A select group of customers is testing the Claude Mythos Preview.
This is the one of the first times an AI company has held back a model over societal risks.
Inside the alarming test results for Mythos.
The new AI model is being heralded—and feared—as a hacker’s superweapon. Experts say its arrival is a wake-up call for developers who have long made security an afterthought.
Claude Mythos is so powerful an AI, Anthropic will not release to the public, but will let organizations use it to test and patch cyber-security vulnerabilities.
Mythos is "the most psychologically settled model we have trained to date."