Microsoft has emerged as a leading voice in the technology industry’s pushback against the Pentagon’s decision to blacklist Anthropic, filing a supporting legal brief in a San Francisco federal court that argues the move threatens critical defense and commercial technology networks. The company is joined by Amazon, Google, Apple, and OpenAI in backing Anthropic’s legal challenge, creating what observers are calling the most unified industry response to a government action in years. The case raises fundamental questions about whether the military can punish companies for setting ethical limits on how their AI is used.
Anthropic’s legal troubles began when it refused to allow its Claude AI to be used for mass domestic surveillance or autonomous weapons as part of a $200 million Pentagon contract negotiation. When talks broke down, Defense Secretary Pete Hegseth labeled the company a supply-chain risk, a designation with severe consequences that has never before been applied to an American firm. Anthropic has since filed two separate lawsuits, arguing the designation violates its constitutional rights and is being used as ideological punishment.
Microsoft has a uniquely powerful reason to back Anthropic beyond solidarity: it uses Anthropic’s AI in military systems it provides to the federal government and is embedded in the Pentagon’s $9 billion Joint Warfighting Cloud Capability contract. The company holds additional agreements worth several billion dollars more across defense and civilian agencies. Microsoft’s public statement framed the issue as one requiring cooperation between government and the private sector to achieve both technological excellence and responsible AI governance.
In its court filings, Anthropic argued that the supply-chain risk label, traditionally applied to firms with ties to China or other adversaries, was being grotesquely misused as a political weapon. The company stated that its concerns about Claude’s reliability in lethal autonomous scenarios were genuine and rooted in deep technical knowledge of the model’s limitations. The Pentagon’s chief technology officer publicly closed the door on renegotiation, stating unequivocally that there was no chance of renewed talks.
The stakes of this case extend beyond any single company. House Democrats have written to the Pentagon seeking answers about whether AI was used in a strike on an Iranian elementary school that reportedly killed more than 175 people. The legal and legislative battles now underway may collectively determine the rules by which artificial intelligence is used in warfare for decades to come.