WSJ: ‘Trump Administration Shuns Anthropic, Embraces OpenAI in Clash Over Guardrails’

Amrith Ramkumar, reporting for The Wall Street Journal (gift link):

Trump’s announcement came shortly before the Pentagon’s Friday afternoon deadline for Anthropic to agree to let the military use its models in all lawful-use cases, a concession the company had refused to make. “We cannot in good conscience accede to their request,” Anthropic Chief Executive Dario Amodei said on Thursday.

The company’s red lines had been domestic mass surveillance and autonomous weapons, areas the Pentagon said Anthropic didn’t need to worry about because the military would never break the law with AI. Defense Department officials said Anthropic needed to fully trust the Pentagon to use the technology responsibly and relinquish control.

OpenAI Chief Executive Sam Altman said the company’s deal with the Defense Department includes those same prohibitions on mass surveillance and autonomous weapons, as well as technical safeguards to make sure the models behave as they should. “We have expressed our strong desire to see things de-escalate away from legal and governmental actions and towards reasonable agreements,” he said, adding that OpenAI asked that all companies be given the chance to accept the same deal. [...]

Shortly after the deadline, Defense Secretary Pete Hegseth said on X that he is designating the company a supply-chain risk, impairing its ability to work with other government contractors.

My short take is that both of these are true:

  • It’s not the place of a corporation to dictate terms to the Department of Defense regarding how its product or services are used within the law.
  • It’s a preposterous, childish (and almost certainly illegal) overreaction to designate Anthropic a “supply-chain risk to national security” in this way. Grow up.

See also: Anthropic’s official response.

★ Monday, 2 March 2026