scientific paid $41.9 million to cancel approximately 15 EH/s of next-generation mining hardware. Keel ceased Bitcoin mining in June, while Cipher does not plan additional mining capex. CoinShares estimates that at least 35 EH/s is scheduled to leave the listed-miner cohort as operators reposition infrastructure.
This is a significant change in incentives: miners are willing to sacrifice future Bitcoin production to preserve power for another computing market.
The Stock Market Has Already Chosen a Favorite
The operational transition is still developing, but investors are already assigning very different valuations to the two business models.
CoinShares calculates that miners with contracted AI or HPC capacity trade at an average 12.9x EV/NTM sales, compared with 3.7x for continuing miners without contracts. The more AI-focused group consisting of Hut 8, Cipher, TeraWulf and Core Scientific averages 19.4x.
That valuation gap matters because it changes corporate incentives.
If securing an AI tenant can materially increase the multiple investors place on a miner’s future revenue, management teams have a financial reason to pursue AI even before those projects become major contributors to earnings.
There is already evidence that the revenue transition can work. IREN generated $70.5 million from AI cloud operations in its latest quarter, overtaking its $66.7 million of mining revenue. HPC leases accounted for 71% of TeraWulf’s Q2 revenue.
But those examples do not mean every mining site has suddenly become an AI asset.
The AI Pivot Has Four Tests
Investors evaluating a miner’s AI strategy increasingly need to separate access to power from a functioning data-center business.
The most useful progression is:
- Energized capacity: How much power is actually secured and available? This establishes the infrastructure advantage but says little about AI revenue on its own.
- Contracted capacity: How many megawatts have customers agreed to take? Contracts provide evidence of demand, although capacity may still require substantial development.
- Billing capacity: How much infrastructure is already operating and producing customer revenue? This is where an AI project moves from backlog into financial results.
- Conversion economics: How much capital is required to move from the first stage to the third? High revenue per MW matters considerably less if development costs consume most of the incremental return.
The distinction is important because AI infrastructure is far more expensive than a conventional Bitcoin mine.
CoinShares estimates mining facilities can cost roughly $700,000 to $1 million per MW, while AI-grade infrastructure can require around $8 million to $15 million per MW. AI workloads demand substantially more sophisticated cooling, networking, redundancy and power reliability.
The relevant return is therefore not simply profit per MW. For shareholders, profit per dollar of conversion capital may ultimately matter more.
More Than 85% of Contracted Capacity Is Still Not Billing
This is where the current AI valuation premium faces its biggest test.
Companies covered by CoinShares have contracted more than 4 GW of AI/HPC capacity, but only approximately 550 MW is currently billing. In other words, less than 14% of the contracted total has reached the revenue-generating stage.
The remaining gap represents enormous potential, but also financing, construction and execution risk.
READ MORE: Bitcoin Is Off Its Peak. Crypto’s Wealth Machine Didn’t
Disclosed AI/HPC backlog exceeds $100 billion while current annualized revenue is around $1.1 billion. Those figures are not directly comparable accounting measures, but together they illustrate how much of the industry’s AI narrative depends on future delivery rather than present operations.
A signed agreement still requires infrastructure to be financed and built before it can contribute recurring revenue.
That creates a useful distinction between contract value and conversion velocity. A smaller AI portfolio entering service quickly could ultimately prove more valuable than a much larger announced backlog facing years of construction.
Miners Own Something AI Developers Cannot Quickly Build
The industry’s strongest advantage may ultimately be neither ASICs nor data centers. It is access to the electrical grid.
CoinShares identifies approximately 2,600 GW of projects waiting in the U.S. interconnection queue, while completed projects in 2025 faced a median wait exceeding five years between requesting interconnection and beginning operation. The report also identifies 225 data-center moratoriums or restrictions across 30 states, with 151 still active.
Many Bitcoin miners secured large power connections before the current AI infrastructure boom.
That gives suitable mining campuses a time advantage. A technology company can purchase computing equipment relatively quickly, but reproducing hundreds of megawatts of energized capacity in a constrained market can take years.
Still, not every mining site deserves a data-center valuation. Location, fiber connectivity, power quality, water availability, construction requirements and customer demand determine whether an existing grid connection can realistically support AI.
The market may therefore begin differentiating between mining megawatts and convertible megawatts.
Bitcoin Can Still Move the Hurdle Rate
There is one further complication: the economics of mining can change much faster than the economics of a long-term AI contract.
CoinShares notes that hashprice recovered toward $38 as Bitcoin rebounded toward $77,000, returning most operators above cash breakeven.
Another major BTC rally could narrow the profitability gap between the two workloads. Miners retaining capacity would participate immediately, while power committed under long-duration AI agreements may no longer be available.
AI therefore offers greater revenue visibility at the cost of some Bitcoin optionality.
That trade-off suggests the strongest operators may not necessarily be those abandoning mining fastest. Companies able to finance AI development selectively while retaining highly efficient mining capacity could preserve exposure to both economic cycles.
For investors, the next phase should make the distinction clearer. Announced gigawatts and headline backlog established the first AI premium. Billing megawatts, conversion capex and cash returns will determine whether companies deserve to keep it.
The information presented in this article is intended for informational purposes only and should not be interpreted as financial, investment, or trading advice. Coinspress.com does not promote or advocate for any particular investment strategy, asset, or cryptocurrency project. Cryptocurrency markets are highly volatile and unpredictable – always perform your own research and seek guidance from a qualified financial professional before making any investment decisions.
Alexander Stefanov
Reporter at CoinsPress
Alex is Editor-in-Chief of Coinspress and co-founder of Millennial Media Group, with nearly a decade of experience covering financial markets - crypto first, then everything else. It started in 2016 with Bitcoin. Like most people at the time, he didn't fully understand it - so he kept digging. Blockchain, tokenomics, the projects, the cycles. That curiosity never stopped, and eventually pulled him into traditional markets too: equities, commodities, macro. Not because he left crypto behind, but because you can't properly understand one without the other. What drives him is straightforward: he wants to know why something is happening, not just that it's happening. Most market coverage stops at the headline - price up, price down, here's a chart. Alex finds that kind of reporting actively unhelpful. If you walk away from an article without understanding the mechanism behind the move, what did you actually learn? He holds a degree in Tourism from New Bulgarian University - not the most obvious path into financial markets, but markets have a way of pulling in people who are simply too curious to stay out. He has authored over 200 in-depth analyses and more than 10,000 articles across crypto and traditional finance. He still thinks every day in markets teaches him something new. That's probably why he hasn't stopped.
