Anthropic said it has cut off access to its AI model Claude to firms linked to the Chinese Communist Party (CCP), foregoing several hundred million dollars in revenue to protect U.S. technological advantage. CEO Dario Amodei described the move as a deliberate decision to prioritize national-security and democratic values over short-term profit.
The company said it blocked Claude usage by entities connected to the CCP, including organizations designated by the U.S. Department of Defense as Chinese Military Companies. Anthropic also reported stopping CCP-sponsored cyberattacks that attempted to abuse the system and said it has pushed for strict export controls on advanced chips to help preserve a democratic edge in AI.
Anthropic described ongoing cooperation with the U.S. Department of Defense and the intelligence community, including deployments of its models inside classified government networks and at National Laboratories. According to the company, Claude is in use across national security agencies for intelligence analysis, modeling and simulation, operational planning, and cyber operations.
The company reiterated that it will not support certain AI uses—specifically mass domestic surveillance and fully autonomous weapons—citing democratic principles and concerns about frontier AI reliability. Amodei said Anthropic resisted pressure from the Defense Department to agree to “any lawful use” and to remove specific safeguards. He added that the department threatened actions such as removing Anthropic from systems, labeling it a “supply chain risk,” and invoking the Defense Production Act, but the company declined to change its stance.
Anthropic said it remains willing to support U.S. national security efforts while keeping what it calls necessary guardrails on AI deployment.
(This story was published from a syndicated feed and is presented as received.)
