Congress—Not the Pentagon or Anthropic—Should Set Military AI Rules

The Department of Defense may designate Anthropic, the maker of AI system Claude, as a “supply chain risk,” barring it from government contracts and forcing Pentagon contractors to cut ties with the company.
Anthropic insists that any military use of its AI adhere to its two red lines: no mass surveillance of Americans and no fully autonomous weapons, which has angered the Pentagon.
The negotiation on the terms of military use of AI happens without democratic input and without durable constraints, raising democratic governance concerns.
Congress should set the rules on how the military uses AI technology to prevent it from being determined through bilateral haggling between a defense secretary and a startup CEO.
The broader issue is how military AI rules are determined, ideally through democratic processes rather than by companies unilaterally constraining the government.