A theoretical computer scientist, citing sources, says AI labs have quietly started probing whether their models can break important cryptographic protocols (Scott Aaronson/Shtetl-Optimized)

A theoretical computer scientist, Scott Aaronson, citing sources, reports that AI labs have begun to quietly investigate whether their models possess the capability to break important cryptographic protocols. This development suggests a growing concern within the AI community regarding the potential security implications of advanced AI models and their interaction with established encryption methods. The probing of cryptographic protocols by AI models could have significant ramifications for data security and privacy.

This report is the first to publicly disclose that AI labs are actively probing their models' ability to break cryptographic protocols, unlike previous discussions that were purely theoretical.