The stakes in this case, how much the government could punish a company for not playing ball, were evident from the start. Anthropic attracted a number of senior supporters who were possibly bedfellows among them, including ex- The authors President Trump’s AI Policy.
But 43 pages of Judge Rita Lynn opinion It turns out that what is really a contract dispute need not reach such a frenzy. This was done because the government ignored it. Current process For how such controversies are managed and the fire is fueled by social media posts by officials that will eventually contradict the positions taken in court. The Pentagon, in other words, wanted a culture war (on top of the actual war in Iran that began hours later).
According to court documents, the government used Anthropic’s Claude without complaint for most of 2025, while the company ran a branding tightrope as a security-oriented AI company that also won defense contracts. Defense employees accessing it through Palantir were required to accept the terms of the government’s specific use policy, which Anthropic co-founder Jared Kaplan said. said “America’s mass surveillance and lethal autonomous warfare” (Kaplan’s statement to the court did not include policy details). Only when the government decided to enter into a direct contract with Anthropic did differences begin.
What drew the judge’s ire is that when these disagreements became public, he had to do more with punishment than with severing ties with Anthropic. And they had a pattern: Tweet first, lawyer later.
President Trump’s Feb. 27 post on Truth Social referred to “leftwing nutjobs” at Anthropic and directed every federal agency to stop using the company’s AI. That was echoed shortly after by Defense Secretary Pat Hegseth, who said he would direct the Pentagon to label Anthropic a supply chain threat.
Doing so requires the secretary to perform a specific set of actions, which the judge found Hegseth did not complete. For example, letters to congressional committees say that less drastic measures have been considered and are not considered feasible, without providing further details. The government also argued that the designation as a supply chain threat was necessary because Anthropic could implement a “kill switch,” but its lawyers later conceded it had no evidence of that, the judge wrote.
Hegseth’s post also stated that “no contractor, supplier, or partner that does business with the United States military may conduct any commercial activity with Anthropic.” But the government’s own lawyers conceded Tuesday that the secretary did not have the authority to do so, and agreed with the judge that the statement had “absolutely no legal effect.”
The offensive posts also led the judge to conclude that Anthropic had solid grounds to complain that its First Amendment rights had been violated. The government, the judge wrote, citing the posts, “set out to publicly chastise Anthropic for its ‘ideology’ and ‘rhetoric,’ as well as its ‘arrogance’ for its unwillingness to compromise those beliefs.”