Anthropic will give defenders what its strongest security model finds — but not the model
2026-08-25AI
Claude Security now scans code with Mythos 5, the model Anthropic keeps most tightly restricted. Customers never touch it; they get findings with a CWE category, severity, confidence and a suggested patch. Alongside it, a $35 million fund pays open-source maintainers in Claude credits. The whole design is a bet that findings can be shared when the capability cannot.