US Government Bans Foreign Access to 'Mythos', Blocks Use by Foreign Nationals Within the Country

Anthropic: "Appears to Stem from a Misunderstanding"

Immediate Suspension of Fable 5 and Mythos 5 Services

The U.S. administration has imposed a complete ban on foreign nationals accessing Anthropic’s top-tier artificial intelligence (AI) models, including 'Mythos'.

According to Yonhap News on June 13, Anthropic announced the previous day, "The U.S. government, in accordance with national security authorities' guidelines, has issued an export control directive suspending all foreign nationals' access to 'Fable 5' and 'Mythos 5' entirely."

Anthropic logo. Reuters and Yonhap News

The directive bans not only overseas access to Mythos, but also prohibits foreign nationals residing in the United States from accessing the models. Foreign employees of Anthropic are also subject to these restrictions.

To comply with the regulation, Anthropic announced that it has immediately suspended Fable 5 and Mythos 5 services for all customers. Anthropic stated, "We received these guidelines from the government at 5:21 p.m. Eastern Time on the 12th," adding, "The letter did not specify concrete details regarding the national security concerns."

However, Anthropic speculated that the government appears to believe users can bypass Fable 5’s safety mechanisms—which are designed to prevent misuse such as malicious hacking or biochemical weapon creation—by employing a method commonly referred to as 'jailbreaking.' Anthropic commented, "We believe this is an issue stemming from a misunderstanding," and added, "We will do our best to restore the service as quickly as possible."

Anthropic further explained that the jailbreaking methods identified by the government are widely used not only on its models but also on others, including OpenAI’s GPT-5.5, and that cybersecurity professionals utilize them daily. Anthropic stated, "While we believe the government should have the authority to block unsafe AI deployment through transparent, fair, and technically sound legal procedures, this particular action does not align with those principles."

According to the U.S. online media outlet Axios, citing government sources, the Department of Commerce decided to take this action due to security risks after a company claimed it had breached Mythos’s security barriers. The sources said the government previously tried to temporarily halt the launch of Anthropic’s model but, failing that, issued an export control notice. The access restrictions are expected to remain in place for several weeks until U.S. national security protocols are further strengthened.

Mythos has triggered what is being called the "Mythos shock" in the global security industry after it was revealed to possess expert-level capabilities in detecting security vulnerabilities. Anthropic publicly announced the development of Mythos in April, but, considering the risks, initially shared it only with select organizations such as government agencies, tech firms, and financial institutions. On June 9, Anthropic released Fable 5, a model with enhanced security.

This content was produced with the assistance of AI translation services.