Ethereum Foundation researcher Justin Drake’s October 7 warning was a worst-case forecast, not a report that AI had broken Bitcoin or Ethereum wallets. He argued that rapid progress in AI-assisted mathematics could eventually undermine the elliptic-curve signatures used by many crypto wallets, potentially before the expected threat from powerful quantum computers. Vitalik Buterin supported taking the possibility seriously but cautioned holders not to rush into moving funds.
1
10
12
What prompted Drake’s warning?
Drake pointed to reports that OpenAI had released 722 mathematical research results. His concern was that AI might help researchers find new mathematical methods that could weaken cryptographic systems—not that those results had already demonstrated an attack on blockchain signatures.
8
15
Drake described a specific threshold for calling a signature scheme “broken”: recovering a private key from an exposed public key in roughly a week using available hardware, such as a large GPU cluster. The “months, not years” timeline was his worst-case estimate, not a confirmed prediction or a measured capability.
1
4
12
What does “bunker mode” mean?
Drake urged the crypto industry to plan a controlled migration of assets to fresh addresses, starting with large and sophisticated holders. The idea is to use addresses whose public keys have not been revealed onchain.
3
5
Under the hypothetical attack Drake described, an address that has never signed a transaction has an extra layer of protection: its public key has not been exposed for an attacker to target in the way he outlined. This is a precaution tied to that scenario, not a guarantee that an address is permanently secure. Drake also called on Binance, Bitbank, Robinhood, Bitfinex and Tether to strengthen cold-storage arrangements.
3
15
How did Vitalik Buterin’s advice differ?
Buterin’s response, reported about 10 hours after Drake’s post, agreed that AI-driven mathematical progress merits serious attention. But he advised against scrambling to move funds immediately, saying that mistakes in wallet migrations had cost him more than hacks.
8
21
The difference is mainly about urgency: Drake advocated planning a controlled migration, while Buterin emphasized that rushed transfers can create their own risks. Buterin also raised a broader concern—that AI could weaken lattice-based post-quantum cryptography, including ML-DSA, within two years.
14
24
Both discussions point toward continued work on cryptographic approaches that rely on hashes, but the reports do not establish a settled replacement or an immediate migration requirement for ordinary users.
10
14
What is known—and what remains uncertain?
The reports describe a risk scenario, not a demonstrated ability to recover wallet keys. No practical attack on the relevant blockchain signatures was reported, and the cited mathematical work was not presented as an OpenAI demonstration against crypto cryptography.
1
15
17
That distinction matters: Drake’s warning is a call to prepare for a possible change in the threat landscape, not evidence that funds are currently exposed to a new AI-powered attack. Buterin’s caution adds a practical counterweight: any migration plan must account for the possibility of losing funds through mistakes during the transfer itself.
15
21