Summary
- Vitalik Buterin, co-founder of Ethereum, has stated that it is "plausible" for advancements in artificial intelligence to compromise encryption methods, particularly lattice-based cryptography, within the next two years.
- This type of cryptography is one of the methods considered for protecting blockchains against quantum computing threats, but Buterin cautions it may also be vulnerable to AI-related risks.
- His comments follow those of Ethereum Foundation researcher Justin Drake, who suggested that the industry should adopt a "bunker mode" to safeguard assets from potential AI risks.
Ethereum's co-founder, Vitalik Buterin, has expressed concern that artificial intelligence could potentially break encryption within a two-year timeframe. He emphasizes the necessity for the crypto industry to fortify itself against both AI and quantum computing vulnerabilities.
This warning comes shortly after Justin Drake, a researcher at the Ethereum Foundation, urged the community to prepare for a "bunker mode" strategy, which involves transferring funds to addresses that have not had their public keys revealed. Drake posited that the advancements in AI could lead to the breaking of encryption before quantum computers reach the capability to do so, a scenario often referred to as Q-Day.
Myriad: Predict Ethereum's next move.Buterin commented on X, stating, "There is a good chance that the concrete security of lattices will take serious hits from the next two years of AI math. If AI can deliver 50 years of mathematical progress in just two years, this could plausibly include significant advancements in breaking lattice encryption."
Lattice-based cryptography is one option being explored by various crypto projects, such as Cardano, to shield themselves against quantum threats. This shift could mean that users might need to consider the short-term risks posed by AI when transferring funds, despite the long-term quantum threat looming.
I don't recommend anyone scramble to move their funds to new wallets today. But we should take the risks to cryptography from AI-accelerated math seriously, and minimize our exposure to not just quantum-vulnerable cryptography, but also potentially AI-vulnerable cryptography.…
— vitalik.eth (@VitalikButerin) October 7, 2026
Isabel Foxen Duke, who leads the Mara Foundation, a project focused on making Bitcoin quantum-resistant, noted, "Potential breaks in cryptography due to quantum computing were primarily thought to affect Elliptic curve cryptography and RSA. However, classical threats (such as those posed by AI) could jeopardize cryptographic types deemed quantum secure, like lattice-based cryptography. We must adopt a more cautious approach given these classical risks."
As a precaution, Buterin advocates for increasing the size of lattices to enhance security. He believes hash-based cryptography may be a more effective method for achieving quantum resistance, although he concedes that AI advancements could also threaten hash security.
Ananda Banerjee, founder of analytics firm Charlie Quant Lab, advised caution, stating, "I wouldn’t treat Vitalik's two-year scenario as a deadline. The post doesn’t demonstrate that these systems have been broken. I would prepare carefully rather than rush to move funds."
Buterin does not recommend any extreme actions in light of this potential threat. Instead, he suggests users keep their assets in addresses that have not been transacted with, thus reducing the likelihood of exposure to attacks.
These AI-related concerns arise just a month after Buterin remarked that AI wouldn't spell doom for crypto security, asserting that security measures would evolve in tandem with technological advancements. His recent statements may indicate a shift in perspective within the industry.