On September 30, the Tokyo District Court issued a ruling that became Japan's first precedent for voice protection against AI. The voice was recognized as part of a person's image, subject to the right to commercial use. Until now, such protection was only extended to appearance. Formally, the actor lost the lawsuit, but the judge formulated a principle that others can refer to.
The plaintiff was 55-year-old Kenjiro Tsuda, one of Japan's most famous voice actors. Audiences know him as the voice of Kento Nanami from 'Jujutsu Kaisen' and Seto Kaiba from 'Yu-Gi-Oh!'. In November 2025, the actor demanded through the court that TikTok remove videos where, in his opinion, an AI-generated copy of his voice was used. There were 188 such videos. They were published by an anonymous account from July 2024 to September 2025. The avatar resembled a character from 'Jujutsu Kaisen', voiced by Tsuda, and the videos themselves were dedicated to urban legends, occultism, and conspiracy theories. At its peak, the account had more than 200,000 followers, and its monthly income, according to the plaintiff, ranged from 500,000 to 750,000 yen – approximately $3,200–4,800. The actor's lawyers insisted that the questionable content gained views only because it was read by the recognizable voice of a popular artist.
The defendants defended themselves on several fronts. TikTok's operator claimed that the videos featured an ordinary male voice, and the resemblance to the actor's voice was a matter of subjective perception. The channel's author assured that the model was trained on recordings of his friend. Moreover, the account was deleted on June 29. Several months after the lawsuit was filed, the defendants decided there was no longer anything to dispute. An expert analysis showed a high degree of similarity between Tsuda's voice and the neural network-generated voiceover. Nevertheless, the lawsuit was formally dismissed. Judge Aya Takahashi stated that TikTok is no longer obliged to remove videos that are no longer on the platform. The actor's lawyer, Takashi Hirano, told reporters that they would not appeal the decision and acknowledged that both sides appear to be winners in the end.
Judge Takahashi formulated a principle for resolving such disputes: a voice, like a portrait, is a symbol of personality, and its unauthorized use may violate the right to publicity if the purpose of such use is solely to profit from its commercial appeal. This logic allowed existing legal protection to be extended to voice. There is still no separate law on voice rights in Japan. The only guideline came from the Ministry of Justice, which indicated that posting videos with an AI clone of someone else's voice on social media for profit could be considered a rights violation. Now, this position has judicial confirmation that other actors, announcers, and musicians can refer to.
Related articles
