Anthropic Faces Ultimatum From Pentagon Over AI Tool

Standoff highlights rifts between Silicon Valley and Washington over AI regulation and national security

Mar. 2, 2026 at 1:39pm

Got story updates? Submit your updates here. ›

A tense standoff has unfolded between the AI startup Anthropic and the U.S. Department of Defense over the use of Anthropic's AI tool Claude. The Pentagon demanded Anthropic allow full use of Claude for any legal purpose, but the company refused. This led to threats from the Trump administration to blacklist Anthropic's technology and a move by OpenAI to deploy its models on the Pentagon's classified network. The feud has sent shockwaves through Silicon Valley as companies grapple with government control over their technologies.

Why it matters

The clash between Anthropic and the Pentagon highlights the growing tensions between Silicon Valley and Washington over the regulation and use of powerful AI technologies. It raises questions about how much control tech companies will have over their creations, especially when the government deems them critical to national security.

The details

The standoff began after a tense meeting at the Pentagon, where Defense Secretary Pete Hegseth gave Anthropic CEO Dario Amodei an ultimatum: allow full use of Claude for any legal purpose by the military, or face consequences. Anthropic refused, prompting Hegseth to threaten designating the company a 'Supply-Chain Risk to National Security,' effectively blacklisting its technology. In response, Anthropic signaled it may pursue legal action, while OpenAI reached an agreement to deploy its models on the Pentagon's classified network.

  • On Thursday at 5:36 p.m. ET, Anthropic posted a statement refusing to acquiesce to the Pentagon's demands.
  • On Friday at 3:47 p.m. ET, President Trump lashed out at Anthropic, calling it a 'RADICAL LEFT, WOKE COMPANY' and directing all federal agencies to cease use of its software.
  • On Friday at 5:14 p.m. ET, Hegseth said he would move to designate Anthropic 'a Supply-Chain Risk to National Security'.
  • On Friday at 8:24 p.m. ET, Amodei defended Anthropic's position and signaled the company may pursue legal action.
  • On Friday at 9:56 p.m. ET, OpenAI CEO Sam Altman announced his company had reached an agreement to deploy its models on the Pentagon's classified network.

The players

Anthropic

An AI startup that developed the AI tool Claude, which is at the center of the standoff with the Pentagon.

Dario Amodei

The CEO of Anthropic who refused the Pentagon's demands to allow full use of Claude for any legal purpose.

Pete Hegseth

The U.S. Defense Secretary who gave Anthropic an ultimatum and threatened to designate the company a 'Supply-Chain Risk to National Security'.

Donald Trump

The former U.S. President who lashed out at Anthropic, calling it a 'RADICAL LEFT, WOKE COMPANY' and directing federal agencies to cease use of its software.

Sam Altman

The CEO of OpenAI who announced his company had reached an agreement to deploy its models on the Pentagon's classified network.

Got photos? Submit your photos here. ›

What they’re saying

“We must not let individuals continue to damage private property in San Francisco.”

— Robert Jenkins, San Francisco resident (San Francisco Chronicle)

“Fifty years is such an accomplishment in San Francisco, especially with the way the city has changed over the years.”

— Gordon Edgar, grocery employee (Instagram)

What’s next

The judge in the case will decide on Tuesday whether or not to allow Walker Reed Quinn out on bail.

The takeaway

This case highlights growing concerns in the community about repeat offenders released on bail, raising questions about bail reform, public safety on SF streets, and if any special laws to govern autonomous vehicles in residential and commercial areas.