Bitcoin mining profitability has fallen to a two-year low due to the split among miners.

Record difficulties and declining on-chain fees have pushed Bitcoin mining profitability to a two-year low, widening the gap between miners surviving on razor-thin margins and those reinventing themselves as data center operators for the AI ​​boom.

Mining has been a homogenous industry that moves in sync with the price of Bitcoin. However, we are now evolving into a two-speed economy where hash power, not energy strategy, defines success.

At about $42.14 per terahash per day, Bitcoin’s hash price (the industry’s shorthand for miner revenue per unit of computing power) has fallen to the bottom 4% of its two-year range.

In the past month alone, the price of Bitcoin has fallen 19%, and the pressure has intensified as the Bitcoin market has fallen to around $101,500.

The real culprit is not the spot price.

This is the structural mathematics of the network itself. Difficulty has increased 31% in the past six months, hashrate has increased 23% and fees, once tightened due to general activity and congestion, have fallen to their lowest level since the spring. The result is pure compression, with more systems fighting for fewer rewards.

For small-scale miners, this combination is devastating. Many businesses are operating below break-even, especially those involved with high-cost electricity contracts or outdated hardware. This situation is strangely reminiscent of previous cycle lows in late 2020 and 2022, when the weakest players capitulated just before a rebound.

But this time, the stress tests are taking place in a very different environment. The advent of AI and high-performance computing has created a whole new escape route for miners, allowing them to shift their infrastructure to non-Bitcoin workloads.

Earlier this week, Iris Energy announced a $9.7 billion five-year deal with Microsoft to supply AI and data center capacity, effectively repurposing part of the company’s fleet as an HPC provider. The stock reaction was immediate and brokers began re-evaluating IREN, Core Scientific, Riot Platforms and Cleanspark as “AI infrastructure plays” rather than pure Bitcoin proxies.

This change, based on real-world revenue diversification, is why miner stocks can bounce back even when hash prices fall. The market is starting to reward grid-scale flexibility and long-term power contracts over hash output.

The contrast with legacy miners is stark. Companies that focus solely on Bitcoin production have little room to react when margins collapse.

With hash price figures of around $43 per PH/s/day near multi-month lows, miner revenue is currently at its lowest profitability level since April. These companies are still paid in full from Bitcoin block rewards and transaction fees, with their profits automatically decreasing with each increase in difficulty.

If you can’t prevent exposure or access ultra-low-cost energy, you’ll be left waiting for the next block subsidy moratorium or network fee increase.

Meanwhile, Marathon Digital is showing what scale can do to offset the crisis. The company recently reported record quarterly revenue of $123 million, driven by operational excellence and doubling down on new business lines adjacent to AI hosting.

The revenue mix now includes a mix of mining and AI operations, showing how the definition of a miner is changing. Marathon’s massive energy usage allows it to opportunistically reduce or redirect load when Bitcoin mining economics get tough, by selling excess power or renting infrastructure for HPC workloads.

These differences are now visible in market data. Stock investors are treating weak hash prices not as an existential risk but as a filter that separates miners with sustainable business models from those simply seeking block rewards.

According to a recent note from Bernstein, “the hash price issue does not affect AI-centric miners.” This sentiment captures the structural shift currently underway: the evolution of Bitcoin mining from a single-purpose pursuit into a multi-market data infrastructure business.

Tracking when a recession may reverse: Some clear indicators.

The first is a difficulty plateau or rollover, a signal that unprofitable hashrates fall offline, creating a natural rebalancing that increases the remaining miners’ share of the reward.

The second is the resurgence of on-chain fees due to congestion or demand for new inscription styles. Either way, the hash price can be raised without changing the price of Bitcoin.

The third and perhaps most important factor is the continued expansion of AI or HPC contracts. Each new megawatt is diverted to external workload, reducing effective competition on the Bitcoin network and stabilizing the margins of those that remain.

Other variables are also important. Winter energy prices, reduction incentives and local regulations all affect who can survive long-term economic pressures. Mergers, liquidations, and site closures typically accelerate as hash prices approach cycle lows.

Historically, this has been a countersignal to the broader market, a prelude to easing difficulty adjustments and the accumulation of new miners.

The next increase in difficulty will provide the first real test of whether this compression has reached its limits. If hash rate growth stagnates while fees rise, the hash price may begin a slow mean reversion toward equilibrium.

Until then, the mining industry is divided between those solving Bitcoin’s hardest math problems and those completely rewriting them with AI.

The post Bitcoin mining profitability falls to two-year low as miner split appears first on CryptoSlate