Anthropic's Project Glasswing Aims to Strengthen Cybersecurity with AI-Powered Vulnerability Discovery

Anthropic has launched Project Glasswing, a cybersecurity initiative that brings together major technology companies to test an advanced AI system for security analysis.

The program focuses on identifying software vulnerabilities using Claude Mythos, an unreleased AI model with strong coding and reasoning capabilities.

Participating companies include Microsoft, AWS, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorgan Chase, the Linux Foundation, Nvidia, and Palo Alto Networks.

Claude Mythos has demonstrated a remarkable ability to identify software vulnerabilities during internal testing, discovering thousands of high-severity flaws in major operating systems and web browsers.