From Months to Minutes: Mythos and AI’s Impact on Enterprise Cyber Risk

Anthropic disclosed Claude Mythos — a frontier AI model discovered thousands software vulnerabilities, including a 27-year-old flaw in one of the world's most security-hardened operating systems — and launched Project Glasswing, a coalition with AWS, Apple, Google, Microsoft, JPMorgan Chase, CrowdStrike, Palo Alto Networks, and others to put these capabilities in defenders' hands first. Members gathered for a timely virtual meeting that focused on what the board needs to understand, the assumptions in our enterprise risk registers that no longer hold, and the specific questions every director should be asking management at the next meeting.