US lifts export ban on Anthropic's most powerful AI models after security review

The US government has lifted export restrictions on Anthropic’s most advanced Artificial Intelligence (AI) models, allowing the company to restore access to its Mythos and Fable systems after a temporary ban triggered by national security concerns.

In a statement on Tuesday, Anthropic said the Department of Commerce withdrew the export controls imposed earlier this month, removing the requirement for a special licence to distribute the AI models.

“We’ve received notice that the Department of Commerce has lifted export controls on Claude Fable 5 and Mythos 5. We’ll begin restoring access tomorrow, and will share an update soon," it said.

The decision comes less than three weeks after Anthropic was ordered to suspend access to the models over concerns that advanced AI systems could potentially be misused by foreign military or intelligence agencies.

The US has intensified scrutiny of powerful AI models amid fears that rapidly advancing systems could be exploited by countries such as China and Russia for cyber operations, intelligence gathering or other sensitive activities.

Anthropic had disabled access to its Mythos 5 and Fable 5 models after the export-control order issued on June 12.

Last week, the government partially relaxed the restrictions by allowing the release of Mythos 5 to select trusted US organisations.

Commerce Secretary Howard Lutnick confirmed the reversal, saying the government had worked with Anthropic to evaluate the models and strengthen safeguards.

“Over the past two weeks, we have worked closely with Anthropic to analyze and approve Fable 5 to ensure alignment across the US Government and strengthen America’s leadership in AI,” Lutnick said in a post on X.

In a separate letter to Anthropic, he said the company had agreed to proactively identify and address security risks linked to its models, cooperate with government standards for future releases and report any malicious activity.

“Anthropic has agreed to proactively detect and address security risks associated with the models; to work diligently with the US government on protocols and standards and releases for Mythos, Fable, and future models; and to inform the US government of any malicious activity,” the Commerce Secretary added.

The restrictions followed concerns raised over possible ways to bypass the safety measures built into Anthropic’s models. US officials had said a trusted partner discovered a “jailbreak” method that could circumvent Fable’s safeguards.

Anthropic argued at the time that such bypasses were “simple” and that similar vulnerabilities existed in other publicly available AI systems.

However, the government's vetting of which companies can gain access to these models has drawn criticism.

OpenAI CEO Sam Altman last week criticised the idea of government-controlled access to AI systems, while acknowledging the importance of safety checks.

“Extensive safety testing is not a bad idea. I just don’t like the idea of the government picking the customers,” Altman said on X.

The ChatGPT maker has also delayed a wider rollout of its upcoming GPT-5.6 model after a request from the US government, limiting early access to a smaller group of vetted partners.

- Ends