AI policy's new power center
Add Axios as your preferred source to
see more of our stories on Google.
Illustration: Allie Carl/Axios
The Pentagon is turning procurement into policy, staking its claim as Washington's most powerful AI player with its decision to ditch Anthropic.
Why it matters: As lawmakers debate guardrails, the Pentagon showed how it can reshape the AI industry with a single contract decision.
- The Defense Department is the federal government's biggest tech buyer, and its requirements for companies to win contracts can become de facto rules and spread beyond military systems.
- In a regulatory vacuum, defense contracts speak the loudest.
What they're saying: "The biggest question is: What kind of business partner does the government want to be?" Jessica Tillipman, associate dean for government procurement law studies at George Washington University, told Axios.
- "They need the AI companies," she said. "The government's a superpower ... but here it's trying to jam a lot of policy."
The big picture: The Trump administration has touted an anti-regulation, pro-AI innovation stance. But it has still regulated AI, just in a different way, former Office of Science and Technology Policy chief Alondra Nelson writes in Science:
- That includes, per Nelson, "intensive state intervention operating through industrial policy, trade restrictions, immigration controls, equity stakes in private firms (selected by the state), the redirection of research funding, and the strategic preemption of state authority."
Zoom in: The Pentagon's power play is rare, rests on questionable legal ground and could have regulatory implications beyond one company.
- Its designation of Anthropic as a "supply chain risk," which means companies must stop using Claude in cases directly tied to the department, is usually reserved for foreign adversaries. Anthropic is suing on the grounds that the Pentagon is violating its free speech rights and lacks congressional authority.
- It's also unclear how regulation-by-contract fits with the administration's AI action plan, which prioritizes rapid AI development and an industry-friendly approach.
- OSTP did not respond to a request for comment.
The Pentagon's actions are hitting Anthropic's contracts far beyond the government.
- At least 100 customers, from pharma to fintech companies, have asked to pause or cancel their contracts, Anthropic lawyer Michael Mongan said during a hearing this week.
- Microsoft is asking the court for a temporary restraining order. Otherwise, Microsoft argues, tech companies will have to act "immediately" to alter their products and contracts, which could potentially "hamper" soldiers.
- A hearing on whether to grant Anthropic temporary relief is set for March 24.
What we're watching: Regulation-by-contract will continue as AI companies seek out opportunities with the government, especially if new draft guidance from the General Services Administration adding "all lawful uses" language to procurement guidelines holds.
The bottom line: The Pentagon's move risks undercutting the White House's declared hands-off, pro-industry approach meant to supercharge AI growth.
- It could also create a new contract-by-contract approach to how AI is governed and controlled, leaving AI companies in the dark about how to work with the government.
