US authorizes Anthropic to disclose model that raised security fears

The United States government has allowed Anthropic to release its Mythos artificial intelligence model to select companies and organizations, revising licensing requirements after ordering an export block earlier this month over national security fears.

Since the export ban in early June, "Anthropic has been working with the U.S. government to address the associated risks," Commerce Secretary Howard Lutnick wrote to the company in a letter dated Friday.

In light of progress in this work, Lutnick wrote: "I have determined that adequate safeguards exist to allow certain trusted partners to access the Claude Mythos 5 Model."

The letter does not include permission for Anthropic to release Fable, a less powerful version of the Mythos.

"We have received notice from the US government that Mythos 5, our most robust cybersecurity model, may be redeployed to a small group of cybersecurity experts and infrastructure providers," Anthropic said in a statement.

"We are working to make the pool of approved providers available and restore their access to Mythos 5 as quickly as possible. We are pleased to see this progress and continue to work with the government to expand access to Mythos 5 and make Fable 5 available for general use again," the statement added.

Earlier this month, Anthropic disabled customer access to both Mythos and Fable to comply with the U.S. government's order to suspend all use by foreign nationals - including Anthropic's own employees.

The export ban and subsequent negotiations with Anthropic highlighted the lack of a consistent regulatory framework around AI, even as the technology advances rapidly and the US tries to stay ahead of global competitors like China.

ChatGPT gains new version and intensifies global competition for AI | MONEY REVIEW

Until recently, the Trump administration took a broadly liberal stance toward regulating artificial intelligence, seeking to keep U.S. companies ahead of Chinese competitors, in particular.

But the recent action against Anthropic, as well as the call for OpenAI to limit its latest model, comes after months of concern from experts that Mythos could pose a security risk, giving hackers and other bad actors access to AI that can find and exploit vulnerabilities exceptionally quickly.

The federal government's relationship with Anthropic, one of the leading AI companies in the US, has also faced other obstacles.

Earlier this year, the government classified it as a "supply chain risk," effectively blacklisting it, due to a disagreement over the military uses of Anthropic's products.

Anthropic filed a lawsuit challenging this designation and has already won at least one initial victory in the ongoing lawsuit.

See jobs that win and lose with artificial intelligence

Source: CNN