Recently, cyberspace administration authorities carried out concentrated rectification targeting the use of artificial intelligence technology to impersonate public figures, cumulatively clearing more than 47,000 items of illegal and non-compliant information and handling 3961 violating accounts. This special governance focused on violations such as fake product promotion, follower-attraction and traffic diversion, fabricating rumors, and vulgar parody, and cracked down hard on four prominent forms: unauthorized impersonation of celebrities' likenesses and voices, fabricating illness rumors, publishing false emotional scripts in the first person, and generating vulgar intimate interaction imagery.
This rectification campaign is deeply embedded in the AI misconduct governance framework that cyberspace administration authorities will continue advancing in 2026. Previously, in September, the Cyberspace Administration of China had already deployed the second phase of the "Qinglang·Rectifying AI Application Misconduct" special campaign, listing impersonation of others as a core rectification target, while the special campaign on false personas in short videos also explicitly targeted malicious marketing carried out through AI-synthesized fake audio and video.
With the popularization of generative artificial intelligence technology, the proliferation of deepfakes and false personas has severely disrupted public perception and market order. This concentrated enforcement has consolidated platforms' primary responsibility and curbed the trend of technology abuse through full-chain crackdowns, carrying important industry warning and normative significance for purifying the online ecosystem, safeguarding the public's legitimate rights and interests, and guiding the artificial intelligence industry toward compliant and healthy development.