The Trump administration is unlikely to broaden export restrictions recently imposed on Anthropic’s advanced AI models to rivals such as OpenAI and other leading developers, according to a government official.
This development follows the Commerce Department’s directive on June 12, 2026, which required Anthropic to suspend access to its latest flagship models, Fable 5 and the more powerful Mythos 5, for all foreign nationals. The restrictions extended even to some individuals inside the United States, including certain company employees, over national security concerns.
Anthropic responded by taking the models offline for all users worldwide, explaining that real-time segmentation by nationality was not feasible. The company had released Fable 5 earlier as a limited version of Mythos 5, which it had already restricted due to reported cybersecurity risks. Officials highlighted a potential jailbreak vulnerability that could enable advanced hacking or malicious applications. Anthropic countered that its models’ capabilities were comparable to those offered by competitors and not uniquely dangerous.
The action represents one of the most direct U.S. government interventions targeting frontier AI models themselves, moving beyond traditional controls on hardware like advanced chips. It escalates an ongoing dispute: earlier this year, the Pentagon labeled Anthropic a supply-chain risk after the company pushed back against contract terms that would expand military use, including potential applications in domestic surveillance and autonomous weapons. Anthropic is contesting that designation in court.
Despite the targeted measures against Anthropic, sources indicate the restrictions stem from company-specific issues and are not expected to expand across the industry. Other major AI firms appear positioned to continue operations without similar immediate curbs. This selective approach fits the administration’s broader AI policy, which features a new voluntary framework for reviewing high-risk systems while seeking to preserve American technological edge.
The episode has rattled the AI sector, prompting concerns about regulatory consistency for companies preparing major model releases. Anthropic has called for clearer, more transparent procedures. As global AI competition intensifies, the administration’s decision to limit the scope of restrictions underscores a strategy focused on addressing immediate risks without broadly disrupting the U.S. innovation ecosystem.