Mining companies heavily betting on AI: stock prices significantly outperform Bitcoin, but the turnaround battle is tough

Crypto mining companies accelerate transformation into AI infrastructure, with stock prices significantly outperforming Bitcoin but profits still under pressure, and funding gaps reaching hundreds of billions of dollars.
(Background: The New York Times exposed corruption among the White House crypto czar, David Sacks involved in insider trading earning hundreds of billions of dollars)
(Additional context: White-haired stock god Serenity calls out SIVE again! Secures an $8.2 million military order, Ka-band chip enters U.S. military satellite supply chain)

Table of Contents

Toggle

• Stock prices outperform Bitcoin, valuation divergence is obvious
• Funding pressure for AI transformation is enormous
• Survival battle upgrades across multiple hurdles

As crypto assets continue to weaken, crypto mining companies are facing increasingly severe survival pressures. In search of new growth trajectories, more and more miners are accelerating into the AI track. This transformation narrative has quickly gained favor in the capital markets, with many miners’ stock prices soaring significantly and even hitting new highs.

However, while AI business injects new growth imagination into miners, the huge capital expenditures behind it, continuous capital投入, and long return cycles are pushing miners into another round of capital consumption. Amidst the ongoing pressure on mining profitability, this high-stakes gamble on AI transformation tests the financial strength and execution capabilities of mining firms.

Mining companies are transforming into the dominant computational power providers of the AI era.

As Bitcoin mining profitability narrows, some miners are even incurring losses, while the explosive growth of AI has driven a sharp increase in global demand for data centers, electricity resources, and GPU computing power. More and more miners are accelerating their shift into AI infrastructure, seeking new growth curves.

For miners, this transformation has natural advantages. Over the long term, to meet large-scale mining demands, miners have accumulated abundant electricity resources, land reserves, substation access, and mature heat dissipation and cooling systems—key assets. Compared to data center operators starting from scratch, miners only need to upgrade existing facilities to quickly enter the AI infrastructure market, offering lower costs and shorter cycles to meet AI computing demands.

Stock prices outperform Bitcoin, valuation divergence is obvious

Since last year, miners’ pace of AI transformation has significantly accelerated. Some miners have decisively downplayed or even exited traditional mining operations, fully shifting to AI computing power and data center operations; others retain some mining machines but gradually reallocate resources and capital expenditure focus toward AI. Today, multiple miners have become important players in AI infrastructure construction.

In terms of timing, CoreWeave, Applied Digital, and Bitdeer started deploying AI computing and data center businesses as early as 2022-2023, representing early industry transformers; while Iris Energy, Terawulf, Hut 8, Riot Platforms, and Bitfarms began ramping up AI infrastructure investments in 2025, coinciding with the rapid expansion cycle of the AI industry.

In terms of stock performance, the market has given high recognition to miners’ AI transformation narratives. The 11 miners’ average increase since the beginning of the year has reached 75.97%, significantly outperforming Bitcoin’s performance during the same period, with most hitting new highs post-transformation. Notably, Bitfarms (129.62%), Hut 8 (131.87%), Terawulf (118.68%), and Riot Platforms (93.71%) stand out as beneficiaries of this AI infrastructure revaluation rally.

In terms of market capitalization, miners have shown clear segmentation. As a successful example of transformation, CoreWeave’s valuation has reached $62.85B, far surpassing other miners and becoming a new industry valuation benchmark; Iris Energy, Terawulf, Hut 8, Applied Digital, and Riot Platforms form a valuation tier of $10 billion to $20 billion; while companies like MARA Holdings, Core Scientific, Bitdeer, CleanSpark, and Bitfarms remain below $5 billion. This segmentation is driven not only by first-mover advantages but also by market differentiation in AI strategic execution, customer resources, and data center progress.

However, from a fundamental perspective, most miners are still in the heavy investment stage of AI transformation. Although many have reported revenue growth in recent quarters, overall profitability remains under pressure. On one hand, fluctuations in crypto asset portfolios drag on profits; on the other hand, AI data center construction requires enormous capital expenditure, with increasing investments in power capacity expansion, infrastructure, and GPU procurement, driving operational costs higher, and many miners still operating at a loss.

AI transformation faces enormous funding pressures

It’s noteworthy that despite generally underperforming results, the stock prices of related miners have surged significantly, indicating that the current market focus is not on short-term profitability but on the growth potential of miners as new-generation computing infrastructure operators.

The sluggish Bitcoin market is making the survival environment for miners even more challenging.

Capriole Investments data shows that as of June 18, the average cost of Bitcoin production is about $63,707, with electricity costs around $50,965, and miner profit margins at only 17.45%. Over the past 30 days, miner profit margins have contracted by 47.8%. Meanwhile, Luxor Hashrate Index data indicates that as of June 18, the daily return per TH/s has dropped to $0.032, a significant decline from $0.053 a year ago.

With mining revenues shrinking, many miners are forced to sell Bitcoin to maintain cash flow, further increasing survival pressure on small and medium-sized miners, and resources are rapidly consolidating toward leading players. Currently, Foundry USA, AntPool, and F2Pool together control 59% of the total network hash rate. In contrast, in 2022, the top three Bitcoin pools accounted for only 44% of the hash rate.

Although traditional mining is sluggish, explosive growth in AI data center demand is re-evaluating miners’ value. VanEck’s latest research report states that the most valuable assets for miners are not mining machines but electricity resources, substation access, land reserves, and data center infrastructure—precisely the core resources most scarce in the current AI industry. Since AI clients are willing to pay much higher electricity prices and rents than traditional mining, AI infrastructure is expected to become the main growth engine for miners over the next decade.

Survival battle upgrades across multiple hurdles

According to Bernstein’s research report, over $90 billion in AI infrastructure collaborations have been announced by large cloud providers, AI cloud service providers, and chip companies, involving about 3.7 GW of power capacity. Currently, competition for electricity resources has become the core of AI infrastructure competition, with Bitcoin miners planning over 27 GW of power capacity. In some U.S. regions, new 1 GW power connections can take up to 50 months, making existing mines critical nodes for AI data center expansion.

However, AI transformation is far from an easy path. VanEck emphasizes that the market is still in the early stages of AI transformation, with enterprise valuations primarily based on the total energized power (Gross Energized Power). Miners with signed AI leases generally command higher valuation premiums, while projects still in planning stages struggle to gain market recognition. Future industry valuation logic will gradually shift from “power capacity” to “project delivery capability,” ultimately returning to core metrics like cash flow, capital return, and tenant quality. Currently, only about 25% of signed capacity has been delivered; whether AI data centers can be completed on time and within budget will be key to determining enterprise valuation.

VanEck also highlights that AI tenant quality will directly impact miners’ valuation levels. Large cloud providers like Microsoft, Amazon, and Google can bring more stable cash flows and lower financing costs, while smaller GPU cloud service providers face higher operational risks and capital costs.

The massive capital投入 required for transformation is also testing miners’ financial strength. VanEck estimates that the capital expenditure for miners shifting to AI infrastructure remains huge, with a short-term financing gap of about $50 billion and long-term capital needs possibly reaching $221 billion.

Under this enormous financial pressure, many miners have already begun raising funds through various means. For example, Iris Energy, Terawulf, Bitfarms, and CleanSpark have issued convertible bonds and other instruments to attract investors with lower coupon rates and future equity conversion options; while Core Scientific, Terawulf, MARA, Bitdeer, and Riot Platforms have sold or liquidated some Bitcoin reserves to fund AI transformation.

Additionally, many miners are locking in future income through long-term AI or high-performance computing (HPC) contracts, securing project financing and reducing operational risks. For instance, CoreWeave has a $6 billion AI cloud service partnership with Jane Street; IREN secured a $9.7 billion AI cloud computing contract with Microsoft; Hut 8 signed a data center leasing agreement worth $9.8 billion; Bitdeer partnered with Norway’s DCI to build the country’s largest AI data center project.

For miners, AI currently offers a development path with far more imagination than traditional mining. However, this transformation is not simply switching from mining to selling computing power; fundamentally, it is a long-term competition revolving around capital, resources, and execution capabilities.