Quoting Matthew Green
AI may break public-key encryption faster than humans can replace cryptographic standards
“The problem here is that the speed of AI producing surprises, and the speed of human beings replacing standards (even with the very best AI assistance) are just orders of magnitude different.”
Cryptographer Matthew Green warns there is a 15% chance AI could undermine confidence in existing public-key encryption algorithms, creating a dangerous asymmetry between AI's ability to produce cryptographic surprises and humanity's ability to replace standards. The core concern is preparedness: recovery from such a surprise requires advance work, not reactive response. This is a credible expert raising a tail-risk scenario that intersects AI capability growth with critical infrastructure security.