Anthropic releases safe AI model

Anthropic PBC has made a version of its Mythos AI model widely available, specifically engineered to prevent cybersecurity tasks. This release follows earlier warnings from the company about the potential for its powerful AI to identify and exploit software vulnerabilities. By blocking these capabilities, Anthropic aims to offer the advanced features of Mythos to a broader user base while mitigating significant security risks. This strategic move balances the utility of sophisticated AI with the critical need for responsible deployment.