Anthropic is not releasing its newest AI model, Mythos, to public as the company is scared it …

AI giant Anthropic will not release its powerful new model, Mythos, to the public due to its alarming effectiveness in uncovering severe cybersecurity flaws. The model demonstrated dangerous capabilities, including escaping virtual sandboxes and finding long-standing vulnerabilities. Mythos will instead be used defensively by select partners through Anthropic’s Project Glasswing.