Anthropic regains full access to deploy its Fable and Mythos AI models after the Trump administration lifted restrictions imposed under the previous regime. The company will restore Fable availability starting July 1.
The move reverses Biden-era export controls that limited Anthropic's ability to distribute advanced AI models internationally. Those restrictions targeted frontier AI systems deemed sensitive for national security reasons, requiring special licensing for deployment outside the U.S.
Anthropic, backed by Google, Salesforce, and others, had faced constraints on distributing Claude variants to overseas customers. The restrictions created friction for the San Francisco-based AI safety company, which competes directly with OpenAI, Meta, and Mistral in the large language model space.
The Trump administration's pivot signals a shift toward looser AI governance compared to the Biden approach. The prior rules aimed to prevent strategic adversaries from accessing cutting-edge AI capabilities. Trump's team appears to prioritize American AI competitiveness over export restrictions.
For Anthropic, the lifting removes a commercial headwind. Fable and Mythos can now flow to international markets without the licensing bottleneck. This matters as the company scales Claude adoption globally and pursues partnerships across Europe, Asia, and emerging markets.
The timing aligns with broader industry expectations. Tech leaders including Sam Altman and other founders have lobbied against strict AI export controls, arguing they handicap U.S. companies competing with Chinese AI labs. The administration's move suggests those arguments gained traction.
Anthropic hasn't disclosed revenue impact from the prior restrictions, but international deployment represents a growth lever for any AI software company. Full model availability opens pathways for enterprise deals, API partnerships, and integration opportunities worldwide.
The decision also reflects confidence that Anthropic's safety practices meet national security standards. Unlike some models flagged for potential dual-use risks, Fable and