Bitcoin mining just hit a stress level JPMorgan says it hasn’t seen before. The bank flagged that the network is now reacting faster and harder to every price move. That’s because a record share of miners are operating right at the edge of profitability. Here’s what the numbers mean — and why it matters beyond the mining world.
What Does “Mining Sensitivity” Actually Mean?
Think of Bitcoin mining like running a factory. It produces one thing: newly minted bitcoin. The factory costs real money to run — electricity, hardware, staff. If bitcoin’s price stays above those costs, you profit. If it falls below, you bleed.
JPMorgan reports that the beta of Bitcoin mining difficulty relative to price has climbed to 0.62 over the past six months. That’s a measure of bitcoin mining sensitivity — how sharply the network reacts to price shifts. In plain English: for every meaningful price move, the network’s total computing power now swings more dramatically than before. That total computing power is called hashrate.
Hashrate is the backbone of Bitcoin’s security. When miners shut down machines, the network’s defenses thin out. The system compensates through what’s called a difficulty adjustment — a built-in recalibration every two weeks. But when Bitcoin falls below production cost, higher-cost miners power down. Hashrate drops, and difficulty adjusts lower. The loop tightens fast.
The $78,000 Problem
JPMorgan places bitcoin’s all-in production cost at about $78,000. That figure covers electricity, hardware depreciation, and overhead across public miners. Bitcoin trades near $63,000. That’s a 17% gap — and it’s held for five consecutive months.
Mining economics have worsened all year. Citing CoinShares’ Q1 2026 mining report, JPMorgan said roughly 20% of miners are currently unprofitable. One in five miners is losing money on every block they mine. That’s not a blip — it’s a structural squeeze.
The stress shows in live data. Mining difficulty fell 10.09% in early June — the second-largest single decline of the year. Bitcoin’s hashrate dropped 12% in June. Two drops of that magnitude in one calendar year are unusual. It signals something deeper than routine network noise.
Bitcoin’s price has been sliding since its October 2025 all-time high above $126,000. The BTC price forecast for 2026 has grown more cautious since. Multiple institutions now point to a potential bottom later in the year.
Miners Are Selling Everything They Can
When you’re running a factory at a loss, you sell inventory just to keep the lights on. That’s exactly what’s happening.
Operators, including MARA, CleanSpark, Riot Platforms, Cango, Core Scientific, and Bitdeer, sold a combined 32,000 BTC in Q1 2026. The purpose: fund operating expenses. That figure surpasses those companies’ total bitcoin sales for all of 2025. It also sets a new quarterly record. The previous high was 20,000 BTC in Q2 2022, during the bear market after the Terra-Luna collapse.
The dollar math is stark. At $63,000 per coin, six companies pushed roughly $2 billion in forced supply onto the market in a single quarter. These aren’t discretionary sellers. They have no real choice.
Miners collectively held about 1.8 million bitcoin at publication — down from 1.86 million at the end of 2023. Treasury drawdowns are now a recurring feature of the environment.
Meanwhile, Michael Saylor’s Strategy and its bitcoin machine showed its own cracks in 2026. That added another layer of caution across the institutional landscape.
Why This Creates a Feedback Loop
Here’s the part that makes this more than just a mining story.
A small price drop can push marginal miners past their breakeven. They shut down machines. Hashrate shrinks. Difficulty adjusts lower — automatically. Then the cycle starts again.
In past cycles, there was more cushion. Miners operated across a wide range of profit margins. Network hashrate didn’t swing hard when prices dipped. Now, a larger share of the network clusters right around breakeven. The buffer is gone. Even modest price declines trigger meaningful shutdowns. That narrows the gap between a functioning network and one undergoing visible contraction.
JPMorgan expects this elevated sensitivity to persist. It will last as long as bitcoin remains well below its $78,000 production cost. Larger and more frequent difficulty adjustments are the likely result.
The AI Lifeline — and Its Reality Check
Faced with this squeeze, miners have a Plan B: pivot to artificial intelligence.
The logic makes sense on paper. Mining companies already own the power infrastructure, land, and cooling systems that AI data centers need. Instead of mining bitcoin, they host AI workloads for tech giants and charge premium rates. AI contracts offer stable, multi-year revenue. That’s more predictable than volatile block rewards squeezed by the 2024 halving.
But there’s a big catch. Asset manager VanEck just flagged a serious gap between the hype and the delivery. Bitcoin miners pivoting to AI face a roughly $50 billion near-term funding gap. Long-term capital needs could reach $221 billion. The industry has so far delivered only about 25% of the AI and high-performance computing (HPC) capacity it has leased to customers.
“Execution, not signing, becomes the next premium,” said VanEck’s Griffin MacMaster and Matthew Sigel. Companies missing construction milestones risk structural de-ratings from investors.
Announcing an AI deal is easy. Actually building the infrastructure to deliver on it is an entirely different problem.
Is There a Silver Lining?
JPMorgan’s analysts aren’t calling for a crash. They noted that weak market sentiment of this kind has, in past cycles, acted as a contrarian bullish signal.
Here’s the theory. When miners capitulate and difficulty drops sharply, the cost of producing each bitcoin falls. That resets the floor lower. Survivors face less financial pressure. Historically, these capitulation periods have preceded strong recoveries — though timing them is nearly impossible.
The bank has shifted from its earlier “overweight” 2026 stance to a cautious one. A stronger second half for crypto has two main prerequisites. First: digital asset treasury companies need to clarify how they’ll meet rising dividend obligations. Second: U.S. crypto market structure legislation needs to be passed. JPMorgan currently assigns less than a 50% probability.
The latest Fed signals on interest rates and bitcoin add more uncertainty heading into Q3. Rate expectations remain hawkish — a headwind for risk assets.
FAQs
What is bitcoin mining difficulty, and why does it keep dropping in 2026?
Bitcoin’s difficulty adjustment keeps block production steady — roughly one new block every 10 minutes. When miners shut off machines and hashrate falls, the network makes mining easier to compensate. In 2026, two 10%-plus difficulty drops have already occurred. That reflects significant miner exits driven by unprofitable economics. You can explore more context in our breakdown of bitcoin’s bear market timeline.
What is hashprice and how does it affect miner profitability?
Hashprice measures how much revenue a miner earns per unit of computing power per day. It’s expressed in dollars per petahash per second. It combines the bitcoin price, network difficulty, and block reward into one number. When hashprice falls, even efficient miners feel the squeeze. In mid-2026, hashprice sits around $33 per petahash per second per day. That’s well below the levels where marginal operators stay comfortable.
Why did bitcoin’s 2024 halving make the current miner squeeze worse?
The halving cuts the block reward miners receive by 50%. It happened in April 2024 and immediately halved mining revenue — without changing electricity or hardware costs. Bitcoin then slid from its $126,000 peak in late 2025. Miners now face pressure from both directions: lower payouts from the halving and a lower coin price. BlackRock’s new Bitcoin income ETF reflects how Wall Street is adapting to this new yield environment.
What’s the difference between a miner shutting down and the Bitcoin network going offline?
Bitcoin’s network doesn’t go offline when miners exit — it self-adjusts. As hashrate falls, difficulty drops. Remaining miners find blocks more easily and cheaply. The network keeps running. What changes is security. A lower hashrate makes certain attack types slightly more feasible in theory. In practice, Bitcoin’s network remains extremely robust. The bigger risk is economic: forced selling from stressed miners creates sustained downward price pressure.
Could the AI pivot actually save publicly listed bitcoin mining companies?
Potentially — but it’s not a quick fix. Companies like Core Scientific have signed multi-year AI hosting deals. Those provide more predictable cash flows than mining. But building the data center infrastructure to deliver on those contracts costs tens of billions. Many miners don’t have that capital today. The companies best positioned are those with investment-grade AI customers, proven construction execution, and strong enough balance sheets to bridge the funding gap.