A federal judge ruled that the Defense Department’s attempts to punish the company behind the Claude chatbot were “illegal and baseless.”
Read more Trump Wants to Rename Lake Ontario. Here Are Other Things He’s Renamed – or Tried To
The Pentagon in March labeled artificial intelligence firm Anthropic a “supply chain risk” – a blacklist of sorts for firms doing business with the federal government – after it refused to loosen restrictions on how its technology could be used by the military.
The 59-page ruling carries broad implications for the administration, national security and the use of artificial intelligence by the government.
What Does the Judge’s Ruling Mean?
U.S. District Judge Rita Lin of the Northern District of California, an appointee of former President Joe Biden, said Thursday that the Pentagon’s designation of Anthropic was unlawful, retaliatory and unsupported by the relevant law. The judge also found First Amendment and due process problems.
The government’s actions “were based on a desire to make a public example out of Anthropic for its ‘arrogance’ in criticizing the government, not based on any articulable basis to believe that Anthropic would actually sabotage its model,” Lin wrote in the decision.
Why Was the Pentagon Angry with Anthropic?
The Pentagon wanted AI companies – not just Anthropic – to permit their systems to be used for any lawful military purpose. Anthropic objected to certain uses, particularly mass surveillance of Americans and autonomous lethal weapons. It has been fighting with the Trump administration for months, not only as it attempts to preserve its own guardrails but as it publicly criticizes the Pentagon policy.
Anthropic CEO Dario Amodei said “frontier AI systems are simply not reliable enough to power fully autonomous weapons. We will not knowingly provide a product that puts America’s warfighters and civilians at risk.”
Anthropic sued the Pentagon in March over claims the government was illegally punishing the company for not allowing the military free range with its technology.
Anthropic’s ethical principles are not unusual among AI companies, says Lauren Kahn, senior research analyst at Georgetown University’s Center for Security and Emerging Technology.
“These sorts of carve-outs or clarifications have existed as long as these companies have existed,” she says.
Kahn says Anthropic’s statements suggest “that they are not inherently opposed to working on these systems, but that they simply do not believe the technology is ready from a technical standpoint to be safely and effectively used.”
Read more British Far-Right Commentator Milo Yiannopoulos Detained by ICE
What Did the Government Do?
The Trump administration officially labeled Anthropic a supply chain risk and moved to prevent federal agencies and contractors from using its technology – a significant punishment for a company competing for government AI business.
The “supply chain risk” designation has been used by the government before, but its purpose has generally been to cut off business with foreign companies deemed national security threats.
Why Did the Judge Reject That?
The judge found that a national security designation was being used to punish Anthropic over its position in a policy dispute rather than because Anthropic actually posed the kind of security risk contemplated by the law.
“The empty invocation of national security is not a blank cheque to punish and retaliate against government critics,” Lin wrote.
In an earlier hearing in late July, Lin called the government’s position “really troubling” and said it seemed to contradict the First Amendment.
Anthropic also argued that the “supply chain risk” label did not apply to American companies.
What Happens Now?
The government is expected to challenge the ruling, and there is a related case still moving through the courts. Anthropic can, meanwhile, continue doing business with non-defense federal agencies under the ruling.
For now, the decision provides a major victory for companies seeking to preserve guardrails over how their products are used in the rapidly evolving world of artificial intelligence.
Anthropic has been in talks with the Trump administration about its new AI model for cybersecurity and biology research, called Mythos.
Kahn says Anthropic’s fight against the Pentagon has “reignited concern on both the government and private sector sides about what it means to have these partnerships, which are critical for national defense and security and for ensuring the U.S. maintains its competitive edge when it comes to warfighting capabilities.”
Read more Kalshi Cannot Block Nevada Oversight of Prediction Markets, US Appeals Court Rules
