In mid-April 2026, the global mining sector witnessed two landmark capital transactions. US-listed mining firm TeraWulf issued 47.4 million new common shares at $19 per share, raising $900 million. Shortly after, Canadian miner HIVE Digital Technologies announced plans for a $75 million private placement of zero-coupon, exchangeable senior notes maturing in 2031.
Both financing rounds are aimed at building AI data centers and procuring GPUs, rather than expanding traditional Bitcoin mining rigs. This is not an isolated occurrence. From Core Scientific liquidating its Bitcoin holdings to secure multi-billion-dollar AI hosting contracts, to Hut 8 signing a $700 million infrastructure lease deal with Google-backed Fluidstack and AI unicorn Anthropic, the "pivot" in mining companies’ core business models has evolved from sporadic strategic shifts to a systemic migration across the entire industry.
The Capital Logic Behind the Two Financings
TeraWulf priced its latest offering on April 14, upsizing the deal from an initially planned $800 million to $900 million. Underwriters also secured a 30-day overallotment option for up to 7.11 million additional shares. The funds raised will primarily go toward constructing an AI data center campus in Horseville, Kentucky, and fully repaying outstanding amounts under its bridge credit facility. Morgan Stanley served as joint bookrunner, with Cantor Fitzgerald as equity capital markets advisor.
At the same time, TeraWulf released preliminary Q1 results as of March 31, projecting revenue between $30 million and $35 million and adjusted EBITDA up to $3 million. By quarter-end, the company held $310 million in cash and equivalents, with total debt at $580 million.
HIVE’s $75 million exchangeable note financing followed a different path. The zero-coupon notes, due in 2031, are issued by HIVE Bermuda 2026 Ltd. and privately placed with qualified institutional investors. Initial buyers have an option to purchase an additional $15 million in notes. HIVE also structured a capped call transaction to minimize dilution for existing shareholders.
HIVE will use the proceeds to purchase GPUs, expand data centers, and for general corporate purposes. Notably, HIVE disclosed it is gradually winding down ASIC mining at its Boden, Sweden facility, converting the site into a Tier III high-performance computing data center. It also deployed its first GPU cluster in Paraguay for early-stage large language model training workloads.
Why Mining Companies Are Accelerating Their AI Pivot in 2026
The AI pivot among mining companies is not new, but data from Q1 2026 highlights the urgency of this trend.
After the 2024 Bitcoin halving, mining profit margins dropped by about 50%, while operating costs remained unchanged. In the second half of 2025, the sustained weakness in the Bitcoin price further squeezed miners’ margins. Industry estimates put the weighted average cash cost for listed miners to produce one Bitcoin at around $79,995 in Q4 2025, with total costs even higher after accounting for depreciation and maintenance. As of April 17, 2026, Bitcoin traded at roughly $74,729.90, creating a severe cost-price inversion, with some miners losing over $19,000 per coin mined.
In Q1 2026, hashprice fell to a five-year low of about $29 per PH per day. Meanwhile, total network hash rate dropped from a peak of roughly 1.16 ZH/s to about 853 EH/s—a 22% decline—directly reflecting mine closures and hashpower migration.
In March 2026, Core Scientific confirmed it had sold about 1,900 Bitcoins for $175 million and planned to liquidate nearly all its holdings to accelerate AI infrastructure expansion. At the same time, MARA Holdings sold over 15,000 Bitcoins to reduce leverage, and Bitdeer’s Bitcoin holdings fell to zero.
Between April 14 and 16, 2026, TeraWulf and HIVE announced their respective financings, pushing the AI pivot to a new scale.
The drivers behind this shift are twofold. First, the structural deterioration of Bitcoin mining economics—rising difficulty, weak prices, and high cash costs—has made traditional mining models unsustainable. Second, the explosive growth in AI compute demand is pulling miners in a new direction. While retrofitting existing mines for AI takes 18–24 months, building new data centers from scratch can take over five years. Mining companies’ power infrastructure thus becomes a rare and valuable asset.
Scale and Structure of the Mining Sector’s AI Transformation
Industry data shows the AI pivot has evolved from tentative experiments to a systemic trend. Let’s quantify this shift across contract size, financing structure, and revenue transformation.
AI Order Book Size: By early 2026, cumulative AI and HPC orders across several miners reached roughly $38.5 billion. Major deals include TeraWulf’s $12.8 billion AI data center contract with Fluidstack, IREN’s $9.7 billion, five-year agreement with Microsoft, Hut 8’s $7 billion infrastructure lease with Google/Anthropic, and Core Scientific’s $10+ billion, 12-year hosting agreement with CoreWeave.
Revenue Transformation: Core Scientific’s AI hosting revenue surged 268% year-over-year in Q4 2025, far outpacing its mining income. HIVE’s latest fiscal quarter saw revenue of $93.1 million, up 219% year-over-year. Although HIVE reported a net loss of $91.3 million due to depreciation from expansion projects, management remains focused on long-term infrastructure growth.
Diverging Financing Strategies: Funding sources for the AI pivot are increasingly diverse. TeraWulf opted for a public equity offering, accepting dilution in exchange for capital. HIVE chose zero-coupon exchangeable notes, trading future conversion rights for immediate liquidity. Core Scientific both expanded its credit lines to $1 billion and systematically liquidated Bitcoin holdings. These approaches reflect each firm’s unique capital structure and risk appetite.
Market Response: Short-term share price declines after financing announcements contrast with longer-term re-ratings. TeraWulf shares are up 68.06% year-to-date, HIVE has gained about 37%, and Core Scientific returned 173% over the past year. Some analysts believe the long-term value of AI hosting contracts is not yet fully reflected in current valuations.
| Comparison | Bitcoin Mining Business | AI Data Center Business |
|---|---|---|
| Revenue Stability | Highly dependent on Bitcoin price, cyclical | 10–15 year contracts, decoupled from BTC price |
| Client Quality | Revenue from network protocol, no direct clients | Investment-grade clients like Microsoft, Google, Anthropic |
| Cash Flow | Highly volatile, halving cuts revenue by 50% | Predictable USD cash flows |
| Capital Allocation | Ongoing ASIC miner purchases | Power infrastructure upgrades, GPU cluster deployment |
| Asset Reusability | Fast ASIC depreciation, rapid tech turnover | Data center infrastructure as long-term assets |
The capital markets’ valuation logic is shifting. Previously, miners’ valuations tracked Bitcoin prices. Post-AI pivot, the market is reassessing these firms using a "power-enabled data center REIT" framework. This shift is the core reason mining companies are receiving positive capital market feedback for their AI transformation.
Dissecting Public Opinion: Support, Skepticism, and Sober Assessment
The mining sector’s AI pivot is seen as an inevitable move in the "compute-first" era. Digital economy scholars at the Shanghai Academy of Social Sciences argue that Bitcoin was merely an early token born of a specific historical phase of compute. Miners liquidating Bitcoin to invest in AI infrastructure is a historic shift from the old token system to a new compute-driven paradigm. According to Yu Jianing, rotating chair of the Academic Committee of the Hong Kong Registered Digital Asset Analysts Association, access to power is now a more scarce strategic resource than chips themselves. Miners, with a decade of experience in power procurement, possess the very capabilities most coveted by AI giants.
The challenges of this pivot are often underestimated. On one hand, Bitcoin ASIC miners and AI GPUs are not interchangeable, and mining site power systems need upgrades to support higher-density compute loads. A modern 100 MW AI data center costs over $4 billion, with about 70% spent on servers and GPUs—far higher capital intensity than traditional mining. On the other hand, miners are often highly leveraged—TeraWulf’s total debt stands at $580 million—and debt pressure may become a constraint before AI projects generate positive cash flow.
Some industry observers caution that the scale and pace of the AI pivot will vary. Not all miners have the capability to build high-density data centers or the power redundancy needed for AI hosting. The gap between leading and smaller miners will widen, and only those with high-quality power assets, strong capital management, and technical reserves are likely to complete the transition.
Industry Impact: Mining Ecosystem, Network Security, and Capital Flows
The large-scale pivot to AI data centers is reshaping the Bitcoin mining industry and the broader crypto ecosystem.
Bitcoin Network Security: Total network hash rate fell from a peak of about 1.16 ZH/s to 853 EH/s—a 22% drop. Lower hashpower reduces the cost of attacking the network. While a 51% attack remains prohibitively expensive, continued hashpower outflows could pose long-term security risks. At the same time, hashpower concentration may rise, with only miners enjoying low-cost electricity able to maintain operations.
Bitcoin Market Supply Pressure: Large-scale liquidation of Bitcoin holdings by miners created significant sell-side pressure in Q1 2026. MARA sold over 15,000 BTC, Riot Platforms liquidated more than 3,700 BTC, and additional sales from Core Scientific and Cango added to the collective selling pressure, weighing on Bitcoin’s market price.
Capital Market Flows: The investment logic for mining stocks is being redefined. Traditionally, investors viewed mining equities as leveraged Bitcoin proxies. Post-AI pivot, some institutions now value these firms using data center REIT models. This paradigm shift means mining stock prices may become less correlated with Bitcoin and more tied to the AI infrastructure market’s outlook.
Industry Value Chain Effects: Reduced demand for mining rigs will pressure ASIC manufacturers. The redistribution of power resources will reshape regional energy markets, and mining operations service providers will need to adapt their business models. The ripple effects of the AI pivot are spreading throughout the crypto value chain.
Multi-Scenario Evolution Outlook
Given current information and industry trends, we can outline three possible evolutionary scenarios:
Scenario 1: Accelerated Pivot—Leading Miners Establish Dual-Core Models
Logic: AI compute demand continues to soar—multiple research firms forecast a 25–36% CAGR for the global AI data center market from 2026 onward, reaching hundreds of billions of dollars by 2034. With power resources as the key bottleneck, miners’ existing infrastructure appreciates rapidly. After initial AI deployments, leading miners use stable cash flows from long-term contracts to support their mining operations, creating a "dual-core" model of AI hosting plus Bitcoin mining. In this scenario, the sector’s valuation multiples rise, and correlation with Bitcoin prices declines further.
Scenario 2: Intensified Segmentation—Only a Few Complete the Pivot
Logic: The capital intensity of AI data center construction far exceeds that of mining—a 1 GW data center costs about $55 billion, with GPUs as the largest expense, followed by power and cooling infrastructure. Most small and mid-sized miners cannot afford the upfront investment or lack adequate power infrastructure for AI hosting orders. After the transition window closes, only those embedded in the AI supply chain command a capital premium; others face acquisition or exit. Industry concentration rises, and Bitcoin hashpower consolidates among the few remaining miners, who nevertheless endure low profit margins.
Scenario 3: Bitcoin Mining Economics Recover, Slowing the Pivot
Logic: If Bitcoin prices rebound sharply to the $90,000–$100,000 range in the next one to two years, mining economics will improve significantly. Some miners with partial AI deployments may slow their transition and redirect resources to mining. However, this does not alter the long-term trend—after the fifth halving in 2028, block rewards will drop to 1.5625 BTC. Even with high prices, per-unit mining output will halve again. Thus, a Bitcoin price rally may delay the pivot but cannot reverse the industry’s structural migration.
These scenarios are not mutually exclusive and may unfold sequentially. In the short term, Bitcoin price volatility will affect the pace of the pivot. In the medium term, successful delivery and positive cash flow from AI contracts will be the key proof points. In the long term, the 2028 halving will further cement the necessity of an AI infrastructure pivot for miners.
Conclusion
TeraWulf’s $900 million equity offering and HIVE’s $75 million convertible notes are milestone events in the 2026 wave of AI pivots among miners—but not the end of the story. From Core Scientific to Hut 8, MARA to IREN, the collective shift to AI data centers is fundamentally a revaluation and redistribution of "compute" as a core asset.
For the Bitcoin mining industry, this transformation means a new network security dynamic, changes in market supply, and a realignment of capital flows. For crypto market participants, the AI pivot is rewriting the old logic of "mining stocks = Bitcoin leverage." For the broader tech sector, miners’ decade-long accumulation of power infrastructure is positioning them as critical asset providers in the AI compute era.
No matter how the transition unfolds, one trend is clear: miners are evolving from "laborers" of the Bitcoin network into "power landlords" of the AI compute infrastructure age. This migration across crypto and AI is only just entering the fast lane.
