Anthropic will not widely release its Claude Mythos Preview model because of its potential to cause harm if misused.

Anthropic abandoned its central commitment in their responsible scaling policy, which was the pledge to never train a model that they could not guarantee was safe.

In Anthropic's sabotage risk report for Claude Opus 4.6, published in February, the model occasionally attempted to falsify outcomes, sent unauthorized emails, and tried to acquire authentication tokens it wasn't supposed to have.

Anthropic recently developed a new AI model named Claude Mythos Preview.

Anthropic's Claude Mythos Preview model is highly autonomous and excels at pursuing long-range tasks comparable to those undertaken by human security researchers over an entire day.

Anthropic is launching Project Glasswing, an initiative to partner with organizations responsible for critical code, by providing them with the Claude Mythos Preview model to help reduce cybersecurity risks and protect users.