Scott Aaronson on AI breaking cryptography and responsible disclosure

Andrew Curran points to a post by computer scientist Scott Aaronson, published four days before October 30, 2026, on the possibility that AI models will soon break widely used cryptographic systems.

Aaronson discusses what responsible disclosure would look like when the parties who need to be notified are effectively the entire world. The available report does not include Aaronson's specific claims about which systems are at risk or what disclosure steps he proposes.