Public bitcoin miners are increasingly repositioning themselves as AI infrastructure companies, and that shift is opening a new debate over what it means for the network they once helped secure. A discussion on X on Friday, led by Capriole Investments founder Charles Edwards and joined by on-chain analyst James “Checkmate” Check and Blockstream CEO Adam Back, framed the issue in stark terms: a rational business pivot for miners may still carry uncomfortable implications for Bitcoin’s long-term security model.
Public Miners Chase AI as Bitcoin Share Shrinks
Edwards argued that the trend is broader and more aggressive than many in the market had appreciated. In a post that drew wide attention, he wrote, “This is a list of all the major public Bitcoin miners. ALL have made statements to pivot to AI. ALL are targeting major shares of revenue from AI from here, not Bitcoin.” He added that, based on company guidance, “current Bitcoin revenue is expected to drop from 90% to just 30% in the next 2-3 years.”
That revenue mix matters because equity markets have already rewarded the shift. Edwards said public miners targeting more than 80% of revenue from AI had seen their stocks rise by more than 500% on average, while those targeting less than 60% AI revenue delivered roughly one-tenth of that growth, with many posting negative two-year returns. His conclusion was blunt: “The market has been voting with its feet. Now the miners are voting with their feet.”
The concern, in his view, is not simply branding or treasury strategy, but capital allocation. Edwards said many miners are “not even planning to upgrade or renew Bitcoin mining hardware at all,” instead running existing fleets through the end of their usable life and reinvesting primarily into AI infrastructure. If that pattern holds, the implication is that listed miners may remain adjacent to Bitcoin while steadily reducing their direct contribution to hashrate growth.
For some industry participants, however, that shift looks less like abandonment than the natural evolution of a difficult business. Checkmate described mining as “a business destined for bankruptcy,” arguing that high equipment turnover and periodic operator failure are part of Bitcoin’s intended design through difficulty adjustment. In that framework, machines and facilities do not disappear; they simply change hands at a lower cost basis when weaker operators wash out.
He also made the case that AI and bitcoin mining are complementary workloads for power buyers. “They buy power, and they compute. AI is baseload always on, Bitcoin is intermittent, can balance loads,” he wrote, calling the pivot “a very rational decision” and suggesting it was “very much like an expected result.” From that perspective, miners are not betraying Bitcoin so much as adapting to the economics of energy and compute.
Edwards did not dispute the business logic. “If I was a miner, I’d probably have done the same,” he wrote. But he drew a distinction between what is rational for a miner and what is healthy for the network, especially if listed operators move from bitcoin-derived revenue to single-digit exposure over a relatively short period. In his framing, the issue is less whether miners should diversify, and more whether Bitcoin can ignore the cumulative effect if nearly all major public players do it at once.
Security Fears Meet the Case for Diversification
The central disagreement turned on whether lower hashrate necessarily means weaker security in a meaningful economic sense. Adam Back suggested the opposite may be true if unprofitable capacity exits and surviving miners earn better margins. “Think you could make the case that 90% Hashrate with a healthy profit margin is better than 100% with the less efficient struggling financially,” he wrote, later adding that if hashrate falls, mining margins rise through arbitrage until equilibrium is restored.
Back’s point was that miner profitability can create its own stabilizing loop. “Higher profit margin adds to positive reflexivity – miners sell less Bitcoin to cover power, and as price rises.” That argument treats hashrate not as a raw number to maximize at any cost, but as an output of competitive equilibrium. If AI absorbs weaker or higher-cost operators while leaving a more profitable core set of miners behind, the end result may be more sustainable than it first appears.
Edwards pushed back by pointing to the recent trajectory of network compute. “HR is already down 25% in last 6 months, and that’s while 90% of the miner revenue is still from BTC,” he wrote, arguing that the concern is not a hypothetical 10% adjustment but a larger decline occurring before the projected revenue shift has fully played out. He tied that to a broader warning that Bitcoin’s “backbone of its security is leaving the industry” just as, in his words, quantum computing is advancing and may require protocol changes in the years ahead.
The exchange also exposed a semantic fault line. Checkmate asked whether miners were truly “pivoting, or diversifying,” saying “the language used here, is doing all the lifting in this debate.” For Edwards, the distinction depended on scale. “When you go from 100% revenue being bitcoin mining to 10% or less being that’s not diversifying, that is pivoting,” he replied, adding that some firms are explicitly planning only to exhaust existing bitcoin fleets rather than reinvest in new rigs.
That leaves the market with two competing readings of the same development. One is that miners are sensibly monetizing access to power and compute in the highest-return segment available, with Bitcoin mining remaining a flexible load and residual business line. The other is that, even if each firm is acting rationally, the aggregate result could be a structural reduction in fresh capital committed to bitcoin mining hardware, and by extension a different security trajectory than the sector has assumed.
The immediate takeaway is not that Bitcoin mining disappears, but that its public-market champions may no longer be pure-play expressions of network growth. If the AI revenue targets cited by Edwards prove directionally right, investors and protocol stakeholders will have to weigh a harder question: whether Bitcoin’s security budget remains robust when the companies once most exposed to it increasingly see their future somewhere else.