U.S. Department of War Seeks to Expand OpenAI and Anthropic Tools to Classified Systems, Report Says

The United States Department of War (The Pentagon) is pushing major AI companies, including OpenAI, Anthropic, Google, and xAI, to deploy their artificial intelligence tools on classified military networks without many of the standard restrictions that typically govern their use, according to a Reuters report published today.

During a White House event, Pentagon Chief Technology Officer Emil Michael told tech executives the military is aiming to make AI models available across all classification levels—both unclassified and classified domains, Reuters said it learnt from “two people familiar with the matter.” “The Pentagon is moving to deploy frontier AI capabilities across all classification levels,” Reuters reports, a defense official who requested anonymity said.

This update could mark a sharp escalation in how the U.S. military plans to use commercial AI technology. Classified networks could handle mission planning, weapons targeting, and intelligence analysis.

OpenAI agrees to fewer restrictions, Anthropic pushes back

Earlier this week, OpenAI announced a deal with the Pentagon that allowed the military to use ChatGPT on genai.mil, an unclassified network now rolled out to over 3 million civilian and military personnel.

“OpenAI will deploy a custom version of ChatGPT on GenAI.mil, approved for the Department’s unclassified work. The system runs in authorized government cloud infrastructure with built-in safety controls and protections for the Department’s highly sensitive data, giving teams secure access to generative AI,” the company said in a blog post.

Expanding to classified systems, where weapons decisions happen, would require new negotiations.

Talks between Anthropic and the Pentagon have been significantly more contentious, Reuters previously reported. According to the report, Anthropic executives have privately told military officials they don’t want their technology used to target weapons autonomously or conduct domestic surveillance.

“Anthropic is committed to protecting America’s lead in AI and helping the U.S. government counter foreign threats by giving our warfighters access to the most advanced AI capabilities,” an Anthropic spokesperson said.

“Claude is already extensively used for national security missions by the U.S. government, and we are in productive discussions with the Department of War about ways to continue that work.”