US judge blocks Pentagon from blacklisting Anthropic

Anthropic sued the US government in March after a high-stakes disagreement led to its abrupt blacklisting.

A US judge has blocked the Pentagon’s blacklisting of Anthropic’s products for government use, ruling that its actions violated the AI giant’s rights.

“The empty invocation of national security is not a blank check to punish and retaliate against government critics,” district judge Rita Lin wrote in a 59-page ruling.

Anthropic sued the US government in March after its refusal to lower guardrails on its AI models led to an abrupt ‘supply chain risk’ designation.

Anthropic, which refused to change its safeguards over use of its AI for fully autonomous weapons or for mass surveillance of US citizens, argued that the designation was unlawful and violated its free speech and due process rights.

“We do not believe this action is legally sound and we see no choice but to challenge it in court,” company co-founder and CEO Dario Amodei said at the time.

The Pentagon, while labelling Anthropic a risk, said that the US military does not allow vendors to restrict the “lawful use” of technology. This was the first time it had publicly designated a US company a supply chain risk via a statute aimed at protecting US military systems.

In yesterday’s (27 August) ruling, Judge Lin said that the Pentagon’s actions threatened to harm Anthropic and quell public debate around its actions.

“The decision to publicly broadcast Anthropic’s punishment via social media – even before the formal, administrative process of making the necessary findings to designate Anthropic a supply chain risk had begun – makes little sense except as an attempt to swiftly make a public example of Anthropic for daring to criticise the administration,” she wrote, referring to social media posts made by US president Donald Trump ordering federal agents to cease use of Anthropic’s technology, days before a formal order was issued.

Anthropic welcomed the ruling and said it remained “focused on working productively with the government to harness AI for our national security so all Americans benefit from this technology”.

Anthropic has maintained that certain uses of AI in a domestic US setting would be incompatible with democratic values. “Using these systems for mass domestic surveillance … presents serious, novel risks to our fundamental liberties”, Amodei wrote in February, following discussions with the US government.

“Frontier AI systems are simply not reliable enough to power fully autonomous weapons. We will not knowingly provide a product that puts America’s war fighters and civilians at risk,” he said.

Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.