OpenAI strikes deal with Pentagon hours after Trump admin bans Anthropic

In a significant move late Friday, OpenAI's CEO Sam Altman revealed that the company has finalized an agreement with the Pentagon, allowing its artificial intelligence tools to be integrated into the military's classified systems. This announcement coincided with President Donald Trump's directive mandating all federal agencies to stop using AI technology developed by rival firm Anthropic, which has been labeled a 'supply chain risk' by Defense Secretary Pete Hegseth. The controversy surrounding Anthropic stems from its refusal to comply with the Pentagon's requests for tighter restrictions on its AI systems, particularly concerning their application in autonomous weapons and the surveillance of American citizens. Altman emphasized that the agreement with the Department of War (DoW) includes vital safety principles, expressly prohibiting domestic mass surveillance and ensuring human oversight in the use of force. "We are committed to building technical safeguards that ensure our models operate as intended, as per the DoW's requirements," Altman stated on social media. He also mentioned that OpenAI plans to deploy engineers to the Pentagon to monitor the safety and functionality of their AI models. OpenAI is advocating for similar terms to be extended to all AI companies, expressing a desire for a collaborative approach rather than escalating tensions through legal means. Meanwhile, Anthropic announced its intention to contest the 'supply chain risk' designation, which typically applies to firms with ties to foreign adversaries. The specifics of the agreement between OpenAI and the Pentagon remain unclear, stirring questions about what distinguishes it from the conditions requested by Anthropic. In response to the situation, Hegseth shared Altman’s announcement, while Under Secretary Emil Michael highlighted the importance of reliable partnerships in the evolving landscape of military technology.