Public Bitcoin miners sold 32,000 BTC in Q1 2026, exceeding total sales for all of 2025 and surpassing the 20,000 BTC liquidated during the Q2 2022 Terra-Luna collapse. The sell-off coincides with hashprice falling to $29/PH/s/day — a five-year low — and average production costs for listed miners...
"We expect further capitulation among higher-cost operators in H1 2026 unless BTC's price recovers materially." — CoinShares, Q1 2026 Bitcoin Mining Report
Public Bitcoin miners sold 32,000 BTC in Q1 2026, exceeding total sales for all of 2025 and surpassing the 20,000 BTC liquidated during the Q2 2022 Terra-Luna collapse. The sell-off coincides with hashprice falling to $29/PH/s/day — a five-year low — and average production costs for listed miners reaching approximately $80,000 per BTC, while Bitcoin trades near $68,600. According to CoinDesk, miners are now losing roughly $19,000 on every coin produced at full cost.
Simultaneously, more than $70 billion in AI and high-performance computing (HPC) contracts have been signed across the public mining sector. Core Scientific, Hut 8, and IREN collectively account for approximately $27 billion of that figure. The industry is bifurcating: miners with cheap power and next-generation ASICs are consolidating hashrate, while those with flexible infrastructure are converting megawatts from SHA-256 to GPU clusters. By year-end, an estimated 85% of global hashrate will be controlled by 12 or fewer entities, and several former miners may derive up to 70% of revenue from AI compute.
This report examines the economic forces driving the restructuring of Bitcoin mining, the security implications of hashrate concentration, and the value-capture dynamics as mining companies evolve into multi-purpose infrastructure operators.
The post-halving margin compression that began in April 2024 has reached a structural inflection point. According to CoinShares' Q1 2026 Mining Report, hashprice — the daily revenue per petahash per second of deployed computing power — fell to $29/PH/s/day in March 2026, down from $63/PH/s/day in July 2025 and $35-37/PH/s/day in November 2025. This represents a 54% decline in eight months.
The weighted average cash cost of producing one Bitcoin among publicly traded miners rose to approximately $80,000 in Q4 2025, the most recent period with audited data. With BTC trading near $68,600 as of April 7, 2026, the gap between production cost and market price has turned negative for most operators. CoinDesk reported on March 22 that miners are losing an estimated $19,000 per BTC produced when factoring in hardware amortization, site maintenance, and corporate overhead.
At the current hashprice of approximately $30/PH/s/day, any miner running hardware below the Antminer S19 XP with electricity costs at or above $0.06/kWh is operating at a loss. CoinShares estimates this covers 15-20% of the global mining fleet. Operators with sub-$0.03/kWh power — primarily those accessing stranded energy, flare gas, or sovereign-subsidized electricity — remain the only segment generating positive cash flow from mining alone.
Mining 1 BTC in the United States now requires more than 750,000 kWh at current difficulty levels. At the national average commercial rate of $0.1363/kWh, the electricity cost alone exceeds $106,000 per coin — well above the spot price.
The profitability crisis triggered the largest miner sell-off on record. According to data compiled by CoinMarketCap and reported by multiple outlets, publicly traded miners — including MARA, CleanSpark, Riot Platforms, Cango, Core Scientific, and Bitdeer — collectively sold more than 32,000 BTC in Q1 2026. This exceeded total miner sales for the entire year of 2025 and surpassed the previous quarterly record of approximately 20,000 BTC set during Q2 2022.
Individual company data illustrates the scale:
The divergence in stock performance is notable. CleanSpark and Riot, which have maintained lower cost profiles and disciplined capital allocation, have outperformed. MARA, which pursued aggressive expansion and now carries higher all-in costs, has underperformed despite generating the highest absolute revenue from BTC sales.
Data from early April shows a continued spike in miner-to-exchange transfers, according to on-chain analytics, suggesting that small-to-medium operators are liquidating wallet balances before shutting down permanently.
The global Bitcoin hashrate peaked at approximately 1.15 ZH/s (1,150 EH/s) in early 2026 before declining to roughly 940 EH/s by April 7 — an 18% drop. CoinDesk reported on January 19 that hashrate had already fallen 15% from its October 2025 high, with the Hash Ribbon indicator — a metric that measures the convergence of short-term and long-term hashrate moving averages — signaling active miner capitulation.
The network responded with several large difficulty adjustments:
The difficulty mechanism is functioning as designed: unprofitable miners exit, difficulty falls, and remaining operators become marginally more profitable. However, the speed and magnitude of the hashrate decline raises questions about network security concentration, discussed below.
The most consequential structural shift in Bitcoin mining is not about mining at all. It is the conversion of mining infrastructure — specifically power capacity, cooling systems, and real estate — into AI and HPC data centers.
More than $70 billion in cumulative AI and HPC contracts have been announced across the public mining sector, according to data aggregated by insights4vc. The largest deals:
The economic logic is straightforward. A megawatt of power capacity dedicated to a 15-year fixed-rate AI lease generates more predictable and higher-margin revenue than the same megawatt pointed at Bitcoin's difficulty-adjusted, subsidy-halving mining algorithm. CleanSpark has publicly stated that Bitcoin mining investment "doesn't make a lot of sense" at current hashprices compared to returns available in AI infrastructure.
CoinDesk reported on March 27 that miners are effectively "becoming AI companies and selling their BTC to fund the transition." The 32,000 BTC sold in Q1 was not primarily panic selling — it was strategic capital recycling from a lower-return activity (mining) into a higher-return activity (AI compute).
By year-end 2026, listed miners could derive as much as 70% of their revenue from AI, up from approximately 30% in Q1, according to insights4vc projections. At that point, calling these companies "Bitcoin miners" will be a misnomer.
The combination of margin compression and AI conversion is accelerating industry consolidation at an unprecedented rate. An estimated 85% of global hashrate is projected to be controlled by just 12 publicly traded entities or sovereign wealth funds by year-end 2026. This represents a significant concentration increase from 2024, when the top 12 entities controlled roughly 35-40% of hashrate.
This trend has direct implications for Bitcoin's security model:
Centralization risk. Bitcoin's censorship resistance depends on a distributed set of miners, none of whom can individually control block production. As hashrate concentrates among a small number of publicly traded companies subject to regulatory jurisdiction, the network becomes more susceptible to coordinated compliance pressure — such as OFAC-compliant block production or transaction filtering.
Regulatory surface area. Publicly traded miners are subject to SEC reporting, OFAC sanctions compliance, and domestic energy regulation. As they absorb a larger share of hashrate, the portion of Bitcoin's block production subject to single-jurisdiction regulatory oversight increases commensurately.
Counterargument. Mining pools, not individual miners, coordinate block production. Even if entity-level concentration increases, the use of multiple pools and the ability of miners to switch pools provides a degree of decentralization at the block-production level. Additionally, the remaining 15% of hashrate from private and sovereign operators provides a baseline of diversity.
The nation-state mining experiment is showing mixed results. Bhutan, which operated a sovereign Bitcoin mining operation through its investment arm Druk Holding & Investments, has begun unwinding its position. Over the past 18 months, Bhutan has sold off a significant portion of its holdings, with the latest transfer of 319.7 BTC occurring in early April 2026, according to on-chain data reported by multiple outlets.
Bhutan's predicament illustrates the challenge: when the operation began, Bitcoin prices were higher and difficulty was lower. At current levels — $68,600 BTC and all-time-high difficulty — the margins for small-scale sovereign mining, even with subsidized hydroelectric power, have narrowed to near zero.
At the institutional level, sovereign wealth funds are taking a different approach. BlackRock CEO Larry Fink has disclosed that sovereign wealth funds are "adding incrementally at $120,000, $100,000" and "establishing a longer position," treating Bitcoin as a strategic reserve asset rather than a mining operation. This represents a shift from the "mine it" thesis to the "buy it" thesis for sovereign accumulation.
The divergence is instructive: sovereign mining makes economic sense only with access to sub-$0.03/kWh power at scale. For most nation-states, purchasing Bitcoin on the open market is more capital-efficient than standing up and operating mining infrastructure.
The restructuring of the mining industry is redistributing economic value across multiple stakeholders:
Winners:
Losers:
The Bitcoin protocol itself is neutral: the difficulty adjustment ensures block production continues regardless of how many miners participate. But the economic value captured by miners — the $10.3 billion in annual block rewards at current prices — is being redistributed from a broad, distributed set of operators to a narrow, consolidated set.
The Bitcoin mining industry is undergoing its most significant structural transformation since China's 2021 ban. The combination of post-halving margin compression, five-year-low hashprices, and the availability of higher-return AI compute opportunities is forcing a three-way split: low-cost operators consolidating hashrate, infrastructure-rich operators converting to AI, and everyone else shutting down.
The 32,000 BTC sold in Q1 was not a sign of panic — it was the funding mechanism for the largest industrial pivot in crypto history. Whether these companies remain "Bitcoin miners" in any meaningful sense by 2027 is an open question.
For Bitcoin's security model, the concentration of hashrate among a small number of publicly traded, regulated entities is a development that warrants monitoring. The difficulty adjustment ensures the network functions, but the political and regulatory surface area of block production is narrowing. The next 12 months will determine whether the mining industry's evolution strengthens or weakens the properties that make Bitcoin distinct.