Anthropic locked down Mythos AI after US foreign-access order
What's new: The company restricted Mythos to 200 partners and then cut off Mythos 5 and Fable 5 after officials moved to bar foreign access.
Anthropic sharply restricted access to its advanced AI system Mythos after the Trump administration ordered the company on June 12 to keep its top models away from foreign nationals in and outside the United States. The San Francisco company had already limited Mythos to 200 partner organizations because it said the tool was exceptionally strong at finding software vulnerabilities that attackers could weaponize. It later shut off customer access to Mythos 5 and Fable 5, a version designed to block cybersecurity tasks. Anthropic said Mythos Preview found thousands of zero-day flaws in testing across major operating systems and web browsers, though outside researchers have not independently verified those claims.