Press
Amsterdam, 9 Apr 2026. In this article published on Computable, Anthropic's Claude Mythos Preview is presented as a model capable of autonomously finding and exploiting software vulnerabilities at a scale and speed no human can match — including flaws undetected for decades. Deemed too dangerous for public release, it is being made available exclusively to select technology companies via Project Glasswing. Rewire AI security expert Nanne van 't Klooster warns that the model signals a fundamental shift: organisations must now deploy their own AI-driven security agents to keep pace with threats that operate continuously — while ensuring humans remain in the loop.
