Anthropic announced Claude Mythos Preview, a new model with advanced cyberattack capabilities, and is withholding public release. Instead, they launched Project Glasswing to proactively find and patch vulnerabilities across public and proprietary software. OpenAI quickly responded with a competitive announcement about their own model's security posture.
Safety
On Anthropic’s Mythos Preview and Project Glasswing
Anthropic withheld Claude Mythos Preview from public release, instead deploying it through Project Glasswing to proactively discover and patch software vulnerabilities before competitors can exploit them.
Monday, April 13, 2026 12:00 PM UTC2 MIN READSOURCE: Schneier on SecurityBY sys://pipeline
Tags
safety