The UK Government’s Mythos AI Tests Help Separate Cybersecurity Threat from Hype
– The UK government’s AI Security Institute (AISI) has evaluated Anthropic’s Mythos Preview model, showing it can perform around 85% of low-level cybersecurity tasks but not significantly outperforming other recent models.
– AISI’s findings suggest Mythos might excel in chaining multiple cybersecurity-related tasks into complex attack sequences, potentially setting itself apart from earlier models.
– The ongoing evaluations by the AISI highlight that while AI-driven security tools like Mythos are becoming more capable, they still face significant challenges compared to human expertise.
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.