Anthropic Defies Pentagon Over AI Safeguards Friday Deadline

Anthropic has refused to bend to the Pentagon on AI safeguards, with a public feud unfolding as officials push for Friday’s deadline. According to Associated Press, Anthropic seeks narrow assurances that Claude will not be used for mass surveillance of Americans or in fully autonomous weapons, while Defence Secretary Pete Hegseth’s ultimatum warned of consequences if the company does not comply. The message from Pentagon spokesman Sean Parnell said they gave Anthropic until 5:01 p.m.

ET on Friday to decide whether to meet the demands or face actions such as designating the firm a supply chain risk. Anthropic’s chief executive Dario Amodei declared the company cannot in good conscience accede to the Pentagon’s final demand to allow unrestricted use of its technology. The dispute highlights the tension between military use of AI and corporate safeguards, with Google, OpenAI and others already involved in contracts to supply AI models to the military.

If Anthropic caves, officials warned it could jeopardise contracts; if it holds firm, the firm faces increased scrutiny and potential contractual or strategic consequences.