A US federal judge has blocked the Pentagon’s decision to blacklist Anthropic, the company behind the Claude AI models, ruling that the government acted illegally when it designated the company as a national-security supply-chain risk.
The dispute began after Anthropic refused to allow the US military unrestricted use of its AI systems for purposes including autonomous weapons and domestic surveillance. The company argued that current AI technology was not reliable enough for autonomous weapons and raised concerns about surveillance of Americans.
Judge Rita Lin said the Pentagon's action was unlawful and lacked a sufficient basis, finding that the government had effectively punished Anthropic for its position on AI safety. Her ruling temporarily blocks enforcement of the blacklist and related restrictions on the company's government contracts.
Anthropic welcomed the ruling, saying it remained willing to work with the US government on responsible uses of AI for national security. The Pentagon has not immediately commented on the decision, and the government could challenge the ruling.