![Protester carrying a sign with a slogan at rally in support of Ukraine, outside Ohio Statehouse,[1] in Columbus, Ohio, United States, 26 February 2022 - bIMG_0195.The slogan refers to the famous state](https://upload.wikimedia.org/wikipedia/commons/6/6b/Rally_in_support_of_Ukraine_in_Columbus%2C_Ohio%2C_United_States%2C_26_February_2022_%2851906586684%29_-_edited.jpg)
Anthropic suspended access to its Claude Mythos 5 and Fable 5 AI models for all users on June 12, 2026 [1].
This sudden shutdown highlights the increasing tension between the global distribution of advanced artificial intelligence and the national security priorities of the U.S. government. Because these models are utilized by developers and enterprises worldwide, the restriction disrupts international AI workflows and underscores the U.S. government's willingness to treat high-level AI as a protected strategic asset.
The company took the action following a directive from the U.S. government to block foreign access to the specific models [1]. This order was issued based on national security concerns, though the government did not specify the exact nature of the risks associated with the Mythos 5 and Fable 5 architectures [1].
The suspension affects users across the globe, including those in Japan [1]. This broad restriction comes despite assertions from the AI developer that the tools do not pose a danger to the public.
Anthropic said there were no serious safety concerns regarding the models [1]. The company said that the current situation is the result of a misunderstanding between the private sector and government regulators [1].
A spokesperson for Anthropic said, "We believe this is due to a misunderstanding, and we are striving to resume service as soon as possible" [1].
The company has not provided a specific timeline for when the models might return to the international market. The suspension remains in effect for all users as the company works to resolve the dispute with U.S. officials [1].
“Anthropic suspended access to its Claude Mythos 5 and Fable 5 AI models for all users”
The suspension of these models suggests a shift toward 'AI protectionism,' where the U.S. government may actively restrict the export of cutting-edge weights or API access to prevent foreign adversaries or allies from gaining a technological edge. This creates a volatile environment for global businesses that rely on U.S.-based AI infrastructure, as access can be revoked instantly based on geopolitical shifts rather than technical failures.
