A US federal judge has temporarily halted the Pentagon from branding Anthropic, a leading AI company, as a supply-chain risk, undermining the Trump administration’s efforts to sever the firm's government contracts. The injunction gives Anthropic some breathing room as its legal fight over how its AI tech is used in military operations heats up.
Legal Battle Over AI and National Security
Anthropic, known for its advanced AI model Claude, has been at the centre of a fierce dispute with the US Department of Defense. The Pentagon, under the Trump administration, accused the company of posing a supply-chain risk, leading to directives aimed at phasing out Anthropic’s technology from federal agencies. Usually, only foreign adversaries get this label, so applying it to a US company shows how serious the government is.
But on Thursday, Judge Rita Lin of the Northern District of California sided with Anthropic, issuing a preliminary injunction against the Department of Defense’s order. She said the designation was probably unlawful and arbitrary because the government didn’t have a solid reason to call Anthropic a threat just because it wanted to limit how its AI was used.
"Defendants’ designation of Anthropic as a ‘supply chain risk’ is likely both contrary to law and arbitrary and capricious," Lin wrote. The judge's decision effectively restores the status quo as it stood before the Trump administration's directives on 27 February, allowing Anthropic to continue servicing existing customers while the legal dispute proceeds.
Origins of the Dispute
The Pentagon’s concerns stem from Anthropic’s attempts to limit how its Claude AI is deployed, especially regarding mass surveillance and autonomous weapons. The company demanded assurances that its technology wouldn't be used for these purposes, clashing with the Department of Defense’s stance that Claude must be available for all lawful uses without restrictions.
The situation escalated fast. President Trump ordered all federal agencies to immediately cease using Anthropic’s technology, while Defense Secretary Pete Hegseth moved to officially classify the company as a supply-chain risk. This label effectively bars Anthropic from current and future defence contracts and forces military suppliers to certify they don't use its AI.
Anthropic fought back by filing two lawsuits against these moves. The company argues that the government's punitive measures violate constitutional protections, including the First Amendment’s safeguard of free speech and the Fifth Amendment's due process clause.
The firm claims the administration is retaliating against it for speaking out about AI safety and responsible use.
Wider Industry Support and Implications
This case drew attention from more than just Anthropic. More than three dozen AI experts from firms like OpenAI and Google, including Google's chief scientist Jeff Dean, have voiced support for Anthropic’s position, filing an amicus brief highlighting the risks posed by the government’s approach to AI regulation and security.
Anthropic’s lawsuit emphasises the economic impact of the Pentagon’s actions. With contracts cancelled and future deals at risk, the company faces significant financial strain. Its reputation within the AI sector and among private customers has also suffered, as the supply-chain risk label carries heavy stigma.
Still, the Pentagon can drop contracts with Anthropic or switch to other AI providers, as long as they don’t use the supply-chain risk label as the reason. The injunction doesn't oblige the Department of Defense to continue using Claude but prevents them from punishing Anthropic solely based on the disputed label.
What’s Next?
The ruling is temporary and starts in a week, giving the government time to appeal.
Anthropic’s second lawsuit, focused on a separate legal basis for the sanctions, awaits judgement from a federal appeals court in Washington, DC. The outcome of these cases will shape how AI technologies are regulated and integrated into national security frameworks.
Meanwhile, the Pentagon continues to deploy Claude AI in ongoing operations, including during heightened tensions with Iran, underscoring the strategic importance of the technology despite the legal wrangling.
As the dispute unfolds, the case raises broader questions about government power, corporate rights, and the future of AI in defence. Can companies impose ethical limits on military use of their technology without risking government retaliation? And how will courts balance national security concerns against constitutional freedoms and commercial interests?
Related Articles
- Musk–Altman trial seeks $134bn, could reshape OpenAI
- Meta's AI push tests ad-era profitability
- AI demand forces memory makers to reroute capacity
Judge Lin’s injunction is a rare example of the courts pushing back against the Pentagon’s control over AI vendors. But the legal and political battle over Anthropic’s fate is far from over, with significant consequences for the AI industry and national security policy still to play out in court.
This article was created with AI assistance.
