The US Cybersecurity and Infrastructure Security Agency is using Anthropic's AI model Mythos. Mythos is being used to audit government code for bugs. The agency is utilizing a private, offensive-grade AI model for this purpose.