The U.S. military has signed agreements with eight major technology companies — including Google, OpenAI, Nvidia, Microsoft, Amazon, Oracle, and SpaceX — authorizing them to deploy artificial intelligence tools on classified Defense Department networks, according to multiple reports including the Financial Times and AP News.
According to The Guardian, the companies agreed to "any lawful use" of their technology under the deals. Oracle confirmed its agreement in a press release, describing it as allowing AI deployment on classified cloud networks.
One notable absence: Anthropic. According to The Guardian, the AI startup was not included in the agreements, reportedly due to a dispute with the Pentagon over concerns about potential AI misuse.
The breadth of the deals is striking. As GeekWire noted, Microsoft and Amazon's inclusion signals a push toward building what officials describe as an "AI-first military." Federal News Network reported that the agreements are specifically designed to bring commercial AI into the heart of classified government infrastructure — a significant shift from the traditionally slow, heavily regulated pace of defense technology procurement.
Meanwhile, Axios reported that Congress has stalled on legislation governing military AI, meaning these executive-branch deals are moving ahead largely without new legislative guardrails in place.
The agreements matter because they mark a formal, large-scale integration of Silicon Valley's most powerful AI systems into sensitive military operations — raising urgent questions about accountability, oversight, and the speed at which commercial AI is being woven into national security decisions.