Anthropic chief government Dario Amodei has stated the corporate has “no alternative” however to problem in court docket the Pentagon’s formal designation of the unreal intelligence agency as a danger to US nationwide safety.
The CEO, writing in a weblog publish on Thursday, insisted nevertheless that the ruling’s sensible scope is narrower than initially urged, signaling that the designation wouldn’t have a catastrophic impact on the corporate.
Amodei stated the Division of Conflict, the identify most popular by the Trump administration for the Division of Protection, confirmed in a letter that Anthropic and its merchandise, together with its widely-used Claude AI mannequin, have been deemed a provide chain danger.
It’s the first time a US firm has ever been publicly given such a designation, a label usually reserved for organisations from overseas adversary international locations, like Chinese language tech firm Huawei.
Amodei, in his weblog publish, stated the corporate disputes the authorized foundation of the motion however sought to reassure prospects.
“It plainly applies solely to the usage of Claude by prospects as a direct a part of contracts with the Division of Conflict, not all use of Claude by prospects who’ve such contracts,” he wrote.
The designation would require defence distributors and contractors to certify that they do not use Anthropic’s fashions of their work with the Pentagon.
However Amodei argued that below the related statute, the intention is “to guard the federal government quite than to punish a provider” and requires the Pentagon to make use of “the least restrictive means obligatory.”
Microsoft, certainly one of Anthropic’s largest companions, agreed with that studying, concluding that Anthropic merchandise can stay accessible to its prospects aside from the Division of Conflict.
Google and Amazon Net Providers (AWS), the opposite main cloud giants on which Anthropic’s Claude is commonly delivered to companies prospects, additionally stated they have been standing by the corporate’s merchandise apart from US navy use.
The dispute erupted after Anthropic infuriated Pentagon chief Pete Hegseth by insisting its expertise shouldn’t be used for mass surveillance or absolutely autonomous weapons techniques.
Washington hit again, saying the Pentagon operates inside the legislation and that contracted suppliers can’t dictate phrases on how their merchandise are used.
Amodei additionally used the assertion to apologise for an inside firm memo leaked to the press this week, through which he advised employees the actions towards the corporate have been politically motivated.
“The actual causes” the Trump administration “don’t like us is that we have not donated to Trump (whereas OpenAI/Greg have donated so much),” Amodei stated, referring to Greg Brockman, the president of ChatGPT-maker OpenAI, who has donated $25 million to Trump.
Amodei known as the memo an “out-of-date evaluation of the present state of affairs,” written below duress on a day that noticed his firm below excessive strain from the federal government.
OpenAI initially swooped in to interchange Anthropic in its contract with the US navy, however that transfer backfired when senior OpenAI employees expressed discomfort with the deal.
OpenAI CEO Sam Altman later stated the deal was “sloppy” and that he was working to revise it.
The standoff with the Pentagon has had some silver lining for Anthropic, which was based in 2021 by former staffers of OpenAI, with a deal with AI security.
The battle has helped propel the Claude app to the highest of obtain rankings on Apple and Google smartphones.
Anthropic additionally indicated to AFP that the variety of paying customers of its Claude mannequin had doubled because the starting of the yr and that its app is presently downloaded greater than one million occasions a day.
Revealed – March 07, 2026 10:24 am IST