Leading AI firm Anthropic announced on Friday that it has cut off access to its advanced Fable 5 and Mythos 5 artificial intelligence models. This drastic measure was taken to comply with an export control directive issued by the U.S. government, which cited “national security authorities.”
The company confirmed it received an urgent order at 5:21 p.m. ET, mandating the immediate suspension of access to these models by “any foreign national, whether inside or outside the United States, including foreign national Anthropic employees.” To ensure complete compliance with this directive, Anthropic made the decision to abruptly disable the models for all of its customers worldwide. The company clarified that this action does not affect its other AI models.
Dario Amodei, chief executive officer of Anthropic, at the AI Impact Summit in New Delhi, India, on Thursday, Feb. 19, 2026.
Ruhani Kaur | Bloomberg | Getty Images
This unexpected shutdown occurred just days after Anthropic proudly unveiled Fable 5 and Mythos 5, touting them as state-of-the-art models that excelled across various industry benchmarks. Fable 5 was particularly significant as it marked Anthropic's first public release of such an advanced offering, equipped with new safeguards designed to prevent responses in high-risk areas.
These recent models built upon the success of Claude Mythos Preview, which previously garnered significant attention from Wall Street and government officials for its sophisticated cybersecurity capabilities. Although not planned for general availability, Claude Mythos Preview had been selectively rolled out to a limited group of companies as part of a specialized cybersecurity initiative known as Project Glasswing.
In its statement on Friday, Anthropic expressed regret to its customers for the disruption, noting that the government had not provided specific details regarding the national security concern. The company voiced its belief in government oversight for unsafe deployments, provided it adheres to a “transparent, fair, clear, and grounded in technical facts” statutory process, and stated, “This action does not adhere to those principles.”
This incident is the latest in a series of confrontations between Anthropic and the U.S. government. Earlier this year, a public dispute with the Department of Defense (DOD) unfolded after negotiations collapsed. The DOD subsequently labeled Anthropic a supply chain risk, alleging the company poses a threat to U.S. national security – a designation typically reserved for foreign adversaries. This label now requires defense contractors to certify that they will not use Anthropic’s Claude models in military-related work.
Anthropic has since initiated legal action against the Trump administration to challenge and reverse this blacklisting, with the litigation still ongoing.
