China is moving to tighten oversight of “digital human” technologies, introducing draft rules that would require clear labeling, limit misuse, and strengthen governance of AI-generated virtual personas.
The Cyberspace Administration of China unveiled the proposal on Friday, opening it for public feedback until May 6. Under the draft, all digital human content must be clearly labeled to prevent confusion with real people. The regulations also forbid developers from using personal data without consent or creating digital humans that bypass identity checks.
To protect minors, the rules prohibit platforms from offering “virtual intimate relationships” to users under 18. Content that threatens national security, promotes separatism, or incites discrimination is also banned. Platforms are urged to restrict sexually explicit material, violent or horror content, and other harmful content.
In addition, the guidelines encourage platforms to take action when users show signs of serious distress, such as suicidal or self-harming behavior.
The move reflects China’s dual approach: encouraging AI adoption while tightening regulation to ensure technologies align with state priorities and “socialist values.” Analysts say the rules aim to close gaps in governance and set clear boundaries for the digital human industry.
According to the regulator, the governance of virtual humans has become an issue that goes beyond standard industry norms, signaling Beijing’s intent to maintain strict oversight over emerging AI applications.
