Last week, Anthropic pulled back the curtain on Claude Mythos Preview, an AI model so capable at finding and exploiting software vulnerabilities that the company decided it was too dangerous to release to the public. Instead, access has been restricted to roughly 50 organizations”, Microsoft, Apple, Amazon Web Services, CrowdStrike and other vendors of critical infrastructure”, under an initiative called Project Glasswing. The announcement was accompanied by a barrage of hair-raising anecdotes: thousands of vulnerabilities uncovered across every major…
First seen on securityboulevard.com
Jump to article: securityboulevard.com/2026/04/mythos-and-cybersecurity/
