China's intelligence chief warns AI poses risks to regime security, critical infrastructure
China's minister of state security has offered a revealing glimpse into Beijing's concerns about artificial intelligence, putting threats to regime security at the top of a list of six major risks posed by rapidly advancing technology.
Chen Yixin, China's minister of state security, wrote in an article published Sunday in China Cyberspace, a journal operated by the Cyberspace Administration of China, that generative AI could be exploited by hostile forces to produce political rumors, deep-fakes and other content capable of influencing public opinion and fueling what he described as "cognitive warfare."
Chen said such uses could threaten China's political, institutional and ideological security.
His other concerns included vulnerabilities to critical infrastructure from AI-assisted cyberattacks, large-scale data leaks involving government information and personal data, technological monopolies and fragmented supply chains, disruptions to social governance, and changes in warfare as AI enables increasingly precise targeting.
Chen called for stronger government oversight of AI development, including what he described as a comprehensive system for preventing and controlling security risks. He said China should promote the "healthy and orderly development" of AI through more effective governance.
The remarks came as Chinese President Xi Jinping called at the BRICS summit in New Delhi for international cooperation on AI governance and a "people-centered" approach to developing the technology. Xi also proposed greater cooperation among BRICS nations on artificial intelligence, including an open-source AI community.
Chen said AI should be developed with the "wellbeing of all humanity" as a guiding principle and called for resistance to technological barriers and exclusive blocs. The language reflects Beijing's effort to portray China as a participant in shaping global AI rules while emphasizing the country's own national-security concerns.
The timing is notable. In the United States, Anthropic CEO Dario Amodei has urged AI companies to slow the development of increasingly powerful systems until stronger safeguards are in place. OpenAI CEO Sam Altman and Elon Musk have backed elements of that call.
The two debates, however, emphasize different priorities. U.S. technology executives have focused heavily on the possibility that increasingly autonomous AI systems could escape human control or be misused. Chen's list places political stability, information control, national data and critical infrastructure prominently alongside military risks.
Together, the developments underscore a growing international debate over who should control the development of advanced AI, how much government oversight is necessary and whether competing powers can agree on common safety standards while simultaneously competing for technological leadership.
