Anthropic Engineers at NSA: Mythos AI for Cyber Ops

Anthropic has reportedly placed about six engineers inside the National Security Agency. These engineers are there to help the NSA use Mythos, Anthropic's most capable cyber model, for offensive operations. This arrangement could help infiltrate networks in countries like China or Iran. This collaboration comes despite Anthropic challenging Pentagon policies on military AI use. The company was even labeled a "supply-chain risk" after refusing to allow its models for mass surveillance or autonomous weapons. That designation led to a procurement ban across the Department of Defense. The NSA arrangement exists because of an explicit carve-out from this ban. Mythos is considered too dangerous for broad release, capable of finding and exploiting zero-day vulnerabilities. This situation highlights the complex relationship between AI development and national security.