China Rejects Anthropic CEO's AI Slowdown Proposal as 'Cold War Playbook'

China has sharply criticized a proposal by Anthropic CEO Dario Amodei to decelerate the pace of artificial intelligence development, framing the call for safety as a geopolitical weapon. The dispute underscores the deepening divide between the world's two largest economies over who controls the trajectory of frontier AI.

In a recent essay, Amodei argued that slowing the improvement of AI capabilities is necessary to prevent the technology from spiraling out of control. He compared the current AI race between the U.S. and China to the nuclear tensions of the Cold War, suggesting that the two nations engage in arms limitation-style negotiations to manage the risks. However, the Global Times, a Chinese state-affiliated newspaper, characterized the proposal as a "Cold War playbook" designed to preserve U.S. technological hegemony.

The Containment Strategy

The Global Times argues that Amodei's safety-centric framing is a facade for a broader containment strategy. According to the publication, the proposal includes three specific measures intended to curb China's progress: banning the sale of powerful AI chips and hardware, cracking down on "model distillation," and preventing the theft of model weights. The newspaper described this approach as a "silent AI Cold War," calling the move "hypocritical and short-sighted" while claiming its true purpose is to establish regulatory monopolies and technological barriers.

Geopolitical Context

This clash occurs against a backdrop of intensifying struggle for AI supremacy. The U.S. government has already implemented strict export controls on high-end semiconductors to limit China's compute capacity, while U.S. labs, including Anthropic, have documented significant unauthorized access attempts from Chinese firms. While U.S. leaders present the concept of "slowing down" as a global safety imperative, Beijing views these efforts as strategic attempts to freeze the existing technological gap in favor of the United States.

Industry Implications

The disagreement highlights a fundamental rift in AI governance. If safety measures are perceived as tools for economic warfare, the likelihood of genuine international cooperation on existential AI risks diminishes. By framing disarmament negotiations through the lens of national security rather than shared human safety, the two powers risk escalating the "AI Cold War," potentially leading to a fragmented global ecosystem where safety standards are dictated by political alignment rather than technical necessity.

What's Next

Observers are now watching to see if the U.S. government will formally integrate Amodei's suggestions into its trade and security policies. While the call for a slowdown has gained traction among some Western tech leaders, the lack of buy-in from China suggests that any meaningful "disarmament" treaty remains distant. The primary point of contention remains whether AI safety can be decoupled from the broader trade war over semiconductors and intellectual property.