China's Supreme People's Court issues first national judicial rules on AI disputes

China's Supreme People's Court released the Opinions of the Supreme People's Court on the Trial of Cases Involving Artificial Intelligence Disputes on 7 September 2026. The document runs to five parts and 24 articles and is the first national-level judicial guidance on AI disputes in China.

Article IV addresses AI face-swapping, voice cloning and so-called "AI resurrection of the deceased". On likeness, the Opinions state that where, without the consent of the natural person, AI is used to process a name, portrait or similar attribute "to generate a virtual digital image that can identify that natural person, and this image is used and published", the People's Court shall support the resulting claims. The same approach is applied to voice: where a natural person's voice is used as training material to generate a synthetic voice that can identify that person, courts are directed to support such claims.

The Opinions do not resolve whether AI-generated works themselves qualify for copyright protection under Chinese law — a question lower courts have addressed inconsistently and which remains open at the national level.

Why it matters for digital humans: This is the clearest signal yet that identifiability, not the technique used, is the operative test in the Chinese market. A vendor building a cloned presenter, a memorial avatar or a livestream host for China now has to treat documented consent from the source performer as a product requirement rather than a contracting nicety, and the rule reaches synthetic voice as squarely as it reaches the face. The guidance is judicial direction to courts rather than new statute, so its practical bite will depend on how provincial courts apply it — and the unresolved copyright question means the ownership of the avatar asset itself stays murky.