Anthropic resists Pentagon demands over military AI usage

The Pentagon, which serves as the U.S. Department of Defense, is pushing artificial intelligence companies to allow the U.S. military to use their technology for all lawful purposes. However, Anthropic is pushing back against these demands, according to a report from Axios. In response to this resistance, the Pentagon is apparently threatening to pull the plug on its $200 million contract with the AI company, signaling a potential end to their commercial partnership. This contract represents a significant financial agreement between the defense agency and the AI developer.

The U.S. government is reportedly making the same demand to other AI developers, specifically OpenAI, Google, and xAI. According to an anonymous Trump administration official cited by Axios, one of these companies has agreed to the demands, while the other two have supposedly shown some flexibility. Anthropic, however, has reportedly been the most resistant to the government’s pressure, setting up a direct conflict over the terms of military deployment.

This standoff follows earlier reports of friction between the AI company and defense officials. In January, the Wall Street Journal reported that there was significant disagreement between Anthropic and Defense Department officials over how the company’s Claude models could be used. The Wall Street Journal subsequently reported that Claude was used in the U.S. military’s operation to capture then-Venezuelan president Nicolás Maduro. Anthropic did not immediately respond to TechCrunch’s request for comment regarding the ongoing dispute.

In response to the reports, a company spokesperson told Axios that the company has not discussed the use of Claude for specific operations with the Department of War. Instead, the company spokesperson stated that the discussions are “focused on a specific set of Usage Policy questions — namely, our hard limits around fully autonomous weapons and mass domestic surveillance.” This position highlights the company’s efforts to maintain its safety guidelines even when facing the potential loss of major government contracts.

Why it matters

The Pentagon’s threat to terminate its $200 million contract with Anthropic highlights a growing friction between AI labs and the U.S. government over the boundaries of military AI deployment. As Washington seeks to integrate commercial AI into national security operations, the standoff underscores the challenges of balancing state demands with the safety policies of private technology developers.