Anthropic’s most dangerous AI model just fell into the wrong hands | The Verge

The group accessed Mythos by using knowledge of Anthropic’s other model formats obtained from a recent Mercor data breach to make “an educated guess” about its online location.

Source: Anthropic’s most dangerous AI model just fell into the wrong hands | The Verge

–Maybe run your all-powerful, terrifying security AI against your own stuff so people can’t just guess a URL?

Leave a Reply