Trump Changes His Position on Anthropic
According to Axios, U.S. President Donald Trump said that after meeting Anthropic CEO Dario Amodei during the G7 summit, he no longer views Anthropic as a national security threat. The statement follows earlier disagreements between the U.S. government and Anthropic over the severity of “jailbreak” vulnerabilities in the Claude Fable 5 and Claude Mythos 5 models, as well as the question of how the government should intervene when large language model safety issues arise.
The dispute had already led to direct policy action. On June 12, the U.S. Department of Commerce introduced export control measures that restricted access to the related models for foreign technical personnel. Those restrictions connected model access permissions with national security review, placing Anthropic’s safety remediation process and regulatory discussions on the agenda of the White House and several federal offices.
Federal Agencies Work on a Technical Framework
After the export controls were introduced, the White House, the U.S. Department of Commerce, the Office of the National Cyber Director and Anthropic held multiple rounds of discussions around safety standards for large models. The parties began building a technical framework designed to evaluate model vulnerability risks and define the boundary for government intervention in model development and access. As described in the report, the framework focuses on risk assessment, remediation requirements and the conditions for administrative action, rather than simply closing or taking over a company.
Trump said Anthropic had actively cooperated with the requested fixes. He also emphasized that the United States’ current priority remains maintaining its competitive advantage over China in artificial intelligence, and that he does not want to restrict the domestic AI industry by shutting down or taking control of companies. Under that formulation, the government’s stance toward Anthropic has shifted from treating the company as a security threat to allowing it to continue operating while regulatory communication and safety remediation proceed.
At the same time, Trump said the government still does not rule out using emergency administrative powers, including the Defense Production Act, to regulate the company if necessary. Even though Anthropic is no longer being treated as a national security threat for now, the U.S. government is retaining authority to take stronger action on model safety, export controls and administrative oversight.