China Rejects Anthropic AI Slowdown Proposal Amid Global Coordination Efforts

China has rejected Anthropic CEO Dario Amodei's proposal to slow down artificial intelligence development, turning a safety argument into a geopolitical issue for the $965 billion-valued private company.

The rejection was made by China's Foreign Ministry spokesperson Guo Jiakun during a regular briefing on September 14. Guo stated that parties should pursue open and inclusive AI development for the common good, rather than spreading threat narratives and engaging in confrontation.

Amodei had proposed three layers of control to ensure safe AI development: embedded outside evaluators inside frontier labs, coordination among companies in democratic countries, and eventual global coordination. China's response treats this sequence as competition dressed in governance language, making the harder stages look more remote without changing Anthropic's ability to train or sell Claude today.

For Amazon shareholders, the immediate issue is narrower: Beijing's words do not alter the AWS contract, but they weaken a premise behind the global coordination Amodei says advanced AI will require. The company has a $15 billion financing option tied to AWS milestones, which could be unlocked as compute-delivery milestones are reached.