The Pentagon is seeking information from Boeing and Lockheed Martin regarding Anthropic’s use of the Claude chatbot. This request comes with a deadline of Friday, which could lead to either a reduction in security measures for Anthropic or potentially a ban on the company, as reported.
Axios mentions that the Pentagon recently warned of possibly categorizing Anthropic as a “supply chain risk,” a designation usually aimed at foreign entities like Huawei. This classification could terminate current contracts and compel other defense firms to cut ties with Anthropic.
Boeing indicated that it does not have any existing contracts with Anthropic, while Lockheed Martin confirmed it had been approached by the Pentagon but refrained from providing further details.
During a recent meeting, Defense Secretary Pete Hegseth reportedly cautioned Anthropic’s CEO, Dario Amodei, that restrictions on military utilization of chatbots must be lifted by 5:01 p.m. ET on Friday.
If these restrictions aren’t addressed, Hegseth warned that the Pentagon might invoke the Defense Production Act to effectively compel Anthropic to adapt Claude for military needs. Some critics have raised concerns about the potential contradictions in the “supply chain risk” designation and the intended use of the Defense Production Act.
Neither the Department of Defense nor representatives from Anthropic immediately responded to inquiries for comments.
Amodei emphasized that Anthropic does not endorse the application of its technology for mass surveillance of citizens or for autonomous weapons. He asserted that the company’s policies do not impede military operations.
The atmosphere during the meeting between Amodei and Hegseth was reportedly friendly yet tense. Hegseth praised Claude’s capabilities even as he conveyed the Pentagon’s demands.
Currently, Claude is the sole chatbot the U.S. military employs in classified contexts. There are indications that Elon Musk’s Grok AI model has recently been approved, and other chatbot systems are anticipated to join the field, providing alternatives to Claude.
A spokesperson for Anthropic noted that the company has been engaged in good faith discussions regarding its usage policy to ensure it can support the government’s national security efforts in line with its model’s reliable and ethical applications.
Earlier this year, the Department of Defense awarded Anthropic a $200 million contract for developing advanced AI capabilities that would bolster U.S. national security.
Relationships between Anthropic and the Trump administration have reportedly soured lately. The company’s connections to the Effective Altruism movement and significant Democratic backers, like Reid Hoffman, have come under scrutiny from the White House.
