Judge Overturns US Government Ban on Anthropic's Claude
A federal judge has nullified a U.S. government designation labeling Anthropic a national security risk, allowing federal contractors to resume deploying the startup's Claude models.
US District Court Judge Rita Lin ruled that the Trump administration's decision to label Anthropic a supply chain risk was "arbitrary and capricious." The government had penalized the AI safety startup for its policy prohibiting the use of its Claude model in domestic surveillance and autonomous weapons. Judge Lin declared the blacklisting an unlawful retaliation that violated the First Amendment and denied the company its Fifth Amendment due process rights.
The court highlighted the contradiction in the government's stance, noting that Secretary Hegseth had recently proposed applying the Defense Production Act to Anthropic, which would treat the firm as essential to national security. Furthermore, federal agencies are actively discussing collaboration with Anthropic on its upcoming model, Mythos. Judge Lin also pointed out that the government's security fears lacked technical merit, as Anthropic lacks any technological backdoor to access, alter, or disable Claude once it is deployed in a government environment.
For enterprise technology leaders and government contractors, this ruling provides immediate legal clearance to resume paused initiatives involving Anthropic's technology. Industry analysts suggest that CIOs and CISOs should evaluate frontier models based on technical merits—such as cost, security, and effectiveness—rather than letting political or contractual disputes dictate risk registers. The decision establishes that commercial disagreements over model behavior cannot be used as a proxy for technical security vulnerabilities.
Although legal experts expect the federal government to appeal the decision to the U.S. Supreme Court, the ruling weakens the competitive arguments used by rival AI vendors who had capitalized on the blacklist. If the case reaches the high court, the legal battle will likely shift from the technical validity of the risk designation to whether the executive branch holds unilateral authority to declare a company a national security threat.
This is our own summary of reporting by Computerworld AI
