China Rejects Anthropic CEO's Call for Worldwide Pacing of Frontier AI

Silicon Valley insiders have recently warned about potential catastrophes that could result from a loss of control over advanced artificial intelligence. [1] Anthropic chief executive Dario Amodei recently pitched a worldwide pacing of frontier AI development, stating this effort would require cooperation with China. [1] Amodei explicitly mentioned China, arguing that it is key to keep the AI lead of democracies over autocracies as large as possible. [1] To achieve this, he advocated for continued export controls on advanced chips and machinery to China. [1]

Analysts generally agree that cutting-edge AI models from the United States are still some way ahead of their Chinese counterparts. [1] The Chinese government considers artificial intelligence a core sovereignty capability and will not want to jeopardize the progress made in closing the gap with US frontier models. [1] Lian Jye Su from Omdia stated that China is definitely not receptive to a slowdown on the terms of US firms, as Beijing sees these calls as a bid to lock in American advantage. [1][2] China's foreign ministry responded to Amodei's comments by accusing him of fearmongering. [1]

The pushback from Beijing highlights a fundamental geopolitical friction in technology development, contrasting Silicon Valley's safety concerns with China's rapid advancement goals. By framing AI as a core sovereignty capability, China signals that it views the pacing proposals from executives like Anthropic's Dario Amodei not as neutral safety measures, but as strategic attempts to maintain the current US lead in frontier models. The advocacy for continued export controls on advanced chips further solidifies this technological rivalry and limits the likelihood of near-term regulatory cooperation. What the sources don't address: Whether the US government officially endorses or plans to adopt Amodei's proposal for a worldwide pacing agreement.