Bitcoin mining is undergoing its most severe structural transformation since the network's inception. The fourth halving in April 2024 cut block subsidies to 3.125 BTC, and the consequences are now fully visible: hash price collapsed to $23.9/PH/s/day in Q1 2026, an 80% decline from the $120 pre-...
"We are entering the zetahash era at exactly the moment when miners can least afford to be here." — Matthew Sigel, Head of Digital Assets Research, VanEck
Bitcoin mining is undergoing its most severe structural transformation since the network's inception. The fourth halving in April 2024 cut block subsidies to 3.125 BTC, and the consequences are now fully visible: hash price collapsed to $23.9/PH/s/day in Q1 2026, an 80% decline from the $120 pre-halving peak. JPMorgan pegs the weighted average all-in production cost among public miners at approximately $78,000 per BTC — while Bitcoin trades near $62,000. The math is simple: the median listed miner loses roughly $16,000 on every coin produced.
The response has been a mass liquidation of treasury reserves and an accelerating pivot toward artificial intelligence infrastructure. Public miners sold more than 32,000 BTC in Q1 2026 alone — exceeding total 2025 sales and surpassing the 20,000 BTC offloaded during the Terra-Luna collapse of Q2 2022. CoinShares projects that 70% of listed miner revenues will derive from AI/HPC contracts by year-end 2026.
The Bitcoin network crossed 1 zetahash per second (ZH/s) in late 2025 — a 10x increase from the 100 EH/s recorded in early 2021. As of July 5, 2026, hashrate sits at approximately 935–945 EH/s, according to CoinWarz data at block 956,770. Industry projections from CoinShares target 1.8 ZH/s by December 2026.
Network difficulty currently stands at 133.87 trillion, following a 10.09% downward adjustment at block 953,568 in late June — the second-largest negative adjustment of 2026 and the 11th largest in Bitcoin's history. Difficulty peaked at 155.97T after a +6.31% adjustment on October 29, 2025.
The June difficulty drop was driven by three converging forces:
For remaining operators, the adjustment provided immediate relief: each active petahash now earns approximately 11% more bitcoin than before the retarget.
The April 2024 halving cut the block subsidy from 6.25 BTC to 3.125 BTC. Transaction fees have failed to compensate: fees comprised less than 1% of total block rewards for most of 2025, with average fees per block at roughly 0.018 BTC, according to CoinShares' Q1 2026 mining report.
Hash price — the revenue a miner earns per petahash per second per day — tells the story:
| Period | Hash Price (USD/PH/s/day) | |---|---| | Pre-halving peak (early 2024) | ~$120 | | Q4 2025 | $36–38 | | Q1 2026 (March low) | $28–30 | | Q1 2026 close (Hashrate Index) | $23.9 | | April 2026 | $33.25 | | Current (early July 2026) | ~$37.52 |
JPMorgan analysts led by Nikolaos Panigirtzoglou estimate all-in production cost (electricity, hardware depreciation, overhead) across public miners at approximately $78,000 per BTC. The CoinShares report places the weighted average cash cost at $79,995 in Q4 2025. With BTC trading near $62,000, the gap implies that miners operating older-generation hardware above 20 J/TH at power rates above $0.05/kWh are cash-flow negative.
According to CoinShares, any machine less efficient than a Bitmain S19 XP running at $0.06/kWh or higher is unprofitable at current hash prices. This applies to roughly 15–20% of the global mining fleet.
CleanSpark's fiscal Q2 2026 results (quarter ended March 31) illustrate the pressure: revenue fell 24.9% year-over-year to $136.4 million, with a net loss of $378.3 million. Marathon Digital reported $238 million in its most recent quarter — up 64% year-over-year — but achieved this partly through an 82% increase in energized hash rate, indicating revenue growth was volume-driven rather than margin-driven.
Public miners collectively sold more than 32,000 BTC in Q1 2026, according to CoinDesk data. This figure exceeds all four quarters of 2025 combined and surpasses the roughly 20,000 BTC dumped during Q2 2022's Terra-Luna crisis.
Key cases:
The trend marks a departure from the "HODL" strategy that defined public miner treasury management from 2020 to 2024. As CoinDesk reported in March 2026: "Bitcoin miners are becoming AI companies and selling their BTC to fund the transition."
The public mining sector has announced more than $70 billion in cumulative AI and high-performance computing (HPC) contracts, according to CoinShares. The strategic logic is straightforward: miners own power infrastructure, cooling systems, and real estate — assets that AI data centers require. The revenue profile is also fundamentally different: AI contracts typically provide fixed, long-term revenue versus bitcoin mining's volatile, difficulty-adjusted returns.
Company-level commitments:
JPMorgan analysts Reginald Smith and Charles Pearce project miners will announce roughly 1.7 GW of additional critical-IT capacity by late 2026, equal to approximately 35% of their approved power footprint. The bank upgraded Cipher Mining to Overweight with a December 2026 price target of $18 (from $12) and upgraded CleanSpark to Overweight with a $14 target.
CoinShares projects that 70% of listed miner revenues will come from AI/HPC by end-2026 — a transformation that, if realized, would effectively reclassify these companies from crypto-native miners to hybrid energy-compute operators.
The Cambridge Centre for Alternative Finance's Digital Mining Industry Report (April 2025, surveying 49 mining companies across 16 jurisdictions representing 48% of global hashrate) documented a material shift in Bitcoin mining's energy composition:
| Source | 2022 | 2025 Survey | |---|---|---| | Zero-emission total | 37.6% | 52.4% | | Renewables (hydro, wind, solar) | — | 42.6% | | Nuclear | — | 9.8% | | Natural gas | 25.0% | 38.2% | | Coal | 36.6% | 8.9% |
Hydropower leads renewables at 23.4%, followed by wind at 15.4% and solar at 3.2%. Natural gas replaced coal as the single largest energy source. Coal's share collapsed from 36.6% to 8.9% over three years.
Stranded gas capture has emerged as a parallel track. Crusoe Energy deployed over 425 modular data centers across seven U.S. states and Argentina, capturing nearly 22 billion cubic feet of natural gas that would otherwise be flared, mitigating an estimated 2.7 million metric tons of greenhouse gas emissions, according to Forbes reporting.
At current fleet efficiency of approximately 16 J/TH and global hashrate near 1 ZH/s, the network consumes an estimated 350–420 GWh/day — roughly 128–153 TWh/year, representing less than 0.5% of global electricity consumption.
On June 25, 2026, GoMining mined the first known Bitcoin block using Stratum V2's Job Declaration feature through the DMND mining pool. The block was constructed using a miner-built template rather than the pool-determined template standard under Stratum V1.
The significance is structural. As of mid-2026, the top four mining pools collectively produce approximately 70% of all Bitcoin blocks. Under Stratum V1, these four operators unilaterally decide which transactions appear in the majority of Bitcoin's block space. Stratum V2's Job Declaration protocol allows individual miners to build their own block templates while still participating in pooled mining — separating hash contribution from transaction selection.
GoMining's block included transactions associated with GoBTC Pay, its open-source instant payments protocol, demonstrating that miners can prioritize application-specific transactions within their own templates.
Adoption remains nascent. The milestone is a technical proof-of-concept rather than an industry shift. Widespread deployment depends on pool operators enabling the protocol and miners upgrading firmware — neither of which has occurred at scale.
The Bitcoin mining industry in mid-2026 is a study in forced adaptation. The halving's economic pressure, compounded by a 20%+ BTC price decline from January highs and persistent sub-1% fee revenue, has rendered traditional mining uneconomic for a significant minority of operators and margin-thin for most others.
The sector's response — liquidating bitcoin reserves, abandoning the HODL treasury strategy, and redirecting power infrastructure toward AI data centers — is rational given the economics but represents a fundamental identity shift. Companies that listed as Bitcoin miners are re-emerging as hybrid energy-compute platforms. Whether this pivot generates sustainable returns depends on execution of long-dated AI contracts and whether the current $70 billion in announced deals translates to actual deployed capacity and revenue.
For the Bitcoin network itself, the consequences are mixed. Hashrate remains near 1 ZH/s despite the exodus, indicating that new entrants or efficiency upgrades are replacing departing capacity. The Stratum V2 milestone offers a potential counterweight to mining centralization, though meaningful adoption is years away. The energy mix continues to shift toward zero-emission sources, driven more by economics (stranded gas and curtailed renewables are cheap) than by regulatory mandate.
The data suggests an industry that has hit the economic floor the halving was designed to create — and is now rebuilding from it, with AI as the subsidy that BTC fees were supposed to provide.