JUNE 27, 2026
US government permits Anthropic to release Mythos 5 AI model to select partners after cybersecurity review
The US government has authorized Anthropic to release its Claude Mythos 5 AI model to a limited group of trusted partners and cyber defenders, revising an export block imposed earlier in June over national security concerns. Commerce Secretary Howard Lutnick wrote to Anthropic in a letter dated Friday stating that "appropriate safeguards are in place to permit certain trusted partners to access the Claude Mythos 5 Model." The authorization does not yet cover Fable, a less powerful version of the Mythos model.
The US government's decision follows weeks of negotiations between Anthropic and federal officials after an export ban earlier in June required Anthropic to disable customer access to both Mythos and Fable in order to comply with an order suspending all use by foreign nationals — including Anthropic's own employees.
In his letter, Commerce Secretary Lutnick wrote that "Anthropic has worked with the US government to address risks associated with the Covered Models," and that progress in that work justified the limited resumption of access. Anthropic said in a statement that it is "working to provision the approved set of providers and restore their access to Mythos 5 as quickly as possible."
Anthropic described Mythos 5 as its "strongest cybersecurity model," intended for a small group of cyber defenders and infrastructure providers. The company added that it "continue[s] to work with the government to expand access to Mythos 5 and make Fable 5 available for general use again." CNN reported, citing a source familiar with the discussions, that conversations between Anthropic and the government are expected to continue into the weekend with an eye toward restoring access to Fable.