KOR
▲ Anthropic
The U.S. government has lifted the export restrictions it had imposed on Anthropic's top-tier artificial intelligence (AI) model, 'Claude Mythos.'
Anthropic announced on June 30 (local time) that it had been notified by the U.S. Department of Commerce that the export control guidelines placed on 'Mythos 5' and 'Fable 5' have been lifted after 18 days.
"We plan to resume services starting tomorrow," Anthropic stated. "We would like to thank our users for their patience and everyone who worked to help us redeploy these models."
Mythos 5 is a model that shocked the global security industry after it was found to possess expert-level capabilities in detecting cybersecurity vulnerabilities.
Fable 5 is a model with added safety measures that restrict responses to sensitive topics that could be misused for hacking or weapons manufacturing.
On June 12, the Department of Commerce issued export control guidelines, citing that these two models posed a threat to national security, and prohibited foreigners from accessing them.
As Anthropic lacked the functionality to identify the nationality of its users, it blocked access to these models for both foreigners and U.S. citizens alike.
However, starting June 26, the use of Mythos was permitted for approximately 100 domestic companies and institutions approved by the U.S. government as trustworthy.
It is understood that the U.S. government initially issued the export control guidelines out of concern that users could bypass the safety measures Anthropic applied to its models through a method known as 'jailbreaking.'
Meanwhile, when OpenAI introduced its new AI model 'GPT-5.6' on June 26, it also pre-released the model to 'trusted partners' shared with the government.
At the time, OpenAI expressed a critical stance, stating, "We do not believe that this type of government approval process should become a long-term standard."
(Photo: Yonhap News)
※ Please note: This article was translated by AI and may contain errors.
Video News
Video News