On September 1, during the press conference of the 2026 National Cybersecurity Publicity Week, Wang Lihong, the Deputy Director of the Cybersecurity Coordination Bureau at the Cyberspace Administration of China, shared insights with CCTV reporters. He highlighted that the artificial intelligence sector is currently grappling with five significant security risks and challenges:

Firstly, inherent technological vulnerabilities pose a threat to the stability and reliability of AI outputs. Secondly, the rapid advancement in model capabilities is disrupting traditional security frameworks, thereby exposing risks of extreme loss of control. Thirdly, the swift evolution of AI application forms is accelerating the emergence of new security risks. Fourthly, there are notable risks associated with the misuse, abuse, and malicious application of AI, which have implications for social order and ethical standards. Lastly, the global AI landscape is confronted with the threat of technological hegemony.