Anthropic Says US Has Lifted Fable 5 AI Model Ban

Anthropic Says US Has Lifted Fable 5 AI Model Ban

Anthropic says the federal government has lifted export controls on two of its AI models.

“We’ve received notice that the Department of Commerce has lifted export controls on Claude Fable 5 and Mythos 5,” the artificial intelligence (AI) company announced on social media platform X late Tuesday (June 30).

“We’ll begin restoring access tomorrow, and will share an update soon. We’re grateful to our users for their patience, and to everyone who worked with us on redeploying the models.”

Anthropic introduced Fable 5 and Mythos 5 AI on June 9, saying it had developed safeguards to keep them from being misused.

Days later, the company announced it had disabled some access to the Fable 5 and Mythos 5 AI models in response to a U.S. government export control directive that cited unspecified “national security authorities.”

That directive also called on Anthropic to suspend access to those models by “any foreign national,” whether within or outside the U.S.

“Our understanding is that the government believes it has become aware of a method of bypassing, or ‘jailbreaking’ Fable 5,” Anthropic said.

While it complied with the directive, the company said it disagreed with the notion that a “narrow potential jailbreak” was enough justification for recalling a model that it had deployed to hundreds of millions of users.

In a blog post Tuesday, Anthropic said it had worked with the government to address potential cybersecurity concerns, and would make Fable 5 available Wednesday (July 1) to users globally on the Claude Platform, Claude.ai, Claude Code, and Claude Cowork.

“We will re-enable access on AWS, Google Cloud, and Microsoft Foundry as quickly as possible,” the post added.

Anthropic has also restored access to Mythos 5 “for a set of U.S. organizations,” the company said, after receiving government approval last week. The company says it will keep working with the government to offer access to the broader set of foreign and domestic partners involved in its Glasswing cybersecurity program.

In other Anthropic news, the company on Tuesday debuted a Claude Sonnet model that can make plans, use tools such as browsers and terminals, and operate autonomously at a level that a few months ago would have required larger and more expensive models.

The startup also introduced an AI workbench for scientists called Claude Science, building on Anthropic efforts in the life sciences it launched last fall.