IT Home On October 14, according to Bloomberg, Anthropic urged the Washington Federal Appellate Court to reconsider a previous decision that upheld the Trump administration’s classification of the artificial intelligence company as a risk to the U.S. national security supply chain.
The company applied to the full panel of judges of the appellate court on Tuesday, local time, to review the ruling made by a three-judge panel on September 25 with a 2-to-1 vote. Anthropic argued that the majority opinion misinterpreted the law incorrectly, allowing the U.S. Department of Defense to take action “without any prior determination process, and there is clear evidence that this determination is a retaliation against Anthropic’s public statements on artificial intelligence safety.”
The legal dispute between Anthropic and the government is of great importance. At a critical juncture in the development of artificial intelligence technology, the court needs to determine whether the federal government or private enterprises hold control over the security mechanisms of artificial intelligence technology. US local developers are competing to introduce the most advanced artificial intelligence models and prove to investors that this technology can be profitable.
Earlier this year, the contract negotiations between the two parties regarding the military use of the company’s artificial intelligence technology collapsed, leading to a public dispute between Anthropic and the Trump administration.
The company once requested assurances that its technology would not be used for mass surveillance of the American public, or for the deployment of autonomous weapons. However, the US side stated that, due to national security concerns, it could not accept any restrictions on the use of the technology.
The U.S. Department of Defense notified Anthropic in March that the company and its products were identified as a risk to the U.S. supply chain. All federal agencies are required to withdraw business directives related to this company. In the past, such risk identifications typically applied only to companies considered rival countries by the U.S.
Anthropic has filed two lawsuits against the government—one in California regarding the Department of Defense’s order for federal agencies to stop collaborating with the company; the other was filed with the District of Columbia Circuit Court of Appeals to challenge the legitimacy of this risk assessment.
In the San Francisco case, Anthropic won. U.S. District Judge Rita F. Lin ordered in August for the government to lift the ban on federal agencies using Anthropic technology. She emphasized that the motivation behind the Department of Defense’s action appears to be retribution, rather than legitimate security concerns.
But in this case in Washington, a panel of three judges ruled on September 25 that the government’s determination was legal.
Judge Gregory Katsas wrote a majority opinion stating that the Department of Defense “had reasonable reasons to fear that Anthropic might modify the design of the Claude model so that it could not perform national security tasks that the Department of Defense considers to be within the scope of the contract authority and necessary.”
IT Home notes that the case number is Anthropic v. U.S. Department of Defense, 26-01049, and the court hearing the case is the District of Columbia Circuit Court of Appeals (Washington).