Anthropic has abruptly suspended access to Claude Fable 5, its most advanced artificial intelligence model, leaving millions of users without access to a tool the company had promoted as one of its most powerful AI systems yet.
The shutdown comes just days after the model's June 9 launch and follows a directive from the U.S. government over what Anthropic described as a national security concern.
In a statement, the company said access to both Claude Fable 5 and Mythos 5 had been disabled for all customers to comply with government instructions.
"The access to Fable 5 and Mythos 5 for all our customers was abruptly disabled to ensure compliance," Anthropic said.
For users who had only recently begun exploring the new technology, the sudden suspension offered few answers. Neither Anthropic nor the U.S. government disclosed specific details about the security concern, fueling questions about what prompted such an extraordinary intervention.
According to Anthropic, the government had identified a potential "jailbreaking" method affecting Fable 5. Jailbreaking refers to techniques used to bypass an AI model's safety restrictions, potentially enabling malicious actors to generate prohibited content or access sensitive information.
Despite complying with the order, Anthropic publicly pushed back against the government's decision.
"We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people," the company said.
The dispute highlights a growing struggle over who should have the final say over increasingly powerful AI systems. While technology companies continue racing to build more capable models, governments around the world are grappling with concerns about security, misuse, and national interests.
The suspension also comes weeks after U.S. President Donald Trump signed an executive order encouraging AI developers to provide the federal government with greater oversight of advanced AI systems.
For many users, however, the immediate issue is far simpler: a tool they had access to one day was gone the next.
As artificial intelligence becomes embedded in workplaces, schools, and daily life, the sudden disappearance of a major AI platform raises a broader question about the future of the technology: when concerns over security collide with public access, who gets to decide what remains online?