Hundreds of billions of dollars in financing, Bitcoin mining companies collectively shifting to AI

Author | Wu Talks Blockchain

TL;DR:

Mining companies collectively transforming into AI infrastructure: North American Bitcoin mining publicly traded companies are undergoing a dramatic identity shift, from highly volatile, cyclical "mining firms" deeply tied to Bitcoin prices, to repositioning as energy infrastructure operators and AI data center platforms, securing billions of dollars in project financing and long-term contracts.

Core moat is large-scale power and rapid deployment capabilities: The fundamental reason AI cloud service providers choose to partner with mining companies is that miners possess the most scarce resources in the AI era — ready access to power grids, mature substation and transmission infrastructure, and fast deployment capabilities, significantly shortening GPU cluster deployment cycles.

Leading companies lock in massive long-term agreements: Typical miners like Core Scientific, Hut 8, Iris Energy, and TeraWulf have deeply integrated with top AI clients, signing long-term IT capacity leasing agreements totaling hundreds of MW. Some companies’ long-term contract base value reaches tens of billions of dollars, with higher renewal potential.

Evolving financing structures drive valuation logic shifts: Miners are gradually adopting project-level debt financing, triple-net leases, and take-or-pay clauses. This shift makes their revenue models more akin to traditional data center REITs, prompting capital markets to reprice them from commodity cycle companies to stable cash flow infrastructure assets.

Transformation faces high capital and execution risks: While embracing infrastructure, miners also face high transformation capital expenditures (ranging from millions to tens of millions USD per MW), high customer concentration, and the challenge of shifting from pure hash rate deployment to AI tech operations. Current market valuations are based on future successful delivery; if execution falls short, valuation logic may reverse.

Industry background: Halving pressure and AI infrastructure demand emerge simultaneously

By 2026, North American Bitcoin miners listed companies are experiencing a significant identity transformation.

Once viewed by capital markets as highly volatile, cyclical, Bitcoin price-dependent "mining companies," they now frequently appear in discussions about AI data centers, power infrastructure, and energy asset revaluation. From long-term AI hosting contracts, billions in financing, to ultra-long-term leasing agreements, a group of traditional miners are attempting to reposition themselves as energy infrastructure operators and AI data center platforms.

This change is not accidental but driven by two concurrent forces: on one side, the ongoing pressure on Bitcoin mining economics after the 2024 halving; on the other, the rapid expansion of AI infrastructure demand.

After the 2024 Bitcoin halving, miners’ profitability is further squeezed. Many listed miners’ total costs (including energy, depreciation, interest, taxes, and equipment) face significant pressure amid Bitcoin price volatility. An increasing number of miners are seeking diversified growth paths.

Meanwhile, demand for AI training, inference, and high-performance computing is surging, with large-scale data centers facing new constraints: not GPU shortages, but power shortages, grid access issues, and the lack of ready-to-deploy infrastructure. Even with ample capital, hyperscalers and AI cloud providers may find it difficult to build hundreds of MW-scale campuses within reasonable timeframes. This supply-demand mismatch makes miners with mature power infrastructure, ready industrial parks, and large energy access capabilities strategically valuable.

The real question is not whether these companies can do AI well, but why capital markets are willing to provide billions of dollars in financing. The answer lies in their control of one of the most scarce resources in the AI era — the ability to rapidly deliver large-scale power and infrastructure.

From “Mining Machines” to “Power Assets”: Repricing

In recent years, the market’s valuation logic for miners has been quite clear: fundamentally, a highly leveraged Bitcoin beta. When Bitcoin prices rise, miners tend to gain higher elasticity; after halving, profit margins compress; when prices fall, survival mode kicks in. For most investors, miners have long been viewed as highly cyclical commodity assets, with core variables revolving around Bitcoin price, network difficulty, and energy costs.

But the explosion of AI is changing this cycle.

AI training, inference, and high-performance computing demand enormous computing power, but the core bottleneck is not the GPU itself. What is truly scarce is the support infrastructure: large-scale stable power, grid access, substation and transmission infrastructure, industrial land, cooling systems, and rapid deployment capabilities. These are precisely the assets North American miners have continuously invested in during previous bull markets. Their large-scale farms built to chase Bitcoin hash rate are now being revalued as foundations for AI infrastructure. For more and more miners, the business model is shifting from simply selling hash rate to also selling power capacity and data center capacity.

Why AI clients prefer to partner with miners

For hyperscalers and AI cloud providers, choosing to work with miners is not just about cheaper electricity. The core bottleneck they face is that even if they can procure GPUs, they may not be able to access sufficient power or immediate data center capacity at scale within a reasonable timeframe. Compared to starting from scratch, miners with existing grid access, industrial parks, and mature power infrastructure can significantly shorten deployment cycles. Therefore, what the market is buying is not just electricity, but “large-scale power capacity that can be rapidly delivered.”

More importantly, in today’s North American data center market, the truly scarce resource is shifting from GPU itself to Time-to-Power. For large AI clusters, waiting years for grid approval, transmission construction, and campus development often means missing model training cycles and business windows. In contrast, some miners already possess scalable power access, mature parks, and potential development of hundreds of MW, enabling them to compress deployment from years to much shorter periods. Under this context, what AI clients are purchasing is not just power capacity but a capability for rapid deployment and continuous expansion of infrastructure.

Case studies: contracts, financing, and valuation shifts

Core Scientific (CORZ)

Core Scientific’s (CORZ) long-term partnership with CoreWeave is an early landmark case, with contracted capacity expanding to about 590 MW of key IT load. The latest disclosed base-term contract value has reached approximately $8.7 billion; with subsequent capacity expansions, total revenue potential is expected to exceed $10 billion. The announced acquisition of CORZ by CoreWeave (all-stock, valuation about $9 billion) in July 2025 was terminated on October 30, 2025, due to shareholder rejection of the merger agreement. CORZ remains an independent entity continuing to develop HPC/AI business.

Hut 8 (HUT)

A standout in execution. Signed long-term leasing agreements with top AI clients, including a 15-year, 245 MW IT capacity lease at River Bend park, with a base-term contract value of $7 billion (including higher renewal potential). At Beacon Point (Texas), a 15-year, 352 MW IT capacity lease, with a base-term contract value of $9.8 billion. The two agreements total 597 MW of contracted AI capacity, with a combined base-term contract value of about $16.8 billion. The company supports development through project financing, using triple-net lease structures to enhance long-term cash flow predictability. The market is increasingly reevaluating the company based on long-term cash flows and infrastructure attributes.

Iris Energy / IREN

The company has signed an agreement with Microsoft to deploy about 200 MW of IT load at the 750 MW Childress campus, with a five-year contract value of approximately $9.7 billion (roughly $1.94 billion annually, based on disclosures and market estimates). This deployment pushes the company toward becoming an AI cloud infrastructure provider, with the market starting to reevaluate its value based on long-term contract revenue and infrastructure cash flow logic. The company also signed hardware procurement agreements with Dell and leverages renewable energy advantages to accelerate deployment.

TeraWulf (WULF)

TeraWulf’s HPC / AI business is gradually becoming a key growth engine. The company collaborates with Fluidstack, including the Abernathy campus’s 168 MW AI compute joint venture (25-year agreement, contract revenue about $9.5 billion), supported by project financing to develop AI infrastructure and accelerate its transition into an AI data center platform.

Summary

The cumulative disclosed AI-related contracts, project revenue potential, and market estimates have already reached hundreds of billions of dollars. Some leading companies’ AI-related revenue contributions are beginning to increase, and financing structures are gradually incorporating more project-level debt, long-term notes, and infrastructure financing tools, reinforcing the infrastructure valuation logic.

Most AI projects are expected to be delivered and operational by 2026–2027, but as of mid-2026, not all are fully deployed, and actual revenue contributions are still ramping up.

The core logic of the financing wave: infrastructure-based valuation

What is most noteworthy about this recent wave of financing is not the contract size itself but the changing financing structures.

Historically, miners relied heavily on equity financing, equipment-backed loans, or cyclical financing tied to Bitcoin prices, with costs deeply linked to crypto market volatility. But as some miners sign long-term AI hosting agreements, ultra-long-term leases, and projects with clear cash flow structures, capital markets are beginning to adopt a different perspective.

Some companies are now obtaining project-level debt financing, non-recourse or credit-enhanced structures, triple-net long-term leases, and take-or-pay arrangements. The core significance of these tools is not just “borrowing more money,” but that their income structures are becoming more long-term, predictable, and increasingly resemble traditional infrastructure cash flows.

This signals a market attempt to reprice miners: from typical cyclical commodity companies toward infrastructure assets and growth-oriented energy platforms. The market is betting less on whether these companies will become the next OpenAI, and more on whether they can continuously deliver hundreds of MW of power capacity with rapid deployment capabilities. The recurring keywords in contracts are not models but power capacity, IT load, and interconnection.

Real risks remain

Market optimism does not mean risks are gone. On the contrary, AI transformation itself may become one of the most capital-intensive and hardest-to-execute transitions in miners’ history.

First, capital expenditure pressures. Transforming a mine into a high-density AI data center is far from simple equipment replacement; it requires complex cooling systems, higher-density power architectures, and substantial upfront investments. For some projects, depending on specifications, costs per MW can reach millions to tens of millions USD, meaning that even with financing, project delivery pace will directly impact returns and balance sheets.

Second, customer concentration risk. Many contracts depend on a few hyperscalers, AI cloud providers, or large model companies. If deployment slows, customer demand shifts, or AI infrastructure investments enter a correction cycle, long-term contract values could be reevaluated. Additionally, miners have traditionally excelled at ASIC deployment, energy management, and mine operation, but operating hyperscale AI infrastructure demands new sales capabilities, technical operations, and a more complex partner ecosystem.

In some ways, the high valuations currently assigned are essentially pricing in future execution success. If delivery speed, customer demand, or financing conditions change, the valuation logic may also reverse.

Deeper question: Are these companies still miners?

If long-term power capacity sales can provide more stable cash flows than simply selling hash rate, if long-term leasing offers higher revenue predictability, and if infrastructure valuations remain above traditional mining valuations, a more fundamental question arises: are these companies still Bitcoin miners in the future?

In recent years, the market has viewed miners as cyclical assets, with core variables being Bitcoin price, network difficulty, and energy costs. But as more companies build business models around power capacity, data centers, and long-term infrastructure contracts, their revenue structures, financing methods, and investor narratives are changing.

Bitcoin halving may not end their growth logic but is forcing miners to redefine themselves. Some may evolve into infrastructure platforms centered on AI data center operations; others may maintain a hybrid “mining + AI” model; slow-to-transform firms will still be affected by traditional mining cycles.

This transformation may ultimately determine not only the fate of these miners but also serve as a key case study for how energy assets are being revalued in the AI era.

From this perspective, what capital truly invests in may never have been just computing power, but the ability to continuously deliver that power — land, grid access, network connectivity, and infrastructure capabilities.

Appendix: This article is based on publicly disclosed company announcements, regulatory filings, financial reports, and market data. Contract values, revenue potential, and development capacities mentioned are primarily derived from company disclosures; some figures represent accumulated revenue potential over contract terms or management guidance, not recognized revenue. The content is for research and discussion purposes only and does not constitute investment advice. Market conditions change rapidly; please refer to the latest disclosures from each company.