The UK Government’s Mythos Tests Help Separate Cybersecurity Threat from Hype
- The UK government’s AI Security Institute (AISI) has published an initial evaluation of Anthropic’s Mythos Preview model, adding independent public verification to Anthropic’s claims.
- While Mythos isn’t significantly different in tests of individual cybersecurity-related tasks compared to other frontier models, it could set itself apart through its ability to effectively chain these tasks into the multistep series necessary for full system infiltration.
- AISI’s evaluation follows a long-standing effort to put various AI models through Capture the Flag (CTF) challenges since early 2023. Mythos Preview can now complete north of 85 percent of Apprentice-level CTF tasks, marking significant improvement over previous models.
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.