The hashprice is dead. Long live the hashprice.
Bitcoin miners are shutting down rigs at a rate not seen since the 2022 capitulation. Hashrate has dropped 21% from its peak of 1.14 ZH/s to 900 EH/s. Hashprice sits at $31.8 per PH/s—roughly half of what it was a year ago. Yet the stocks of several mining companies have more than doubled over the same period. Something is off. The market is not paying for Bitcoin mining; it is paying for a pivot.
This is not a paradox. It is a transformation. The same companies that once competed for block rewards are now competing for AI compute contracts. The same infrastructure that housed ASICs is being retrofitted for GPUs. The same power contracts that secured cheap electricity for Proof-of-Work are now being used to power Proof-of-Intelligence. The narrative has shifted from 'mining blocks' to 'allocating energy.'
Let me step back. In 2018, when I started my education platform, I spent countless nights dissecting the philosophy of mining. Satoshi's vision was decentralization through distributed energy. Miners were the guardians of the network. But after the fourth halving, the economics have changed. Miner revenue has collapsed. Hashpower is concentrating in three pools. The decentralization dream is hollowing out. And now, the survivors are not holding the line; they are pivoting.
But pivoting to what? The answer is AI/HPC infrastructure. The thesis is simple: the scarcest asset in the AI boom is not compute—it is low-cost power and land with existing grid access. Bitcoin miners have exactly that. They own substations, cooling systems, and long-term power purchase agreements. They are, in essence, energy arbitrageurs with a side of computational churn.
The data is stark.
CoreWeave, the AI cloud giant, was originally a crypto miner. Now it is valued at $19 billion. Riot Platforms recently signed a 20-year, $9.1 billion contract with Anthropic—a deal that transforms a mining company into an AI infrastructure provider. The total value of AI/HPC contracts secured by public mining companies now exceeds $70 billion. That is not a rounding error. That is a sector redefinition.
The market has priced this in, but not equally.
Companies that have aggressively pivoted—WULF, IREN, CIFR—have seen their stocks rise over 100% in the past year. Their enterprise value multiples now sit at 12.3x, compared to 5.9x for pure-play miners like MARA, which has seen its stock fall 40% over the same period. The divergence is a statement: investors are no longer buying Bitcoin optionality; they are buying AI revenue streams.
But here is the contrarian angle: the market is betting on a flawless execution.
In my years auditing mining operations, I have seen the fragility of these promises. A mining facility designed for ASICs has a PUE of 1.3 at best. An AI data center requires PUE below 1.1, with liquid cooling, fiber interconnect, and 99.999% uptime. The retrofit costs are not trivial. GPU clusters require different O&M skills. The management teams that were once experts in Bitcoin mining are now competing with veteran data center operators like Equinix and Digital Realty. The learning curve is steep.
And then there is the contract risk. The $9.1 billion deal with Anthropic is a 20-year commitment. But AI is a fast-moving field. What if the demand for large language models plateaus? What if a new architecture makes current GPUs obsolete? The contract is a double-edged sword: it provides revenue visibility, but it also locks the miner into a specific technology path. If the AI client renegotiates or cancels, the miner is left with stranded assets.
We do not build walls; we build bridges for value.
This is the spirit of the pivot. The miners are not abandoning Bitcoin; they are diversifying the use of their energy assets. They are building bridges between the energy grid and the AI compute market. But the bridge must be built with care. A bridge that collapses is worse than no bridge at all.
Ideas have no gas fees, only gravity.
For months, the narrative has been that miners must pivot or die. That narrative is partly manufactured. VCs who want to raise funds for AI infrastructure love to tell a story of inevitable convergence. But the gravity of reality is different. The cost of capital is rising. The time to convert a mining site into a Tier 3 data center is 12 to 18 months. The market is pricing in the end state, not the intermediate risk.
Let me give you a specific signal. In my own analysis of public mining companies, I have seen that the 'pivot' thesis is strongest for those with existing interruptible power contracts. But interruptible power is not suitable for 24/7 AI workloads. The miners that succeed will be those that can upgrade their grid connection to firm power. That requires new substations, transmission upgrades, and regulatory approvals. It is not a simple swap.
Freedom is a protocol, not a permission.
The Bitcoin mining industry was built on permissionless innovation. Anyone with an ASIC and cheap power could participate. The AI pivot is different. It requires permission from utilities, hyperscalers, and large AI labs. The miner becomes a service provider, not a sovereign node. That changes the culture. The ethos of 'code is law' is replaced by 'the contract is king.'
Culture is the new consensus mechanism.
The companies that navigate this transition will be those that retain the hacker spirit while adopting enterprise discipline. They will build a culture that values both decentralization and reliability. They will understand that the future is written in code, but felt in spirit.
So what is the takeaway?
Bitcoin miners are not dying; they are evolving. The pure-play mining model is under pressure, but the infrastructure layer is being repurposed. The market is correctly pricing in the value of energy assets, but it may be underestimating the execution risk. The winners will be the miners that can deliver on their AI contracts without overleveraging their balance sheets.
In the chaos of the chain, find the signal.
The signal is clear: the next phase of digital infrastructure will be built on the foundations of Bitcoin mining. But the path from hash to AI is not a straight line. It is a winding road of capital, culture, and grit.
Truth is not mined; it is remembered.
And the truth is that the miners who remember their original purpose—to secure a decentralized network—will also build the most resilient AI infrastructure. The two are not mutually exclusive. They are complementary. The future is not a choice between Bitcoin and AI. It is a synthesis.
Let me close with a reflection. I have spent nearly a decade in this industry. I have seen projects rise and fall. I have seen narratives shift from DeFi to NFTs to gaming to AI. The one constant is that the infrastructure layer always accrues value. The miners are the infrastructure. They are the soil. The crop may change, but the soil remains.
Now, the question is: will the soil be fertile enough to grow both Bitcoin and AI? I believe it can, but only if the farmers are wise. They must not overwater one crop at the expense of the other. They must maintain the balance.
The future is written in code, but felt in spirit.
And the spirit of this pivot is not greed. It is survival. It is adaptation. It is the relentless drive to turn energy into value. That is the essence of mining. That is the essence of Bitcoin. And that is the essence of the coming revolution.
We do not build walls; we build bridges for value. And those bridges will carry us into a new era of decentralized compute.
