Aaronson noted that among the 722 mathematical results recently published by OpenAI, there is a striking lack of breakthroughs in cryptography, adding that the US government has censored academic results on quantum cryptanalysis. Rumors also suggest that OpenAI's most recent release of mathematical findings is the first of three planned batches.
Wednesday, October 21, 2026
Story
AI Companies Use Internal Models to Break Cryptographic Protocols
AI13 days ago
The newest models developed by leading artificial intelligence labs are being deployed in private attempts to crack the foundations of modern encryption. Theoretical computer scientist Scott Aaronson reported that some companies have quietly begun using their non-public models to probe whether they can break important cryptographic protocols and primitives, according to sources.
Looking for more context or analysis?
Tilt chat provides reliable, tailored insights, alerts, and automations.