Publicly traded Bitcoin miners sold 32,000 BTC in Q1 2026 — more than their combined sales for all of 2025 — as hashprice collapsed to $28–30/PH/s/day, a post-halving record low. The quarter marked the largest sell-off by listed miners on record, exceeding even the approximately 20,000 BTC liquid...
"We are no longer a Bitcoin company." — Ben Gagnon, CEO, Bitfarms
Publicly traded Bitcoin miners sold 32,000 BTC in Q1 2026 — more than their combined sales for all of 2025 — as hashprice collapsed to $28–30/PH/s/day, a post-halving record low. The quarter marked the largest sell-off by listed miners on record, exceeding even the approximately 20,000 BTC liquidated during the Terra-Luna crisis in Q2 2022.
The forced selling reflects a structural profitability crisis. According to CoinShares' Q1 2026 mining report, the weighted average cash cost to produce one bitcoin among publicly listed miners reached approximately $80,000 in late 2025. With BTC trading near $69,200 in March, miners were losing an estimated $19,000 on every coin produced. Three consecutive negative difficulty adjustments — the first such streak since July 2022 — confirmed widespread capitulation, with the network's 30-day average hashrate falling 5.8% quarter-over-quarter to 1,004 EH/s.
Simultaneously, the industry's pivot to artificial intelligence infrastructure accelerated sharply. More than $70 billion in cumulative AI and high-performance computing (HPC) contracts have been signed across the public mining sector. CoinShares projects 70% of listed miner revenues will come from AI by end-2026. The economic arithmetic is stark: Bitcoin mining yields roughly $1 million per megawatt annually, while AI compute generates $10–20 million per megawatt. That 10–20x spread is reshaping capital allocation across the sector.
Hashprice — the daily revenue earned per petahash per second of deployed computing power — has become the mining industry's definitive profitability metric. In Q1 2026, it fell to approximately $28/PH/s/day in late February before recovering modestly to $30–33/PH/s/day by quarter's end.
For context, the commonly cited breakeven threshold for the industry sits at approximately $35/PH/s/day. At $33/PH/s/day, an estimated 20% of the global fleet operates at a loss, according to CoinShares. The decline was driven by three converging factors:
At a hypothetical $100K BTC, hashprice would recover to roughly $37/PH/s/day — above breakeven for most operators. Below $80K, CoinShares forecasts continued pressure, though the expected shutdown of unprofitable rigs should act as a natural floor by reducing network difficulty.
The Q1 2026 sell-off was not marginal. Publicly listed miners collectively dumped 32,000 BTC in three months, generating an estimated $2.2–2.5 billion in proceeds at prevailing prices.
Key individual company actions:
| Miner | Q1 2026 BTC Sold | Estimated Proceeds | Notes | |-------|-------------------|-------------------|-------| | Riot Platforms | 3,778 BTC | ~$289.5M | Largest single-quarter sale in company history | | Core Scientific | ~1,900 BTC (Jan.) | ~$175M | Announced plans to liquidate substantially all remaining treasury | | Bitdeer | Treasury to zero | Undisclosed | Shifted strategy to treat BTC as liquidity source, not long-term hold | | Marathon Digital | Multiple tranches | >$1B (recent months) | CEO cited "directing computational power toward most productive use" | | CleanSpark | Significant sales | Undisclosed | SG&A of $17,848/BTC among lowest in peer group |
According to Cointelegraph, the 32,000 BTC figure exceeded the roughly 20,000 BTC that publicly listed miners sold during Q2 2022, making it the largest quarterly miner liquidation on record. The strategic rationale varied: some operators sold to fund AI infrastructure buildouts, others to cover operating losses, and several — notably Core Scientific — to service debt obligations ahead of the CoreWeave acquisition.
Bitcoin's mining difficulty experienced a 7.76% downward adjustment in March 2026, the second-largest negative adjustment of the year. The network's hashrate retreated to approximately 942.83 EH/s, down roughly 15% from the October 2025 peak.
Three consecutive negative difficulty adjustments — the first such sequence since July 2022 — constitute a textbook capitulation signal. Mid-generation hardware (S19j Pro-class rigs at ~29.5 J/TH) running at average industrial electricity costs of $0.05/kWh was operating below breakeven by year-end 2025, and conditions worsened into Q1 2026.
According to BingX, citing CoinShares data, 15–20% of legacy rigs are now unprofitable. At a $30/PH/day hashprice, any rig weaker than an S19 XP paying $0.06/kWh or more is losing money. A further modest decline of roughly 0.7% was projected for the next difficulty cycle in mid-April.
The current hashrate of approximately 1.084 ZH/s reflects a partial recovery, but the trend remains downward from peak levels. The United States maintains its position as the largest mining market at 37.4% of the global hashrate, equivalent to roughly 375 EH/s.
Energy costs have become the decisive variable in miner survival. Several factors intensified the squeeze in Q1 2026:
The efficiency frontier has shifted dramatically. Best-in-class hardware (Antminer S23 Hydro) operates at approximately 10.8 J/TH, representing a 7x improvement from the 98 J/TH standard of 2018. But even these machines face margin pressure when electricity exceeds $0.10/kWh and BTC trades below $75K.
The scale of the mining industry's pivot to AI infrastructure is without precedent. Over the course of 2025 and early 2026, Bitcoin miners signed GPU co-location and cloud service deals with hyperscalers worth over $70 billion in aggregate, according to CoinShares.
Major announced contracts:
| Company | Deal Partner | Value | Duration | Capacity | |---------|-------------|-------|----------|----------| | Core Scientific | CoreWeave | $10.2B | 12 years | ~590 MW across five sites | | TeraWulf | Multiple | $12.8B contracted | Ongoing | ~2.8 GW across five sites | | Hut 8 | Google-backed | $7B | 15 years | River Bend campus |
CoreWeave agreed to acquire Core Scientific in an all-stock transaction, gaining approximately 1.3 GW of gross power across Core Scientific's national data center footprint with an incremental 1 GW+ of potential expansion capacity. CoreWeave estimated approximately $500 million in annual run-rate cost savings by end-2027 from eliminating lease overhead.
The economic logic is straightforward. According to industry analysts cited by CoinDesk, Bitcoin mining yields roughly $1 million per megawatt annually, while AI compute generates $10–20 million per megawatt. Gross margins in Bitcoin mining have fallen to approximately 60%, while AI cloud infrastructure generates margins of approximately 85% with lower energy overhead per unit of revenue.
As Nick Hansen, CEO of the Luxor mining pool, told DL News: "Resisting the urge to transition to AI" will be Bitcoin miners' biggest challenge in 2026.
The market is pricing the AI pivot into mining equities with a clear premium. According to CoinShares:
The market is paying more than double for AI exposure. Individual stock performance reflects this divide:
CoinShares projects that listed miners could derive as much as 70% of their revenue from AI by end-2026, up from roughly 30% at the time of writing. This represents a wholesale identity transformation: from Bitcoin miners to AI infrastructure operators that happen to mine Bitcoin on the side.
The hashrate decline raises legitimate questions about Bitcoin network security, though context is important. The current hashrate of approximately 1 ZH/s remains historically high. However, the direction matters: a 15% decline from peak, combined with an accelerating reallocation of power capacity from mining to AI, could structurally reduce the hashrate ceiling.
The capitulation dynamic carries a built-in stabilization mechanism. As unprofitable miners exit, difficulty adjusts downward, improving margins for remaining operators. The question is whether the AI pivot creates a permanent "pull" on mining capacity that keeps hashrate below previous peaks even when BTC prices recover.
CoinShares notes that some cannibalization of existing mining facilities is likely as miners convert to AI data centers. The $70 billion in HPC contracts creates long-term contractual obligations that cannot easily be reversed. Once a facility is converted to host NVIDIA GPU clusters for CoreWeave, it does not return to SHA-256 hashing.
The Bitcoin mining industry is undergoing a structural transformation that goes beyond a cyclical downturn. The Q1 2026 data — record BTC liquidations, consecutive negative difficulty adjustments, and hashprice at post-halving lows — describes an industry where the core product has become economically subordinate to a new revenue stream.
The $70 billion in AI infrastructure contracts is not a hedge. It is a replacement. When a megawatt of capacity generates 10–20x more revenue running AI workloads than hashing SHA-256, the economic calculus is unambiguous. The companies that can execute the transition — Core Scientific, TeraWulf, Hut 8 — are being rewarded with valuations more than double their pure-play peers.
The implications for Bitcoin's security model remain an open question. The network's difficulty adjustment mechanism provides a self-correcting floor, but the structural reallocation of power capacity to AI creates a permanent drag on hashrate growth. The mining industry that emerges from this shakeout will be smaller, more efficient, and increasingly peripheral to the business models of the companies that once defined it.