Why China is pushing back on US warnings over rapid AI development

Beijing has rejected recent U.S. calls for a slowdown in artificial‑intelligence development, arguing that such pleas are an attempt to cement America’s lead while China strives to close the gap in frontier AI capabilities. The rebuff was voiced after Anthropic chief executive Dario Amodei urged “worldwide pacing of the frontier” and explicitly said cooperation with China would be essential to keep democracies ahead of autocracies; Lian Jye Su of Omdia noted China is “definitely not receptive” to a slowdown on U.S. terms. The Chinese foreign ministry dismissed Amodei’s essay as “fearmongering, confrontation and vicious competition,” insisting that such rhetoric would only hinder global AI governance. At the same time, China rolled out the third version of its AI safety governance framework, a detailed mapping of risks—including agent misuse, cybersecurity threats from advanced models, and the speculative prospect of recursive self‑improvement—signalling that Beijing is seeking to balance rapid development with state‑directed safety measures.

The core of the dispute lies in differing assessments of the strategic stakes and the practical ability to manage AI risks. Analysts agree that U.S. frontier models still outpace Chinese equivalents, yet the Chinese government treats AI as a “core sovereignty capability” and is unwilling to let external pressure dictate its pace. Omdia’s Su argues that Beijing views slowdown requests as a bid to lock in U.S. advantage, while Gabriel Wagner of the Beijing‑based AI safety consultancy Concordia sees more common ground between Amodei’s concerns and China’s recent safety framework, which expands focus from merely controlling AI output to regulating AI actions. Meanwhile, U.S. President Donald Trump labeled existential AI fears a “hoax,” underscoring the political divide. The Chinese Communist Party, keen to safeguard political control, has instituted an iterative regulatory regime—maintaining an algorithm registry since 2022, issuing new rules on emotional dependency on companion AIs, and shifting oversight toward what AI systems do rather than just what they say.

Despite the geopolitical tension, China continues to push AI integration across its economy, embedding the technology in its latest Five‑Year Plan and offering local incentives such as funding and low‑cost premises to attract startups. Education entrepreneur Boris But, who pivoted to AI‑driven study tools after clients began relying on chatbots, was recently invited to pitch his firm, Hampshire Education, to a city near Shanghai aspiring to become a tech hub. Industrial actors like Black Lake Technologies report that safety and controllability are “major concerns” actively discussed as factories adopt AI‑enhanced processes. While Chinese firms conduct less voluntary monitoring than their U.S. counterparts, the state’s centralized authority and rapid regulatory updates are seen by some observers as giving China a potentially stronger capacity to respond to an AI disaster, even as both nations remain unprepared for a loss‑of‑control scenario. The upcoming meeting between President Xi Jinping and President Trump in Washington is expected to feature AI on the agenda, though analysts doubt it will produce a breakthrough in aligning the two superpowers’ divergent approaches.

⚡ Effects Interpreter