Project Glasswing

Anthropic launched Claude Mythos Preview, an AI model identifying software vulnerabilities, via Project Glasswing. Initially available to select companies including Apple, Google, and Microsoft, the initiative prioritizes responsible AI deployment for defensive cybersecurity. This strategic rollout aims to preempt misuse of the model’s advanced coding and reasoning capabilities, with Anthropic committing significant usage credits to foster secure digital landscapes.