Anthropic still on the side Pentagon reaches AI Deals with Tech Companies to use their AI for Warfare

Washington is accelerating its adoption of artificial intelligence – drawing on a broad coalition of leading technology firms to sharpen its military edge.

The US Department of Defense has struck agreements with seven technology companies to deploy their AI capabilities within classified networks. The partners include Google, Microsoft, Amazon Web Services, Nvidia, OpenAI, Reflection and SpaceX. The aim is to enhance decision-making for warfighters in complex operational environments, the Pentagon said. The move underscores Washington’s determination to secure technological superiority on the battlefield.

Concerns over autonomy and civil liberties

The Pentagon has significantly stepped up its efforts in recent years, viewing AI as a key enabler for shortening decision cycles and gaining operational advantage. At the same time, concerns are mounting. Critics warn of potential intrusions into privacy and the risk of machines selecting targets independently. Some of the agreements seek to address these issues. Provisions in certain contracts require human oversight in autonomous or semi-autonomous missions, and stipulate that AI must be used in line with constitutional protections and civil liberties.

Anthropic still sidelined amid dispute

AI firm Anthropic remains outside the Pentagon’s new framework and is now locked in a legal dispute with the Department of Defense. The company had sought assurances that its technology would not be used in fully autonomous weapons or for the surveillance of US citizens.Defence Secretary Pete Hegseth rejected such constraints, insisting the military must retain the option to use AI for any lawful purpose.Tensions deepened after President Donald Trump in a rare move blocked the use of Anthropic’s chatbot Claude across federal agencies and sought to designate the firm as a potential supply chain risk. 

OpenAI takes a central role amid multiple providers

OpenAI had already announced a deal with the Pentagon in March, effectively filling the gap left by Anthropic in classified environments. The company confirmed that the agreement remains unchanged. “We believe the people defending the United States should have the best tools in the world,” OpenAI said. While companies such as Amazon and Microsoft have long supported classified military work, others are new to the field - notably Nvidia and the start-up Reflection. Both are developing open-source AI models, a priority for the Pentagon as it seeks to establish an “American alternative” to China’s rapidly advancing AI ecosystem.

Capabilities already in use

The new tools are already being deployed. Through the platform “GenAI.mil”, military personnel, civilians and contractors are using AI applications in day-to-day operations. According to the Pentagon, processes that once took months can now be completed in days - equipping warfighters to act more quickly and with greater confidence in the face of evolving threats.