The Trump administration moved abruptly Friday to block foreign access to Anthropic's latest artificial intelligence models, reigniting a bitter clash between Washington and the San Francisco startup over AI technology controls.
The restrictions came without warning and immediately triggered a new round of recriminations between the two camps. The administration has made tightening AI export rules a priority, viewing the technology as strategically vital and concerned about its proliferation to rival nations.
Anthropic, founded by former members of OpenAI, has found itself repeatedly caught in the crosshairs of Trump administration policy. The company positions itself as a responsible player in the AI space, emphasizing safety and transparency, but that stance has not shielded it from Washington's aggressive posture on which companies can export what models and to whom.
The foreign access cutoff represents one of the administration's most direct moves yet against a major AI developer. Unlike earlier regulatory skirmishes that played out through notices and guidance documents, Friday's action was immediate and sweeping in scope.
Administration officials have signaled that AI export controls will remain a cornerstone of their technology policy going forward. Anthropic has not publicly signaled whether it will challenge the restrictions or seek exemptions.
The clash underscores a widening gap between how Washington wants to manage AI development and deployment and how the private sector prefers to operate. For Anthropic specifically, the restrictions come at a time when the company is racing to compete with rivals like OpenAI and Google in the race to build more powerful models.
Author Sarah Mitchell: "The administration is playing hardball on AI like it's a weapons system, and Anthropic got caught in the blast radius."
Comments