JUST IN: Anthropic decided to limit access to Mythos, its new AI model, arguing that it could discover and exploit critical vulnerabilities in globally used software.

However, the move also sparked a broader debate about whether the company is protecting the internet or strengthening its market position against rivals and firms that practice model distillation.
Anthropic stated that Mythos is too capable of finding and exploiting security flaws, so it will only share it with large, critical infrastructure organizations.