About two years ago, education entrepreneur Bert Hsu experienced what he describes as a pivotal realization: his company needed to integrate artificial intelligence into its offerings or risk obsolescence. Serving affluent parents seeking overseas education for their children, Hsu observed clients arriving with insights gleaned not from rival firms but from AI chatbots. In response, he began developing his own AI study tool, adopting the approach, "If you can’t beat them, join them."
China’s rapid AI development is now regarded as second only to that of the United States, marking an intensifying contest between the two global powers as advancements in AI capabilities accelerate. While analysts generally agree that the most advanced U.S. models still hold a technological edge, growing concerns within Silicon Valley about the potential consequences of "loss of control" over AI have cast a cautious tone within the industry.
Dario Amodei, CEO of AI firm Anthropic, recently called for a coordinated international effort to pace the development of frontier AI technologies, explicitly suggesting cooperation with China. However, Beijing appears resistant to slowing progress on terms set by U.S. companies. Analysts such as Lian Jye Su from Omdia, a technology research group, said that China views AI as a core element of national sovereignty and is unwilling to hinder its advancements, interpreting calls for a slowdown as attempts to maintain U.S. dominance.
Amodei emphasized the need to preserve a lead by democracies over autocratic regimes in AI capabilities, advocating for continued export controls targeting China’s access to advanced chips and machinery. China’s foreign ministry responded by criticizing such measures as fearmongering that disrupt global AI governance.
Despite this official pushback, experts caution against interpreting China’s stance as outright rejection of concerns about AI risks. Earlier this week, China released its third iteration of an AI safety governance framework, outlining potential risks such as autonomous agents, cybersecurity threats, and the hypothetical scenario where AI systems might improve themselves independently.
Gabriel Wagner, a researcher at Concordia, a Beijing-based AI safety consultancy, noted significant common ground between China’s framework and the perspectives expressed by Amodei, counterbalancing the polarized rhetoric. This contrasts with the approach in the United States, where former President Donald Trump recently dismissed existential fears about AI as a “hoax.”
As President Trump prepares to meet Chinese President Xi Jinping in Washington next week, AI is expected to be a key topic, though analysts remain skeptical about any major agreements emerging from the talks.
China’s Communist Party has actively sought to manage AI development within strict political and operational boundaries. According to Lian Jye Su, China’s centralized control over critical infrastructure and commerce positions it better than the U.S. to respond effectively to any AI system failures. Nonetheless, Carnegie Endowment for International Peace expert Leia Wang points out that Chinese companies engage less frequently in voluntary AI monitoring and evaluation compared to their American counterparts, and she doubts either country is fully prepared for a potential loss-of-control event.
Chinese AI regulation is evolving rapidly, characterized by iterative updates. Since 2022, China has maintained an algorithm registry and recently shifted focus from controlling AI-generated content to regulating AI behavior. New national laws have been introduced to address risks such as emotional dependency on companion AI systems.
Simultaneously, China is actively promoting AI adoption across its economy, embedding AI advancement as a priority in its latest Five Year Plan. Local governments offer incentives, including funding and reduced rents, to foster tech innovation hubs. Industrial manufacturing, crucial to the nation’s economy, is also undergoing AI-driven transformation, with companies emphasizing the importance of AI reliability over abstract safety concerns.
Daily life in China reflects widespread digital integration, with comparatively less public debate over data privacy facilitating rapid AI adoption. For example, A-fu, a digital health service employing AI chatbots modeled after renowned doctors and based on clinical studies, has attracted 150 million users, signaling growing public engagement with AI technologies across various sectors.