Anthropic to Open Mythos AI to EU's ENISA

The European Commission and ENISA are close to gaining access to Anthropic's Mythos AI model through Project Glasswing so EU cybersecurity actors can study its AI-driven vulnerability discovery and exploitation capabilities; Mythos has reportedly found thousands of vulnerabilities and can autonomously generate exploit chains, raising dual-use concerns about accelerating attacks even as Anthropic vets limited partners and negotiates safe usage terms.