The government's AI standoff could decide who really controls America's military tech

• Anthropic was designated a supply chain risk amid a dispute about the military's use of its tech.
• Hours later, OpenAI struck a deal with the Department of Defense to use its AI models.
• The conflict may determine who ultimately controls the country's most powerful AI models.

This week, in a dramatic escalation reshaping how artificial intelligence is integrated into national security — and who controls it — Anthropic was officially blacklisted by the Trump administration while OpenAI swooped in to win a new defense contract.

On Friday evening, the Pentagon designated Anthropic as a supply-chain risk, barring its technology from use by defense contractors after a transition period. That came just hours after President Donald Trump directed federal agencies to stop using Anthropic's AI tools, citing the company's refusal to agree to the military's proposed use of its Claude model.

Anthropic CEO Dario Amodei said he could not "in good conscience" allow the tech to be used for mass domestic surveillance or to independently direct autonomous weapons — two use cases he says violate the company's ethical guardrails.

Amid the standoff, Anthropic's rival, OpenAI, announced that it had reached a deal with the Department of Defense to deploy its own AI models in classified environments.

The dispute has turned into a broader confrontation between the Pentagon and the private AI sector — not just over military contracts, but who ultimately sets the terms for how these powerful systems are used.

A clash of principles and contracts

Anthropic argued that restrictions around its system's use for surveillance and autonomous weapons systems were not adequately reflected — or enforceable — in the government's draft contract language.

Defense officials responded that they need to be able to deploy Claude for any "lawful use" — a term that would give the military broad discretion, even though mass domestic surveillance is illegal under several statutes.

Dean Ball, senior fellow at the Foundation for American Innovation, described the standoff as "uncharted territory" rooted in competing principles: Anthropic's insistence on setting contractual limits on how its tech is used, and the Pentagon's view that defense policy should prevail over corporate priorities.

"This is a matter of principle for both sides," Ball told Business Insider.

Shortly after Anthropic was declared a supply chain risk, OpenAI published a blog post saying it had agreed with the Pentagon on a framework that explicitly includes guardrails similar to some of those sought by Anthropic.

OpenAI outlined three main "red lines" guiding its collaboration with the Department:

• No use of its technology for mass domestic surveillance;
• No use in directing autonomous weapons;
• And no use for other high-stakes automated decisions, such as "social credit" systems.

OpenAI says its agreement is structured to protect these limits through layered safeguards, and that any use involving autonomous weapons or surveillance must comply with existing statutes and Department directives.

In a series of Ask-Me-Anything-style posts, OpenAI CEO Sam Altman said he is prepared for a potential dispute over the legality of specific government requests. However, he said that OpenAI would not agree to allow the government to use its technology for mass domestic surveillance.

"I am terrified of a world where AI companies act like they have more power than the government," Altman added. "I would also be terrified of a world where our government decided mass domestic surveillance was ok."

Potentially 'existential' stakes

Legal scholars say the government's threats to enact the Defense Production Act and designate Anthropic as a supply-chain risk, in this context, are unusual.

Eric Chaffee, a business law professor at Case Western Reserve University, described the move as a "gamble," noting that the recent Supreme Court decision that overturned Trump's tariff policies has pushed back on expansive executive actions without clear statutory backing.

And while national security authorities generally enjoy deference in court, any litigation would unfold in legally unsettled terrain.

"The government, at this point, has ultimately outsourced a lot of its work to private entities," Chaffee said. "As a result of that, because of the reluctance against state-run corporations, there are going to be tensions, and ultimately, in a context like this, figuring out how a private entity interacts with the government is going to be a complex process."

For the Pentagon, removing Anthropic from military AI pipelines could cause disruption. Signum Global Advisor policy analyst George Pollack said Claude is deeply embedded in defense planning and readiness systems and that transitioning away from them risks operational friction.

He also argued that sidelining a leading US AI firm contradicts the stated goals of maintaining American leadership in the technology.

For Anthropic, the stakes are "existential," Ball warned. He said the dispute could send a chilling message to entrepreneurs about the risks of doing business with the federal government if companies can be penalized for insisting on ethical guardrails.

OpenAI's agreement with the Department appears aimed, in part, at breaking the impasse that engulfed Anthropic's negotiations. In its blog post, OpenAI said it requested that the same terms be made available to all AI labs and urged the government to resolve its dispute with Anthropic.

Representatives for OpenAI and Anthropic did not immediately respond to requests for comment from Business Insider. It was not immediately clear whether Anthropic, or any other leading AI company, had been offered similar contractual terms to those that OpenAI said it had agreed to.

Ultimately, how the conflict is resolved could influence not just one company's fortunes, but the balance of power between the US government and private AI developers in setting the rules of engagement for next-generation technologies.