Anthropic's Claude Mythos AI discovers critical vulnerabilities across all major operating systems and browsers, but the company responsibly shares findings with cybersecurity firms instead of exploiting them.

Claude Mythos, an AI model developed by Anthropic, has identified critical security vulnerabilities affecting every major operating system and browser

Rather than exploiting these vulnerabilities, Anthropic is launching Project Glasswing to distribute the findings to cybersecurity companies for patching

This responsible disclosure approach aims to secure global software infrastructure before attackers can weaponize the vulnerabilities

The discovery represents a significant moment in AI safety and cybersecurity cooperation between AI labs and the security community