How AI Is Draining Crypto: Half the Developers Gone, $70 Billion of Compute Sold
In the first quarter of 2026, publicly listed Bitcoin miners sold a record 32,000 BTC to fund AI data centers. Over the same twelve months, crypto lost 56% of its active developers. The machines and the people are leaving together, and those are the only two metrics that can actually kill a blockchain. Here's what's happening, and which loss will be harder to reverse.
The miner math stopped working
By early 2026, producing one Bitcoin cost a listed miner roughly $65,000 to $75,000 all-in, which is about what the coin was selling for. And the clock is against them: the 2028 halving (the rule that cuts the block reward in half every four years) will drop rewards from 3.125 to 1.5625 BTC.
Wall Street made the exit easy. A mining company trades at roughly 6 to 7 times its revenue. The same company hosting AI workloads trades at 12 to 15 times. Same building, same power line, double the valuation. Listed miners have now signed more than $70 billion in AI contracts, and AI already brings in about 30% of revenue at the biggest names.
Company | The move | The numbers (as reported, spring 2026) |
Core Scientific | Exited bankruptcy in Jan 2024, went all-in on AI hosting | $10B+ in contracts, AI is about two-thirds of revenue |
CoreWeave (ex-Atlantic Crypto) | Pivoted from Ethereum mining to GPU cloud | 43 data centers, ~250,000 GPUs, ~$67B backlog with Microsoft, Meta, OpenAI |
Riot Platforms | Reportedly sold 1,100 BTC to buy Texas land for AI | Stock jumped ~11% within hours |
The most valuable "crypto mining" company of this cycle is one that stopped mining crypto.
AI is buying the electricity, not the hardware
Mining chips, known as ASICs, are built to do exactly one job, and that job is useless for AI. So what are Microsoft, Meta, and Google actually paying for? Time, in the form of electricity. A new US grid connection takes 3 to 7 years. Transformers have lead times of up to 4 years. Construction adds another 18 to 24 months. Roughly 2,000 gigawatts of projects sit stuck in US connection queues, double what the country has installed.
A mining farm, with power, permits, and cooling already in place, skips that entire line. The miners' real asset was never hash power. It was the power contract.
What this does to Bitcoin's security
In the short term, the protocol absorbs the hit. When a January 2026 Texas storm knocked 10 to 15% of global hashrate offline, blocks slowed from the usual 10 minutes to 12 or 13. Then the difficulty adjustment (a built-in re-tune every 2,016 blocks, about every two weeks) kicked in and block production went back to normal.
But that algorithm fixes block times, not business models. The 2028 halving cuts miner revenue in half again, and one industry estimate says up to half of today's miners could switch off if the price doesn't compensate. What makes this cycle different is that exiting miners finally have a better-paying buyer for their megawatts. The result is a shrinking security budget: less money paid out to keep the chain expensive to attack.
The developer migration: where the builders went
The miner exodus is loud, full of press releases and stock pops. The developer exodus is silent, and the GitHub data makes it measurable.
According to GitHub activity tracked by analytics firm Artemis, weekly active crypto developers fell from about 8,700 to roughly 4,600 in one year, a 56% drop. Weekly code commits collapsed from around 850,000 to 210,000, down about 75% since early 2025. No major chain was spared:
Ecosystem | Weekly active developers (early 2026) | 3-month change |
Ethereum | ~2,800 | −34% |
Solana | ~940 | −40% |
Base | ~380 | −52% |
Now look at where the same platform is growing. GitHub added roughly 36 million developers in 2025 alone, bringing its total above 180 million, and platform-wide commits rose about 25% year over year. AI-related repositories passed the 4.3 million mark, and generative AI projects attract more than a million monthly contributors. TypeScript, the language behind modern web apps and much of today's AI tooling, gained over a million new contributors in a year and overtook Python as GitHub's most-used language.
Same platform, opposite directions. The talent didn't quit tech. The engineer who wrote smart contracts (programs that run directly on a blockchain) in 2024 is shipping AI agents and LLM tooling in 2026.
And this loss is harder to reverse than hashrate. Hashrate is mercenary by design: it returns within months when margins improve. Ecosystems don't work that way. Fewer developers means fewer experiments, fewer audits, and slower tooling. Talent compounds wherever it lands, and right now it's compounding somewhere else.
The takeaway
Yes, there's a counterpoint. The developers who stayed skew senior: newcomers fell 58%, while contributors with more than two years of experience grew 27% and now write about 70% of all commits. Consolidation, not collapse. But a smaller, more serious room is still a smaller room.
AI never had to beat crypto in the market. It outbid crypto for the two things no blockchain can print: electricity and engineers. Through 2028, watch hashrate and weekly commits, not the price chart.