Shanghai Court Rules Against Platform in AI Voice Synthesis Lawsuit

China News Service reported that the Shanghai First Intermediate People's Court concluded a lawsuit filed by a voice actor, Ms. Wang, against a company for AI-generated voice infringement, representing Shanghai's first case regarding the protection of natural persons' voice rights involving AI-synthesized speech. To establish infringement, Wang used a judicial appraisal comparing the notarized audio with her on-site voice sample, which determined that relatively similar and highly similar matches reached 90%. The company argued the audio was generated by its AI, but training materials could not be confirmed because a former employee had departed, adding it was previously unaware of Wang's profession and had not collected or used her voice. The court held that using a natural person's voice as training material without consent to imitate tone and timbre and generate identifiable synthetic speech constitutes an infringement of voice rights. Judge Wu Dan noted that once a rights holder shows reasonable acquisition channels and high similarity, the burden shifts to the platform to prove legal data sourcing or lack of relation.

Judge Wu Dan stated that this rule for allocating the burden of proof carries general judicial significance, resolving structural evidence imbalances in artificial intelligence infringement cases and preventing platforms from evading legal responsibility by citing technical opacity or departed employees who cannot verify data.