Scientist: AI labs quietly test whether models can break cryptography

Theoretical computer scientist Scott Aaronson, citing sources, says AI labs have quietly started probing whether their models can break important cryptographic protocols. No official confirmation or details have been provided.

• Scott Aaronson cited sources on his Shtetl-Optimized blog
• Labs are testing models against cryptographic protocols
• The probes are being run quietly, with no official data