Microsoft has anchored a massive technology industry coalition backing Anthropic’s legal challenge against the Pentagon’s supply-chain risk designation, filing an amicus brief in a San Francisco federal court that calls for urgent judicial intervention. The brief argued that the designation, which has never before been applied to a US company, threatens to destabilize critical technology supply chains relied upon by both government and commercial actors. Amazon, Google, Apple, and OpenAI have added their names to a separate supporting brief, making the case one of the most broadly backed legal disputes in technology industry history.
Anthropic’s clash with the Pentagon originated in collapsed negotiations over a $200 million contract to deploy AI on classified military systems. The company refused to agree to terms that would have allowed its technology to be used for mass domestic surveillance or to power autonomous lethal weapons, prompting Defense Secretary Pete Hegseth to label it a supply-chain risk. The designation set off a chain of events that included the cancellation of Anthropic’s existing government contracts and a public statement from the Pentagon’s technology chief that renegotiation was not an option.
Microsoft’s filing is informed by its own direct integration of Anthropic’s technology into systems it provides to the US military and its status as a partner in the $9 billion Joint Warfighting Cloud Capability contract. The company also holds additional federal agreements worth several billion dollars spanning defense, intelligence, and civilian agencies. Microsoft publicly argued that ensuring the military has access to the best AI technology and preventing that technology from being misused for surveillance or unauthorized warfare were goals that could and must be achieved together.
Anthropic’s dual lawsuits in California and Washington DC claimed that the supply-chain risk designation violated the company’s First Amendment rights by punishing it for publicly advocating responsible AI use. The company stated in court filings that it does not have sufficient confidence in Claude’s reliability in lethal autonomous warfare scenarios, which it said was the genuine and urgent reason for the restrictions it sought to impose. Anthropic noted that the Pentagon had applied a foreign adversary designation to a domestic US firm for the first time in history.
Congressional scrutiny of AI in military operations is adding another dimension to the case, with House Democrats formally asking the Pentagon whether AI was involved in a strike in Iran that reportedly killed more than 175 civilians at a school. The lawmakers specifically inquired whether human review was applied before and after any AI-assisted targeting decisions. These parallel inquiries are turning Anthropic’s legal battle into a broader national reckoning with the question of accountability in AI-driven warfare.