China's Diffusion‐Forward AI Strategy: The “ AI Race” in Political Economic Context

ABSTRACT

The United States and China are pursuing fundamentally different artificial intelligence strategies, a divergence that has been largely obscured by the prevailing focus on large language model competition and frontier model capabilities. Drawing on official policy documents, original registration data for generative AI services in China, and a patent‐based case study of humanoid robotics firm UBTECH, this paper documents China's “diffusion‐forward” strategy: a state‐directed effort to embed AI as a general‐purpose technology across the physical economy, with particular emphasis on manufacturing, industrial robotics, and embodied AI. We show that China's approach is enabled by distinctive political‐economic institutions—decentralized but hierarchical governance, the investor‐state model, and campaign‐style industrial policy—and is already producing measurable commercial outcomes. These findings reframe the AI competition debate: the decisive contest may be less about which country achieves artificial general intelligence first and more about which political economy can more rapidly diffuse AI into productive activity.