The UK Gov’s Mythos AI Tests Help Separate Cybersecurity Threat from Hype
- The UK government’s AI Security Institute (AISI) has published an initial evaluation of the Anthropic model Mythos Preview, adding independent public verification to earlier reports.
- AISI findings indicate that while Mythos isn’t significantly different in tests of individual cybersecurity-related tasks compared to other recent models, it could potentially set itself apart through its ability to effectively chain these tasks into multistep series necessary for full system infiltration.
- Since 2023, AISI has been evaluating various AI models through Capture the Flag challenges. The performance of subsequent models has improved steadily; Mythos Preview can now complete more than 85 percent of Apprentice-level CTF tasks, demonstrating significant progress in cybersecurity capabilities.
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.