Draft rules to regulate digital human services

China's cyberspace regulator has released a draft regulation to strengthen oversight of digital human services, as authorities move to address risks arising from rapidly advancing artificial intelligence technologies.

The Cyberspace Administration of China said on Friday that the proposed rules, now open for public comment, aim to promote the healthy and standardized development of digital human services while safeguarding public interests and maintaining online order.

The draft defines digital humans as virtual figures that exist in nonphysical environments and simulate human appearance and behavior using technologies such as computer graphics, digital image processing and artificial intelligence. These avatars may be created based on real people or generated entirely by algorithms, and can replicate voices, actions, interactivity and even personality traits of real people.

Authorities said the technology has seen rapid growth in recent years, injecting new momentum into sectors such as public services, entertainment and commercial marketing. At the same time, it has raised concerns over misuse, including impersonation of public figures, fraud, misleading advertising and violations of personal rights.

Under the draft, service providers, users and content distribution platforms would be required to clearly label digital humans with prominent identifiers throughout their display, in line with existing rules on AI-generated content.

The proposal also sets strict limits on content and services involving minors. It bans the provision of virtual intimate relationships, such as virtual family members or romantic partners, to minors, as well as services that may induce excessive spending, encourage harmful behavior or negatively affect their mental and physical health.

In addition, the draft prohibits the creation or dissemination of content involving sexual innuendo, violence, horror or discrimination, and calls on all parties to actively prevent such material from circulating online.

The regulation places strong emphasis on protecting personal information and legal rights. It requires explicit consent before using an individual's likeness, voice or other personal data to create digital humans, and prohibits activities that infringe on rights such as reputation, privacy and intellectual property.

Experts say the draft reflects a shift toward "full-chain governance", extending responsibility beyond platforms to include technology providers and end users.

"With the lowering threshold of AI tools, ordinary users can now create and publish digital humans," Yu Xun, a professor at East China University of Political Science and Law, was quoted by China Central Television as saying. "This requires clearer allocation of responsibilities among all parties involved."

The draft also outlines a framework covering the entire life cycle of digital human services, from creation and operation to dissemination, and introduces "service users" as key actors bearing responsibilities similar to providers in areas such as content review, data handling and risk response.

Officials said the move builds on China's existing regulatory system for algorithms and AI, reinforcing mechanisms such as security assessments and content labeling.

Authorities added that the rules are designed to strike a balance between encouraging innovation and preventing abuse, supporting the use of digital humans in areas such as governance and public services while ensuring the technology is applied in a safe and ethical manner.

In a commentary regarding the draft, Du Cuilan, deputy director of the National Computer Network Emergency Response Technical Team/Coordination Center of China, highlighted that the draft measures represent a balanced approach between development and security, guiding the industry while addressing key risks.

"Digital humans are highly deceptive and realistic," Du noted. "There are significant risks of them being used to generate harmful content, spread rumors, incite crime or maliciously induce consumption."

She added that the draft sets clear safety boundaries while leaving room for industry growth within a framework of responsible innovation, emphasizing the need to safeguard national security, uphold social ethics and protect rights such as personal information, reputation and intellectual property.

The public consultation period will run until May 6, during which feedback from industry stakeholders and the public will be collected before the regulation is finalized.

lishangyi@chinadaily.com.cn