Anthropic Pulls New AI Models After U.S. Government Use Restrictions
Anthropic pulled access to its newly released AI models Mythos 5 and Fable 5 on June 17, 2026, after a U.S. government order that restricted who could use them.[1]
CBS reported the access change was a direct response to the government's new limits and said the outlet did not detail which agencies or user categories were affected.[1] Anthropic's other Claude models remained available to customers.
Anthropic released Claude Fable 5 and Claude Mythos 5 on June 9, 2026. Fable 5 added safeguards to block high-risk tasks such as biology and cybersecurity work, while Mythos 5 removed some guardrails and was initially offered only to vetted U.S. partners through Project Glasswing. On June 12 the U.S. Commerce Department issued an export control directive ordering Anthropic to block access to both models for any foreign national worldwide, including foreign-national employees inside the United States. Officials cited national security authorities and concerns about jailbreaks and cybersecurity vulnerabilities.
Security commentators called the directive the first-ever export ban on AI models and warned it shows how quickly a government order can cut off a single-vendor customer's access. Others said the move underscores debate over treating powerful AI as a controlled technology and the risks enterprises face when they rely on one provider.
The mainstream summary does not mention that the U.S. Commerce Department's directive was the first-ever export ban on AI models, a significant milestone in the regulation of artificial intelligence. This unprecedented action reflects a growing trend of treating advanced AI technologies with the same scrutiny as controlled weapons, as highlighted by various commentators on social media. @matthewhirschey points out that this marks a shift in how frontier AI is perceived, with the government now exercising national security authority to enforce such restrictions. Additionally, the summary overlooks the fact that Anthropic's Fable 5 and Mythos 5 were their most advanced models at the time, which further emphasizes the implications of this ban on the company's capabilities and the broader AI landscape. This context is critical, as it underscores the potential vulnerabilities associated with these technologies and the risks enterprises face when relying on a single vendor, a point echoed by @predictionguard.
Show source details & analysis (1 source)
📊 Relevant Data
The U.S. Commerce Department issued an export control directive on June 12, 2026, ordering Anthropic to suspend access to Fable 5 and Mythos 5 by any foreign national worldwide, including those inside the United States and foreign-national Anthropic employees.
Statement on the US government directive to suspend access to Fable 5 and Mythos 5 — Anthropic
The directive cited national security authorities and was linked to concerns over potential jailbreaking of the models' safety guardrails or cybersecurity vulnerabilities.
Anthropic suspends all access to Mythos model after US government bans foreign nationals use — CNN
Fable 5 and Mythos 5 were Anthropic's most advanced and powerful models at the time of the directive, with access to all other Claude models remaining unaffected.
Statement on the US government directive to suspend access to Fable 5 and Mythos 5 — Anthropic
📌 Key Facts
- On June 17, 2026, Anthropic pulled access to its newly released AI models Mythos 5 and Fable 5.
- The step followed a U.S. government directive that restricted who could use the models.
- CBS reported the access change as a direct response to the government’s new limits, though specific agencies and user categories were not detailed.
📰 Source Timeline (1)
Follow how coverage of this story developed over time