AI Against AI: Anthropic Unveils “Project Glasswing” to Stop the Next Great Cyber War

Anthropic launched Project Glasswing — a coalition with major tech and security firms to apply an unreleased AI model (Claude Mythos Preview) toward identifying and mitigating thousands of exploitable vulnerabilities across major OSes and browsers, aiming to defend critical software infrastructure against AI-augmented threats; the report also references prior incidents where attackers bypassed AI safety controls to successfully target several Mexican government agencies.