Over the weekend, the White House made what appeared to many to be a shocking move: it banned the foreign use of Anthropic’s latest AI models, including by Anthropic’s own noncitizen employees. The move has divided commentators. Those sympathetic to Anthropic, which has withdrawn consumer access to the models in response, have claimed that the White House is retaliating against the company for the restrictions it previously placed on the Pentagon around the use of its large language models for military purposes. Critics have replied that the company is simply getting what it asked for, given its recent pleas for aggressive government regulation of the industry.

The restrictions will limit the use of Mythos and Fable, two of the company’s most advanced models, to US citizens, on grounds of national security. Ironically, Anthropic has long advocated for the US government to see its trade and industrial policy as national security issues. Despite the appearance of conflict, the White House and Anthropic are less divided than they may seem.