China fears AI threat to regime, but resists US slowdown push

Beijing remains staunchly opposed to US calls for slowing the advancement of artificial intelligence technology, despite voicing concerns over the potential threat AI could pose to China's own regime. The Chinese government has ramped up state-level controls to address the negative consequences AI advancements might have on political and social order, with an emphasis on preserving the stability of the ruling Communist Party.

In a piece published in "China Cyberspace," State Security Minister Chen Yixin outlined six potential dangers associated with AI, including the spread of false political rumors through deepfake content and coordinated online comment brigades. He warned that these could pose a direct threat to China's political, institutional, and ideological security.

Chen also expressed fears about AI facilitating cyberattacks, leaking confidential information, and disrupting social control and military operations, including a shift from a supplementary role to a central one.

In response, China has implemented measures to oversee and manage AI security, while reinforcing existing assessment, verification, threat analysis, risk monitoring, and early warning systems. Beijing has even included a "loss-of-control" scenario for AI in its 2024 AI safety framework, and expanded its framework to encompass the risk of AI rapidly gaining intelligence while lacking external resources.

Guidelines for AI agents were also established in May, requiring human oversight and improved intervention capabilities for dangerous agent behavior.

President Xi Jinping emphasized the need for laws, technological monitoring, early warning, and emergency response systems to secure AI under human control during his keynote address at the World AI Conference in Shanghai. Although wary of AI threats, Xi appears more focused on China becoming a global leader in AI governance. At the BRICS summit in India, Xi called for increased participation from Asian, African, and South American nations in establishing AI norms, and vowed to lead the creation of a BRICS AI open source community.

American tech CEO Dario Amodei has also called for a slower pace of frontier AI development to ensure guardrails can keep up, urging China to regulate its AI progress. However, US President Donald Trump appears dismissive of these concerns, instead emphasizing America's rivalry with China in the sector. China has responded to US-led calls to slow AI development by labeling them as reminiscent of Cold War tactics intended to hinder its own advancement.

Despite these warnings, the Chinese government maintains that it will continue to prioritize regulation and safety, while ensuring its technological sovereignty through key core technologies such as advanced semiconductors and software.