The UK Government’s Mythos AI Tests Offer Insight into Cyber Threats
- UK government’s AI Security Institute (AISI) conducted an initial evaluation of Anthropic‘s Mythos Preview model, adding independent public verification to previous reports.
- AISI found that while Mythos isn’t significantly different from other recent frontier models in individual cybersecurity tasks, its ability to effectively chain these tasks into multi-step attack sequences could set it apart.
- Since 2023, various AI models have been tested through Capture the Flag challenges. AISI’s evaluations show that Mythos Preview can now complete over 85% of “Apprentice” level CTF tasks, indicating significant improvement in cybersecurity capabilities compared to earlier iterations.
Originally published at arstechnica.com. Curated by AI Maestro.
Stay ahead of AI. Get the most important stories delivered to your inbox — no spam, no noise.