Inside the whirlwind 24 hours that led the White House to slap export controls on Anthropic

The Trump administration imposed sweeping export controls on Anthropic following a 24-hour effort to persuade the AI company to voluntarily withdraw a newly released model that officials deemed a security threat. According to two administration officials and a senior White House official, the decision came after intense discussions between Anthropic CEO Dario Amodei and key officials, including Treasury Secretary Scott Bessent and White House Cyber Director Sean Cairncross. The White House had grown concerned about the potential risks posed by the model, prompting multiple high-level calls to address the issue before it escalated. The officials, who spoke anonymously due to restrictions on discussing the matter, confirmed that the negotiations were urgent and involved direct appeals to Anthropic to prevent further deployment. The move highlights the administration’s challenges in regulating rapidly evolving AI technologies, where risks can emerge quickly and require swift responses. The export controls signal a broader effort to manage the security implications of advanced AI models, particularly those that could be misused or exploited. Anthropic, a leading AI research lab, had released the model despite internal and external warnings about its risks, forcing the White House to intervene. The episode underscores the tension between innovation and regulation in the AI sector, where companies and governments must balance progress with safeguards. The details of the calls and the specific concerns raised by officials have not been previously disclosed, adding to the secrecy surrounding the decision. The export controls mark a significant step in the U.S. government’s approach to AI oversight, setting a precedent for how future models may be scrutinized and restricted.