China Drafts Sweeping Rules for AI 'Digital Humans' to Protect Minors

The Cyberspace Administration of China (CAC) released draft regulations on April 3 governing the creation, deployment, and commercial use of AI-generated "digital virtual persons," establishing one of the most detailed regulatory frameworks for synthetic human avatars anywhere in the world. The rules are open for public comment until May 6.

Key Provisions

The draft imposes mandatory labeling requirements on all digital humans, ensuring that users can distinguish AI-generated avatars from real people across social media, e-commerce, customer service, and entertainment platforms. Companies deploying digital humans must obtain explicit consent before using any individual's personal data, likeness, or voice to create an AI avatar.

The most aggressive provisions target minors. The rules would ban virtual intimate relationships for users under 18, prohibiting AI services from creating virtual family members, romantic partners, or companions designed to simulate emotional bonds with children. Services that induce excessive spending or promote harmful behavior among minors would also be prohibited, a measure aimed at the growing market of AI companion apps and virtual influencers popular with younger demographics.

Additionally, the regulations prohibit using AI avatars to bypass facial recognition or voice authentication systems, closing a loophole that security researchers had flagged as a growing threat as deepfake technology improves.

Part of a Broader Asian Wave

China's draft rules arrive amid a surge of AI regulation across the Asia-Pacific region. South Korea's AI Basic Act, Vietnam's AI Law, and Taiwan's AI Basic Act all took effect in 2026, each establishing national frameworks for governing artificial intelligence. Together, these laws represent a coordinated, if uncoordinated, regional response to the rapid commercialization of generative AI.

However, China's approach stands apart in its specificity around digital humans. While other countries have focused on broad principles like transparency and accountability, Beijing is targeting a concrete and rapidly growing market segment where the risks to consumers, particularly young users, are already visible.

Industry Impact

The regulations could reshape the business models of companies operating AI companion and virtual influencer platforms in China, a market estimated to be worth over $10 billion by 2027. Major platforms including Baidu, Tencent, and ByteDance all offer digital human services that would fall under the new rules. Compliance will require significant technical and operational changes, including age verification systems, content moderation for AI-generated interactions, and audit trails for consent management.

Industry groups have signaled they will participate in the public comment period, though few have publicly opposed the draft provisions given the political sensitivity of child safety issues in China.