人工智能公司Anthropic起诉特朗普政府，指控五角大楼官员因其对AI安全的立场而非法报复该公司，将其列入供应链风险黑名单。
Anthropic filed two federal lawsuits on Monday against the Trump administration alleging that Pentagon officials illegally retaliated against the company for its position on artificial intelligence safety.
Defense Department officials last week designated Anthropic a supply chain risk, citing national security concerns. It followed CEO Dario Amodei's announcement that he would not allow the company's Claude's AI model to be used for autonomous weapons, or to surveil on American citizens. The lawsuit says the administration's decision to place the firm on what is effectively a blacklist that blocks Pentagon suppliers from using Claude is an attempt to punish the company over its AI guardrails.
"The federal government retaliated against a leading frontier AI developer for adhering to its protected viewpoint on a subject of great public significance — AI safety and the limitations of its own AI model — in violation of the Constitution and laws of the United States," the lawsuit states, adding that Trump officials "are seeking to destroy the economic value created by one of the world's fastest-growing private companies."
A Pentagon spokesperson declined to comment.
The lawsuits, filed in the U.S. District Court for the Northern District of California and the federal appeals court in Washington, D.C., allege the Trump administration violated the company's First Amendment rights and exceeded the scope of supply chain risk law by using the label against Anthropic. The suit is asking a federal judge to block Pentagon officials from enforcing the blacklist designation.
It's the latest turn in what has been a contentious standoff pitting the Pentagon against Anthropic over the company's safety rules that govern its powerful services.
Lawyers for Anthropic say in the suit that Claude was not developed to be used for lethal autonomous weapons without human oversight, nor to be deployed to spy on U.S. citizens, and using the tools in these ways represent an abuse of its technology.
"Allowing Claude to be used to enable the Department to surveil U.S. persons at scale and to field weapons systems that may kill without human oversight would therefore be inconsistent with Anthropic's founding purpose and public commitments," according to the suit.
Pentagon officials have disputed that the fight with Anthropic is over lethal weapons and mass surveillance, instead claiming that private companies cannot dictate how the government uses technology in scenarios like warfare and tactical operations, claiming all of its uses would be "lawful."
The supply-chain risk designation follows a meeting in February between Defense Secretary Peter Hegseth and Amodei. National security experts say such a label typically applies to foreign adversary contractors that could potentially sabotage U.S. interests. It is highly unusual, experts say, to use the blacklist against an American company.
After Pentagon officials informed Anthropic of the designation, President Trump said on social media that all federal agencies would stop using Anthropic's tools.
Anthropic was the first AI frontier lab permitted to be used by the U.S. officials on classified networks. But since the feud began, Pentagon officials have said Elon Musk's xAI and OpenAI's ChatGPT have now been cleared for use in classified systems.
The Wall Street Journal has reported that Anthropic's Claude has been used in military operations, including the raid that led to the arrest of Venezuelan leader Nicolás Maduro and for intelligence assessments and identifying targets in the U.S.'s ongoing conflict with Iran. (NPR has not independently confirmed The Journal's reporting.)
While Anthropic has strongly resisted the administration on lethal weaponry and mass surveillance, the company says in the suit that since 2024 it has partnered with national security contractors, like Palantir, to assist the government in operations including "rapid processing of complex data, identifying trends, streamlining document review, and helping government officials make more informed decisions in time sensitive situations."