Anthropic Claude AI Breaks Encryption Algorithms

Anthropic's Claude Mythos Preview AI model discovered new vulnerabilities in cryptographic systems, including a practical key-recovery attack on HAWK-256 (a post-quantum signature scheme under NIST review) and weaknesses in AES variants that researchers had previously failed to crack. The findings raise questions about AI capabilities in cryptanalysis and the security of systems being standardized for federal use.

This is a long-running storyline that has developed over 6 days. The homepage highlights its most recent activity, so the outlet count there reflects the latest wave. The totals above cover the full run.