Overview
- Anthropic rejected Pentagon demands for unrestricted use of its Claude model in domestic mass surveillance and fully autonomous lethal weapons, citing democratic values and safety concerns.
- President Donald Trump ordered all federal agencies to immediately stop using Anthropic’s technology and announced a six‑month transition to unwind government reliance.
- Defense Secretary Pete Hegseth began executing threatened measures by directing Anthropic’s addition to a procurement‑restriction list and invoking a 1950‑era law as potential leverage.
- OpenAI said it reached an agreement to deploy its models on the Pentagon’s classified network with explicit prohibitions on nationwide mass surveillance and requirements for human responsibility in any use of force.
- Hundreds of employees at Google and OpenAI, along with worker groups at Amazon, Microsoft and Google, publicly backed Anthropic, while the startup called the designation legally unfounded and said it will challenge it in court.