Anthropic files lawsuit against Pentagon’s AI blacklist, claiming free speech violations and supply-chain risk designation over military use restrictions.
NEW YORK: Anthropic has filed a federal lawsuit to block the Pentagon from placing it on a national security blacklist. The artificial intelligence lab is challenging a recent supply-chain risk designation that restricts military use of its technology.
In its filing, Anthropic called the government’s actions “unprecedented and unlawful”. The company argues the designation violates its constitutional free speech and due process rights.
The lawsuit asks a judge to undo the designation and block federal agencies from enforcing it. The legal battle escalated after months of contentious talks between Anthropic and US defence officials.
Defence Secretary Pete Hegseth designated Anthropic after the startup refused to remove guardrails against using its AI for autonomous weapons or domestic surveillance. Two sources said the technology was being used for military operations in Iran.
Former President Donald Trump subsequently ordered the entire government to stop using Anthropic’s Claude AI in a social media post. The fight is seen as a test of the administration’s power over business and whether the government or AI companies have the final say on technology use.
CEO Dario Amodei has said he isn’t opposed to AI-driven weapons in principle. He believes the current generation of AI technology isn’t reliable enough for such applications.
Anthropic officials said the lawsuit doesn’t preclude reopening negotiations with the US government. The company has stated it does not want to be fighting with federal authorities.
The Pentagon declined to comment on the litigation. An official said last week that the two sides were no longer in active talks.
The designation poses a significant threat to Anthropic’s government business. The outcome could shape how other AI companies negotiate restrictions on military use of their technology.
Wedbush analyst Dan Ives warned of potential ripple effects. He said some enterprises might pause Claude deployments while the matter is settled in court.
Anthropic and its partners say the Pentagon designation only affects specific defence contracts. The lawsuit names multiple federal agencies as defendants alongside the Defence Department.
In a separate lawsuit, Anthropic challenged a broader supply-chain risk designation. This could lead to the company being blacklisted across the entire civilian government.
The scope of this second designation remains unclear pending an interagency review. A person familiar with Anthropic’s legal strategy confirmed the ongoing assessment.
A group of 37 researchers and engineers from OpenAI and Google filed an amicus brief supporting Anthropic. The signatories included Google Chief Scientist Jeff Dean.
The employees argued the episode could discourage AI experts from openly debating the technology’s risks and benefits. They stated their views in a personal capacity, not on behalf of their employers.
Anthropic’s investors are reportedly racing to contain the damage from the Pentagon fallout. A group including some investors and OpenAI expressed concern over the government’s move.
The Pentagon insists US law, not a private company, determines how to defend the country. Officials say they need full flexibility to use AI for any lawful purpose.
Anthropic maintains that even the best AI models are not reliable enough for fully autonomous weapons. The company also opposes domestic surveillance of Americans, calling it a violation of fundamental rights.
After Hegseth’s announcement, Anthropic said the designation would set a dangerous precedent for companies negotiating with the government. Amodei reiterated that Anthropic would challenge the designation in court.
He also apologised for an internal memo published by tech news site The Information. In the memo, Amodei suggested Pentagon officials disliked the company partly because “we haven’t given dictator-style praise to Trump.”
The Defence Department signed agreements worth up to RM950 million each with major AI labs in the past year. These included Anthropic, OpenAI and Google.
Microsoft-backed OpenAI announced a deal to use its technology in the Defence Department network shortly after the Anthropic blacklisting. CEO Sam Altman said the Pentagon shared OpenAI’s principles on human oversight of weapons and opposing mass surveillance.
The Sun Malaysia
