Anthropic Expands AI-Powered Security Scanning to 200 Partners, Finds 10,000+ Critical Flaws

Anthropic has expanded Project Glasswing to roughly 150 new organizations across 15 countries as of June 2, 2026. The program uses Claude Mythos Preview to scan code for vulnerabilities in critical infrastructure—power, water, healthcare, communications, and hardware. Initial partners uncovered over 10,000 high- or critical-severity flaws. New participants include vendors serving governments and over 100 million people. All organizations undergo Anthropic's security vetting before access.