Ethereum Foundation researcher Justin Drake has urged the crypto industry to start planning for what he calls “bunker mode,” warning that artificial intelligence could break the elliptic-curv
Ethereum Foundation researcher Justin Drake has urged the crypto industry to start planning for what he calls “bunker mode,” warning that artificial intelligence could break the elliptic-curve signatures securing Bitcoin and Ether wallets “in months, not years.” In a post on X on October 7, Drake called for a “controlled mass migration” of funds to fresh addresses whose public keys have never appeared onchain. Within hours the warning split the field: Ethereum co-founder Vitalik Buterin agreed the risk is real but told holders not to rush.
Drake’s worst case is not the quantum attack the industry has spent years preparing for. He wrote that it is now “reasonable to brace for the possibility that ECDSA breaks before qday,” defining a break as fast private-key recovery, “e.g. in one week,” on available hardware such as “a large GPU cluster.” That is a classical-computing scenario, and it would land years earlier than Ethereum’s December 2029 target for moving onto quantum-resistant cryptography.
The trigger is OpenAI’s release of 722 mathematics manuscripts on October 6, produced by an unreleased frontier model across roughly 4,000 problems. Drake said the drop showed “mathematical superintelligence is upon us,” and argued that elliptic curves, with their “rich structure,” are more exposed than hash functions, which are designed to leave little pattern to exploit. The distinction reframes the quantum threat regulators have flagged: AI would not need a quantum computer to hunt for a shortcut.
Drake’s recommendation is deliberately gradual. Large and sophisticated holders should move “the bulk of their funds” to addresses that have never signed a transaction, rotating any remainder when they do sign. He stressed this “does not require new cryptography or new wallets,” and added that wallets holding under 50 BTC enjoy partial cover from “Satoshi’s shield,” roughly 20,000 exposed addresses holding 50 BTC each.
Buterin agreed the industry should take AI-accelerated math seriously but pushed back on urgency. “I don’t recommend anyone scramble to move their funds to new wallets today,” he wrote, noting he has “lost more money in botched migrations than I have lost in all hacks combined.” Buterin went further than Drake in one direction, warning that lattice-based cryptography, the foundation of quantum-resistant standards such as ML-DSA, could take “serious hits from the next two years of AI math,” and said Ethereum’s roadmap is already shifting toward “hash-only” signatures.
No practical break of Bitcoin or Ethereum wallet keys has been demonstrated, and Drake himself noted the “striking under-representation of cryptographic breakthroughs” among OpenAI’s manuscripts. As Bitcoin traded below $83,000 on October 8, tokens tied to quantum-resistant cryptography moved the other way, with Algorand gaining about 9% and Starknet roughly 4% since midnight, according to CoinDesk market data.
The concrete next steps are modest: sophisticated actors reducing public-key exposure where it is cheap, and exchanges hardening cold storage. Drake said he will address institutions in London next month and push for the Ethereum roadmap’s hash-based cryptography timelines to be “revisited and accelerated.” For most holders, the takeaway both researchers agree on is measured: no panic, but less unnecessary public-key exposure from here on.
