Anthropic to Open Mythos Model Access to EU

🔑 Enhanced Key Takeaways

• •Anthropic has granted the European Union's cybersecurity agency, ENISA, access to its Mythos AI model through 'Project Glasswing,' making ENISA the first EU institution to join this controlled-access cybersecurity initiative.
• •Mythos is an advanced AI model specifically designed for defensive cybersecurity, capable of autonomously identifying and exploiting zero-day vulnerabilities in major operating systems and web browsers, having reportedly discovered over 10,000 such vulnerabilities.
• •The agreement follows weeks of contentious negotiations, including EU officials traveling to San Francisco, as the US government had previously restricted Anthropic's broader distribution plans for Mythos due to security concerns.
• •The EU AI Act, set for full enforcement in August 2026, lacks a mechanism to compel American companies to share powerful AI models with European regulators, making this agreement a significant precedent for international AI governance.
• •Anthropic launched Mythos as 'Claude Mythos Preview' in April 2026, initially as a limited, controlled release due to its powerful capabilities and potential for misuse, but now plans a wider public rollout of 'Mythos-class models' in the coming weeks.