China's cyberspace regulator cracks down on AI impersonation of public figures

According to a post on the "网信中国" WeChat official account, Chinese internet authorities have been pressuring platforms to crack down on the use of AI technology to impersonate well-known public figures for fraudulent product promotion, follower acquisition, and misleading online content. As of November 6, 2026, the coordinated campaign had resulted in the removal of over 47,000 pieces of illegal and non-compliant information and the handling of 3,961 accounts. A report by IT之家 on the same day expanded on the scope, noting that the crackdown specifically targets false endorsements, follower-drawing schemes, the fabrication of rumors, and vulgar parodies. The involved accounts have been dealt with in accordance with relevant laws and platform agreements.

In its announcement, the authorities outlined four main categories of typical violations currently being addressed. The first involves the unauthorized use of AI to mimic celebrity portraits, synthesize their voices, or steal the portraits of their family members to promote products and drive sales, with violations found across multiple accounts on platforms such as WeChat, Douyin, and Xiaohongshu. The second category consists of stealing public figure portraits and employing AI effects to fabricate rumors, such as false illness claims, to mislead the public. The third involves impersonating public figures in the first person to publish opinions, share personal experiences, or use emotional rhetoric to induce users to follow them, with such content located on platforms like Douyin, Xiaohongshu, Bilibili, and Kuaishou. The fourth type utilizes AI to generate videos of well-known figures participating in vulgar dances or intimate interactions to gain followers and generate profit.

This campaign operates within the broader framework of Chinese authorities' ongoing governance of AI-related chaos in 2026. According to prior reporting by IT之家, the Cyberspace Administration of China (CAC) launched the second phase of the "Clean Internet: Addressing AI Application Chaos" special action in September. "Impersonating others" was explicitly designated as a key priority. By early September, the CAC reported having cleared over 5.61 million pieces of illegal information and penalized more than 40,000 accounts. As of the latest announcement, the platforms have completed the phased reporting of the crackdown, all identified violating accounts have been handled, and internet authorities are continuing to advance their regulatory campaign against AI impersonation.