Bitcoin Miners Are Becoming an AI Utility Whether the Industry Likes It or Not
On a single day in early May 2026, three data points landed in financial terminals that should have been read together. Cointelegraph reported that Bitcoin miners are on track to earn more from artificial intelligence services than from Bitcoin mining by the end of the year. Goldman Sachs announced its best quarterly performance in five years. And KKR committed $10 billion to build AI-dedicated power plants and data centers. Individually, each item is a market brief. Taken together, they sketch a structural realignment: the energy infrastructure that Bitcoin miners spent years building is becoming the most sought-after asset class in the AI economy.
The implications are not obvious at first glance. Bitcoin mining has always been portrayed as a speculative, environmentally controversial activity — a calculation game played with cheap electricity and specialised hardware. That framing is not wrong, but it is incomplete. What Bitcoin miners actually built, at scale, was a distributed network of industrial-grade power procurement, custom cooling systems, land rights near substations, and operational expertise in running computation at megawatt scale. Those capabilities did not disappear when the Bitcoin price stabilised or when ESG pressures mounted. They sat dormant, waiting for a buyer.
AI is that buyer.
The Halving That Forced the Reckoning
Bitcoin's block reward halving — which cuts miner revenues in half roughly every four years — has been the industry's central stress test. The most recent halving reduced per-unit earnings forProof-of-Work miners, compressing margins for operators without access to very cheap power. The standard response has been either to relocate operations to lower-cost jurisdictions or to diversify into adjacent revenue streams. AI hosting contracts represent a structural diversification, not a tactical hedge.
The distinction matters. A tactical hedge is what happens when a mining firm rents spare capacity to a crypto cloud service during a quiet market. A structural diversification is what happens when a mining operation signs a multi-year contract to provide colocation for AI inference clusters, using the same power infrastructure it already owns. In the former case, the revenue stream is cyclical and dependent on crypto market conditions. In the latter, the revenue stream is anchored to corporate AI demand — which is growing on a trajectory that no cryptocurrency market has ever approached.
The shift in revenue composition — AI earnings potentially surpassing Bitcoin earnings by year-end — reflects this structural change at the operational level. It is not a narrative rebranding. The hardware is the same; the customer is different.
KKR's $10 Billion Bet on the Power Constraint
KKR's commitment to build AI-dedicated power plants and data centers is the clearest signal that institutional capital has identified the bottleneck in AI expansion. Training and running large language models requires sustained, massive electricity supply. The grid capacity to deliver that supply, particularly at the locations where AI companies want to operate — near talent clusters, coastal hubs, renewable-rich regions — is not keeping pace with demand.
This is where Bitcoin miners become relevant to the conversation in a way that is difficult to replicate from scratch. A greenfield data center requires securing power agreements, navigating permitting, contracting construction, and commissioning cooling — a process that can take years. A Bitcoin mining facility that pivots to AI colocation has solved most of those problems already. The site has power rights. The electrical infrastructure is rated for continuous high draw. The cooling systems are designed for dense computational loads.
KKR is not the only firm that has worked this out. Several mining operators have announced or are negotiating AI hosting agreements with technology companies, treating their existing facilities as immediately available infrastructure rather than stranded assets. The energy transition that Bitcoin miners were supposed to undergo — away from carbon-intensiveProof-of-Work — is instead being redirected into a different industrial transition: from cryptocurrency to AI compute.
Who Controls the Power, Controls the Model
The AI industry's power problem is structural, not cyclical. Data centers currently under construction or in planning stages are competing for grid connections that may not exist for three to five years in many high-demand markets. The capital being deployed by firms like KKR reflects a view that owning the power generation and delivery infrastructure is as strategically valuable as owning the models themselves.
This is the realignment that the three May data points reveal. Goldman Sachs's strong quarterly performance partly reflects a broader environment in which financial institutions are profiting from the capital flows into AI infrastructure. KKR's $10 billion commitment signals that private equity has reached the same conclusion. And the Bitcoin miners' pivot toward AI earnings is the industrial consequence of that conclusion playing out in practice.
The workers, communities, and regional economies that anchored themselves around Bitcoin mining operations may find that their existing infrastructure is suddenly more valuable than it was six months ago — not because the Bitcoin network has changed, but because a new class of buyer has arrived with entirely different priorities.
What Remains Uncertain
The pace of this transition is not uniform. Not all Bitcoin mining operations are positioned to pivot toward AI colocation. The sites best suited for the transition — those with access to reliable, high-capacity power; proximity to population centres or fibre interconnects; and operational expertise in running dense computational loads — represent a minority of the global mining estate. There is also the question of whether large AI companies will prefer purpose-built facilities over converted mining sites, which may carry legacy constraints in layout and power density.
The sources do not indicate which specific mining operators have secured AI contracts, at what scale, or on what financial terms. What the pattern suggests is that the market has moved ahead of the reporting: institutional capital is already pricing the transition, and the revenue shift that miners are experiencing is the first publicly visible confirmation of a structural repositioning that has been underway for at least a year.
The question for investors, energy planners, and policymakers is no longer whether Bitcoin mining infrastructure will be absorbed into the AI economy. It is whether that absorption happens fast enough to relieve the power constraints that are currently limiting AI expansion — or whether the demand for compute outpaces the speed at which even converted mining facilities can be brought online.
Wire provenance
This editorial synthesis draws on the following public wire/social posts:
- https://t.me/cointelegraph
- https://t.me/cointelegraph
- https://t.me/cointelegraph