Bitcoin Miners Shift to AI: Risks and Trade-offs

Article by: Prathik Desai

Compiled by: Luffy, Foresight News

Over the past year, every major publicly traded Bitcoin miner has faced a survival crisis. Reviewing their financial reports reveals that they are shedding their identity as "miners."

Today, they have rebranded and repositioned themselves as "energy infrastructure platforms," "vertically integrated AI cloud service providers," or "digital infrastructure companies built on power, land, and computing power."

Behind this rebranding is a shift in revenue structure, with the company’s market capitalization now exceeding its valuation during its time as a pure Bitcoin mining company.

On the surface, the story seems wonderful: struggling miners seize the opportunity and embrace the surge in AI inference computing demand, finding a path to self-rescue. But if we take a broader view, they are also giving up something crucial.

This article will explain why miners have made a complete 180-degree turn, fully exiting Bitcoin mining, and how this may eventually backfire on them.

In April 2024, the Bitcoin block reward halved, forcing miners to choose: either hold onto mined Bitcoin and wait for higher prices, or upgrade their hardware to shift toward high-performance computing (HPC) and diversify their business. From under $70,000 in October 2024, Bitcoin surged to $124,000 by October 2025, prompting some miners to continue mining and hold onto their Bitcoin output.

However, the market liquidation event on October 10 caused $19 billion in market capitalization to vanish within 24 hours, triggering a downturn that has yet to fully recover.

The current price of Bitcoin is nearly half its peak from October last year, at approximately $63,000. At this level, selling Bitcoin after mining yields almost no profit. Bitcoin mining costs are influenced by electricity prices and overall network difficulty. The combined mining cost rose from around $40,000 in February 2024, reaching a range of $90,000 to $110,000 between October 2025 and July 2026; by August this year, the combined Bitcoin mining cost hit a new high, surging to $140,000.

Source: MacroMicro

At the same time that the Bitcoin mining environment continued to deteriorate, massive amounts of capital flooded into the AI inference sector.

Bitcoin miners use ASIC chips to run complex algorithms in order to earn block rewards. This hardware infrastructure shares highly similar underlying requirements with those needed for high-performance computing and AI inference. This allows miners to transition into AI inference without starting from scratch, significantly lowering the barrier to entry. For miners seeking profitable alternatives, shifting to AI appears natural.

Almost the entire listed mining sector has collectively changed course. The shifts in revenue structure over the past few quarters clearly reflect the scale of this transformation.

Taking Core Scientific as an example: In the second quarter of 2025, revenue from data center hosting services for AI and high-performance computing amounted to just $10.6 million; during the same period, revenue from mining and selling cryptocurrency reached $62.4 million. Twelve months later, their positions were almost entirely reversed. In the second quarter of 2026, hosting revenue surged to $136.7 million, while mining income declined by 65% to $21.5 million.

The custody business accounted for 83% of the company's total revenue, up from 14% last year.

TeraWulf’s revenue from high-performance computing rentals now accounts for 71% of total revenue. This business generated no income in the same period last year but reached $32 million in Q2 2026; during the same period, cryptocurrency mining revenue declined by approximately 75%, dropping to just $13 million.

Many miners are still converting their existing mining facilities and hardware to support AI and high-performance computing workloads.

The shift from Bitcoin mining to AI computing power has turned into a collective rush. In addition to reallocating existing capacity to AI, former Bitcoin miners are aggressively securing electricity and land, signing massive demand contracts. The total value of already finalized contracts has reached tens of billions of dollars.

Core Scientific has committed up to 2.5 GW of available computing capacity to AMD and CoreWeave, with a total potential revenue exceeding $24 billion over the contract period. Hut 8 has secured 949 MW of contracted computing power with a contract base value of $26.6 billion, along with $7.5 billion in new project financing.

TeraWulf has signed a 20-year lease agreement worth approximately $19 billion and acquired a gigawatt-scale facility in Kentucky to meet computing demand. Over the past six months, Riot Platforms has completed multiple leases totaling 241 MW, corresponding to contracts valued at approximately $10 billion. Just last week, IREN delivered its first AI cloud deployment nodes to Microsoft under a five-year contract worth $9.7 billion; it has also partnered with NVIDIA for 5 GW of capacity and is continuously acquiring power sites in Texas and Spain to support computing demand.

This series of large-scale contracts has ignited market enthusiasm. Compared to miners' past business models, signing 20-year leases with top cloud providers means more stable cash flow. After a full year of continuous mining losses, locking in contracts for up to fifteen years appears to be a solid and realistic move.

But many market commentators have underestimated the risks involved.

Although mining is arduous, Bitcoin mining possesses a characteristic that AI托管 businesses lack: a self-repairing mechanism. When mining becomes unprofitable, miners shut down their machines; when enough miners exit, the Bitcoin network’s difficulty automatically adjusts downward. For the remaining miners, the same equipment and electricity costs yield higher returns.

After a large number of miners shifted to the AI sector, Bitcoin’s network difficulty has dropped from its 2025 peak of approximately 156 trillion to 127.5 trillion. Each company that exits mining increases the mining rewards for those who remain.

However, the AI computing power industry lacks this automatic adjustment mechanism. If a large number of participants enter, leading to an oversupply of computing power, prices will be continuously pushed lower.

Although overall demand for AI computing power is rising, prices are likely to fall rather than rise—seemingly contradicting basic economic principles. In fact, multiple AI companies have already begun lowering their service prices. In the industry’s early stages, scarce hardware and supporting resources allowed providers to charge high premiums; today, major cloud providers worldwide are competing for megawatt-scale power resources, and this scarcity has attracted a surge of expansion from Bitcoin miners and data center operators. Once new supply fills the demand gap, computing power prices will drop rapidly.

When prices drop, the 15- to 20-year long-term leases signed by miners at today’s high prices will later appear excessively costly.

Another challenging aspect of this transition is that the path is largely one-way: converting a mining facility into an AI computing park is straightforward, as the power infrastructure is already in place. But once you replace ASIC miners with H100 GPUs and sign a 20-year lease, you’ll be locked into that contract for the next two decades, making it nearly impossible to revert to mining.

When the next market cycle reverses and Bitcoin prices rise above the $60,000 low, combined with a decline in overall network difficulty, mining will once again become profitable. By then, former miners who have fully transitioned away from mining will be bound by long-term lease contracts and forced to watch opportunities slip away. Even worse, if the actual returns from AI computing power fall short of the optimistic projections made at the time of signing, they will face a double blow.

So, shouldn't miners just stick to waiting for the mining cycle to turn around instead of pivoting to AI? I don’t think so. For most businesses, transformation is a survival imperative driven by necessity. With total mining costs reaching $140,000 and the coin price at only $63,000, continuing to mine is not commercially viable—AI is the only lifeline available right now.

Yet, the will to survive drives companies to make extreme choices. The risk lies precisely with those who have gone all-in on AI: under pressure to survive, they completely abandon Bitcoin mining, destroy or fully replace their ASIC miners, and lock into long-term contracts in a nascent industry that will still experience price corrections.

However, not all miners have opted for a complete 180-degree shift; some companies have taken a more cautious approach.

Marathon Digital, the publicly traded mining company with the largest Bitcoin treasury, adopted a cautious approach. In the second quarter of 2026, it sold 30% of its Bitcoin holdings to reduce debt, but instead of abandoning its Bitcoin mining equipment, it chose an alternative path to fund its artificial intelligence expansion.

MARA secured $150 million in funding through a 2026 credit facility, using its remaining Bitcoin as collateral, to finance the expansion of its AI business. As of June 30, it still held over 35,000 Bitcoin on its balance sheet while maintaining a mining operation with flexible load capacity.

Hut 8 has not abandoned mining either. It spun off its Bitcoin mining operations into a new subsidiary, American Bitcoin Corp., which is wholly owned by the parent company, while the parent company shifts its focus to high-performance computing and AI infrastructure.

By maintaining their mining rigs operational and holding onto their Bitcoin inventory, these companies retain a choice that fully committed transitioners have abandoned. When mining becomes profitable again, they can simply switch their power load back to mining; companies deeply locked into long-term AI contracts, however, can only watch from the sidelines.

Treating the old mining business as a burden and betting everything on AI computing is essentially wagering on an unproven future. However, historical patterns are easy to anticipate: with each emerging technology—chips, computers, smartphones—prices have consistently fallen as the industry matures, and signs of price softening are already emerging in the AI computing sector.

Companies like Marathon and Hut8 haven’t missed out on the AI boom. Instead of liquidating everything to purchase GPUs, they’ve maintained a portion of their mining capacity and Bitcoin holdings, treating AI as a diversification strategy rather than betting all their capital on an immature market still subject to price revaluation. As long as ASIC miners remain operational, they can shift their load back to Bitcoin mining when mining profits become more attractive and the AI computing premium fades.

From a broader perspective, these companies are essentially power operators, with their hardware consisting of ASIC miners or H100 GPUs. Only those companies that can flexibly allocate power resources between mining and AI computing based on which business yields higher returns will avoid being entirely at the mercy of market cycles and narratives.