Judge Blocks Pentagon’s AI Company Ban, Citing First Amendment Concerns

A federal judge in San Francisco has temporarily blocked the Trump administration’s recent decision to label Anthropic, an artificial intelligence company, as a “supply chain risk.” The injunction, issued Thursday by Judge Rita F. Lin, also halts a directive ordering federal agencies to cease using Anthropic’s Claude AI model. This legal challenge stems from a dispute over the military’s potential employ of the technology and Anthropic’s concerns about its application, particularly regarding autonomous weapons, and surveillance.

The core of the conflict centers on Anthropic’s stated unwillingness to allow its Claude model to be used in the development of autonomous weapons systems or for domestic surveillance purposes. The Pentagon, however, maintains that it should have the discretion to determine how purchased technologies are utilized, even if that differs from the vendor’s preferred guidelines. This disagreement escalated when President Trump ordered all federal agencies to stop using Claude, followed by the Pentagon’s designation of Anthropic as a supply chain risk earlier this month. The Pentagon stated this designation was necessary to protect national security.

A Designation Usually Reserved for Adversaries

Judge Lin’s order raises significant questions about the appropriateness of applying the “supply chain risk” designation to an American company. She noted in her ruling that this label is typically reserved for foreign intelligence agencies or entities considered direct threats to national security, not domestic technology firms. The judge questioned whether the government’s actions were genuinely motivated by national security concerns or were, instead, a form of retaliation against Anthropic for publicly voicing its objections to certain applications of its AI technology. She wrote that the government’s actions “appear designed to punish Anthropic.”

The legal proceedings highlight a growing tension between the rapid development of artificial intelligence and the need for clear guidelines and oversight regarding its use, particularly within the defense sector. Anthropic argues that the supply chain risk designation will severely damage its business, potentially barring it from contracts with Pentagon contractors and the federal government more broadly. The company also alleges a violation of its First Amendment rights to free speech.

Broader Support for Anthropic’s Position

Anthropic is not alone in its challenge. A range of organizations have filed amicus briefs supporting its case, including Microsoft, the American Civil Liberties Union (ACLU), and a group of retired military leaders. The ACLU argues that the government’s actions set a dangerous precedent, potentially chilling open debate about the ethical implications of AI. The involvement of retired military leaders suggests concerns within the defense community about the potential consequences of restricting access to innovative AI technologies.

The Pentagon’s Perspective and Concerns

Lawyers representing the Pentagon argued in court that Anthropic’s restrictions on the use of its technology rendered it untrustworthy. They expressed concern that Anthropic could potentially update Claude in a way that could compromise national security. The Pentagon maintains it needs the flexibility to utilize AI tools as it deems necessary for defense purposes, and that vendors should not be able to dictate how their products are used. This position underscores a fundamental disagreement about the balance between vendor control and government needs in the realm of advanced technology.

A Landmark Copyright Settlement for Anthropic

This legal battle with the Pentagon unfolds against the backdrop of another significant legal development for Anthropic. In September 2025, the company reached a $1.5 billion settlement in a landmark copyright infringement lawsuit brought by a group of authors. The settlement, which is awaiting court approval, will compensate authors approximately $3,000 for each of the estimated 500,000 books used to train Anthropic’s Claude models. This case was considered a pivotal moment in the ongoing debate about the fair use doctrine and its application to generative AI systems. It suggests a potential shift towards AI companies being required to compensate creators for the use of their copyrighted works in training AI models.

What Comes Next: Awaiting Court Review and Potential Implications

The preliminary injunction issued by Judge Lin will remain in effect until the court can fully consider the merits of Anthropic’s case. The Pentagon and the Trump administration will have an opportunity to present their arguments, and Anthropic will be able to further demonstrate the potential harm caused by the supply chain risk designation. The outcome of this case could have far-reaching implications for the relationship between the government and AI companies, as well as for the development and deployment of AI technologies within the defense sector. It also sets a precedent for how companies can navigate ethical concerns related to their technologies while working with government entities. Anthropic expressed gratitude for the court’s decision, stating its continued commitment to working with the government to ensure the safe and responsible development of AI.

Founded in 2021, Anthropic has quickly turn into a prominent player in the generative AI landscape with its Claude family of large language models. As Politico reported earlier this month, this dispute with the administration could potentially undermine former President Trump’s broader AI agenda.