How Dangerous Is Anthropic’s Mythos AI?

Anthropic last month announced its new Claude Mythos Preview model is so effective at identifying software security vulnerabilities that the company will not release it to the public, restricting access to a select group of companies for internal scanning. However, the UK’s AI Security Institute found that OpenAI’s already-available GPT-5.5 model has comparable capabilities, raising questions about the actual danger posed by Anthropic’s restricted release.

Last month, Anthropic made a remarkable announcement about its new model, Claude Mythos Preview: it was so good at finding security vulnerabilities in software that the company would not release it to the general public. Instead, it would only be available to a select group of companies to scan and fix their own software. The announcement requires context—but it contained an essential truth. While Anthropic’s model is really good at finding software vulnerabilities, so are other models. The UK’s AI Security Institute found that OpenAI’s GPT-5.5, already generally available, is comparable in capability. The company Aisle ...