Artificial intelligence company Anthropic has temporarily withdrawn its latest AI models after receiving a directive from the U.S. government to limit access for foreign nationals due to national security concerns.
The company confirmed that it has suspended access to its newest models, Mythos 5 and Fable 5, after U.S. officials raised concerns over a potential security vulnerability that could allow users to bypass certain safety protections built into the systems.
According to reports, the U.S. government had been evaluating Anthropic’s advanced AI models for several weeks before their wider release. Officials reportedly expressed concerns about the technology’s potential misuse, particularly in areas related to cybersecurity and critical infrastructure.
Anthropic said the government’s directive had a broader impact than expected, forcing the company to temporarily disable access to the models for all users, including those within the United States.
The move has intensified tensions between the AI company and the U.S. administration over how powerful AI systems should be regulated and deployed.
In a statement, Anthropic argued that while governments should have mechanisms to address unsafe AI deployments, such actions should follow a transparent, technically grounded, and clearly defined process. The company also maintained that the identified issue represented a limited vulnerability and did not justify the withdrawal of commercially deployed models.
Anthropic warned that if similar standards were applied across the AI industry, the release of many advanced AI systems could be significantly delayed.
The company is already involved in a legal dispute with the U.S. government over separate national security concerns and has faced increasing scrutiny regarding its AI safety policies and approach to deploying advanced models.
The latest development highlights the growing debate worldwide over balancing rapid AI innovation with national security, cybersecurity, and public safety concerns. As AI systems become increasingly powerful, governments and technology companies are facing mounting pressure to establish clear rules governing the development and release of frontier AI technologies.
Industry experts say the incident could influence future regulations and strengthen calls for formal review processes before the launch of highly advanced AI models.
