“`html
- The UK government’s AI Security Institute has conducted an initial evaluation of Anthropic‘s Mythos Preview model, providing independent public verification to earlier reports by Anthropic.
- This evaluation shows that while Mythos isn’t significantly different in individual cybersecurity tasks, it excels at chaining these tasks into the multi-step series needed for full system infiltration, setting it apart from previous models.
- The assessment comes as part of a broader series of Capture the Flag challenges run by the AI Security Institute since early 2023, demonstrating increasing sophistication in AI cybersecurity capabilities over time.
“`
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.