The Current State of Bitcoin Mining: May 2026
Lumerin Protocol7 min read·Just now--
Bitcoin mining is going through its most consequential transformation since the first halving in 2012. Here is what the data actually shows.
The Network Is Under Pressure, and the Protocol Is Handling It
Six times in 2026, the Bitcoin network has automatically reduced its mining difficulty. The most recent reduction, a 2.3% drop on May 1 at block height 947,520, brought difficulty to 132.47 trillion. That figure sits 10.7% below where it started the year.
This is not a malfunction. The difficulty adjustment mechanism is doing precisely what Satoshi designed it to do. When fewer miners are online, block times lengthen past the 10-minute target, and the protocol responds by making the next set of blocks easier to find. Average block time in early May sits at around 10 minutes and 28 seconds, which is slightly above target and explains why the algorithm has been trimming difficulty consistently throughout the year.
Network hashrate peaked above 1,000 exahashes per second, crossing the symbolic 1 zettahash threshold, before beginning a gradual decline around April 19. As of early May, the seven-day simple moving average sits near 965 EH/s, with daily readings ranging between 899 and 977 EH/s. The next difficulty adjustment is projected for around May 17, and current estimates point to a modest upward revision of approximately 0.5%, which would suggest the hashrate is beginning to stabilize.
Hashprice Is the Number That Matters Most
Hashprice measures daily miner revenue per petahash of computing power. It is the single most important profitability signal in the industry, because it captures the combined effect of Bitcoin price, block subsidy, transaction fees, and network difficulty.
Hashprice collapsed to between $28 and $30 per petahash per day during Q1 2026, some of the lowest readings on record. Since then, it has recovered meaningfully. As of May 3, hashprice stands at $37.52 per PH/s per day, up from $34.39 at the previous difficulty epoch and approximately 13.65% above its late-April reading.
The recovery is being driven by two converging factors. Bitcoin has climbed back toward $80,000 to $81,000, breaching that level for the first time since January 31. And falling difficulty is reducing the computational cost per block for miners who stayed online, expanding their share of a fixed subsidy pool. Miners who survived the Q1 shakeout are now operating in a notably better revenue environment than they were 60 days ago.
The Production Cost Problem Is Real
The economics of 2026 Bitcoin mining are severe for operators carrying high costs. According to CoinShares’ Q1 2026 mining report, the weighted average cash cost to produce one bitcoin among publicly listed miners reached approximately $79,995 to $90,000. Bitcoin spent most of the first quarter trading well below that range, producing losses on a per-coin basis for many large operators.
Minerstat’s hardware data clarifies where the dividing line sits. Mining equipment older than the Bitmain Antminer S19 XP is operating at a loss at electricity rates of $0.06 per kilowatt-hour. That obsolete hardware represents an estimated 15 to 20 percent of the global fleet. At the top of the efficiency curve, the Antminer S23 Hydro delivers 1,160 terahashes per second. Operators running that generation of hardware at sub-$0.04 electricity are the ones with comfortable margins in the current environment.
The spread between winners and losers in Bitcoin mining has never been wider. Energy cost and hardware generation now determine survival in a way that pool choice or geographic location simply cannot offset.
Pool Concentration Deserves Attention
Mining pool distribution data for the seven-day period ending May 3 shows significant concentration at the top. Foundry USA produced 31.51% of 987 blocks over that window. AntPool and ViaBTC together add another 26.84%. The three dominant pools collectively account for 58.35% of total network hashrate.
MiningPoolStats reports that 115 distinct entities are currently contributing computational power to Bitcoin. The tail of the distribution is long, but the head is very concentrated. Foundry USA’s position in particular represents a level of block production share that was uncommon even five years ago.
It is worth noting the distinction between pool concentration and mining centralization. Individual miners pointing hashrate to a pool can redirect to a competitor at any time, and they retain full custody of their hardware and operations. Pool concentration reflects coordination efficiency and fee competitiveness rather than ownership of the underlying infrastructure. Still, the ongoing concentration trend is one of the metrics that serious observers of Bitcoin’s security model are watching closely.
Public Miners Are Becoming AI Companies
The dominant story in Bitcoin mining in 2026 is not the difficulty curve or the hashprice recovery. It is the wholesale transformation of publicly listed mining companies into artificial intelligence and high-performance computing infrastructure providers.
CoinShares projects that publicly listed miners could derive up to 70% of their revenues from AI by December 2026, up from approximately 30% in Q4 2025. The cumulative value of AI and HPC contracts announced across the public mining sector now exceeds $70 billion. These are not exploratory pilot agreements. They are decade-long infrastructure commitments with major technology counterparties.
The contract ledger is striking in its scale. IREN secured a five-year partnership with Microsoft projected to generate $1.94 billion in annualized revenue at an 85% project-level EBITDA margin. CoreWeave’s deal with Core Scientific runs for 12 years and is worth $10.2 billion. TeraWulf has $12.8 billion in contracted HPC revenue. Hut 8 signed a $7 billion, 15-year lease for AI infrastructure at its River Bend campus.
The financial logic behind these moves is straightforward. Bitcoin mining gross margins have compressed from approximately 90% during the 2021 cycle to around 60% today. AI cloud infrastructure delivers margins near 85% with more predictable, contract-backed revenue. When the 2024 halving cut block rewards to 3.125 BTC and Bitcoin’s price retreated from its cycle peak near $126,000, the relative attractiveness of AI infrastructure became difficult to ignore.
MARA Holdings, historically one of the most publicly committed Bitcoin accumulation advocates, sold over $1 billion in BTC to fund its AI transition. Core Scientific liquidated $175 million in March 2026 for the same purpose. Publicly listed miners sold more than 32,000 Bitcoin during Q1 2026 alone, a record that surpassed their total BTC sales for all of 2025.
The Opportunity for Efficient Operators
The AI pivot by large public miners has created an unambiguous benefit for smaller and mid-scale operators running efficient hardware. Every machine that leaves the Bitcoin network reduces difficulty for the machines that remain. The surviving miners on modern ASICs are capturing a larger percentage of the same 3.125 BTC block subsidy per block.
Industry analysis of the current environment suggests that the compound effect of falling difficulty and a recovering hashprice translates to a 15 to 20 percent profit improvement at the individual miner level compared to Q4 2025 economics. For a 10-machine fleet on Tier 3 hosting, the difference amounts to roughly $7,000 per year in net profit. For a 50-machine enterprise deployment, the figure approaches $35,000 per year.
These projections carry real uncertainty. Hashprice could fall if Bitcoin’s price retreats. Difficulty could jump sharply if sidelined hashrate returns quickly. CoinShares projects the network could ultimately recover to 1.8 ZH/s by year-end 2026 as new hardware deployments come online. If that timeline is accurate, the current window of favorable difficulty economics will be shorter than many operators assume.
What to Watch Over the Next 60 Days
The May 17 difficulty adjustment will serve as a first real signal of whether the hashrate decline has bottomed. An upward adjustment, even a modest one, would indicate that network computing power is stabilizing and that the acute phase of the AI-driven exodus is slowing. A seventh consecutive downward adjustment would suggest that economic pressure on the mining cohort is still intensifying.
Bitcoin’s price trajectory is the master variable for everything else in this equation. A sustained move above $85,000 to $90,000 would bring production costs back into alignment with spot price for most efficient operators, slow the economic rationale for the AI pivot, and create conditions for new ASIC deployment at scale. A return to $68,000 to $70,000 would accelerate the structural shift already underway.
The first concrete AI revenue disclosures from major public miners in their Q2 2026 earnings reports will also be closely watched. The gap between announced contract value and recognized revenue has been wide through H1 2026. When large-scale AI revenue starts flowing through income statements, it will reshape how investors and analysts value the sector entirely.
The Bottom Line
Bitcoin mining in May 2026 is a study in simultaneous stress and adaptation. The protocol is absorbing a meaningful reduction in global hashrate with zero drama, adjusting difficulty automatically and maintaining its 10-minute block cadence across six consecutive reductions. Hashprice is recovering from historic lows. Efficient operators are benefiting from a competitive landscape thinned by high production costs and a compelling alternative in AI infrastructure.
The publicly listed mining sector is undergoing a genuine identity transformation. Companies that built billion-dollar businesses producing Bitcoin are now deploying that infrastructure for machine learning workloads and signing decade-long data center contracts. Whether this constitutes a permanent structural shift or a temporary response to compressed margins depends almost entirely on where Bitcoin’s price goes from here.
The Bitcoin network has no opinion on any of this. It simply adjusts.
Data sourced from Hashrate Index, Minerstat, CoinWarz, CoinShares Q1 2026 Mining Report, and industry sources. All figures as of May 5, 2026. This article is for informational purposes only and does not constitute financial or investment advice.
