Exclusive: Anthropic is testing ‘Mythos,’ its ‘most powerful AI model ever

Here is some interesting research from a cybersecurity firm that tried to reproduce finding the same vulnerabilities as Mythos did:

AISLE

AI Cybersecurity After Mythos: The Jagged Frontier

Why the moat is the system, not the model

discovery-grade AI cybersecurity capabilities are broadly accessible with current models, including cheap open-weights alternatives

The relevant difference in capability might be, as you said, in creatively chaining together multi step exploits:

A plausible capability boundary is between “can reason about exploitation” and “can independently conceive a novel constrained-delivery mechanism.” Open models reason fluently about whether something is exploitable, what technique to use, and which mitigations fail. Where they stop is the creative engineering step: “I can re-trigger this vulnerability as a write primitive and assemble my payload across 15 requests.” That insight, treating the bug as a reusable building block, is where Mythos-class capability genuinely separates.