China Launches Qinglang 2026 Campaign Targeting AI Deepfakes and Fraud
Primary region China
Tags Surveillance · Policy
Regions China
China's Cyberspace Administration launched the Qinglang 2026 campaign on May 1 targeting AI-generated deepfakes, voice cloning fraud, disinformation, and unauthorized use of digital likenesses. The campaign targets AI voice cloning and face-swapping scams that surged in China, including impersonations of acquaintances and recreations of deceased individuals. Companies failing to comply with mandatory registration and security assessment requirements for generative AI services face account suspension or platform bans. The CAC's Administrative Measures for Digital Virtual Human Information Services were open for public comment until May 6.
Strategic interpretation
The Qinglang 2026 campaign reflects Beijing's dual approach to AI: promoting domestic AI industry growth while tightening control over content and applications deemed socially destabilizing. The focus on deepfakes and voice cloning addresses genuine fraud concerns but also provides a framework for suppressing AI-generated political content. The digital virtual human regulations, if adopted, would establish the world's most comprehensive governance framework for synthetic media.