Anthropic has announced it will abruptly disable its most advanced AI models for all users after the US government ordered it to suspend access for foreign nationals, citing national security concerns. The company received an export control directive to block access to its Fable 5 and Mythos 5 models for all foreign nationals, without being provided specific details of the security issue.
Government Concerns Over Jailbreak Vulnerabilities
According to Anthropic, the US government believes there is a method to bypass safeguards in Fable 5, potentially allowing it to be used for identifying software vulnerabilities. The company stated that it has only been given verbal evidence of a narrow, non-universal jailbreak and disagrees that such a finding warrants recalling a commercial model deployed to hundreds of millions of users.
The order comes amid ongoing tensions between AI developers and regulators over how to assess risks from jailbreaks. Anthropic had recently called for greater US oversight of AI, including the ability to block models with unacceptable risks, but criticized Friday's action as not following principles of fair and fact-based regulation.
Background of US-Anthropic Relations
Anthropic's relationship with the US government has been strained since the company refused to allow its AI models to be used for domestic surveillance and fully autonomous weapons systems. In response, the government placed Anthropic on a supply chain blacklist, set to take effect later this year.
The directive marks an escalation in US efforts to curb foreign adversaries' AI capabilities, moving beyond controls on chips and tools to directly restricting access to AI models themselves. Anthropic confidentially filed for a US IPO last month, edging ahead of rival OpenAI.
Impact on Users and Industry
Anthropic said the net effect of the order forces it to abruptly disable Fable 5 and Mythos 5 for all customers to ensure compliance, while access to other models remains unaffected. The company believes there is a misunderstanding and is working to restore access as soon as possible.
The Pentagon's chief information officer, Kirsten Davies, expressed support for prioritizing national security over revenue cycles, stating, "Some things are simply more important than revenue cycles, clickbait and pre-IPO valuation. America First. Always."
Experts warn that Mythos models in the wrong hands could accelerate sophisticated cyber-attacks, particularly in sectors like banking. Anthropic had worked with the US government on safety ahead of the Fable launch and noted that rival models show similar capabilities in uncovering minor bugs.
