A Japanese voice actor has taken TikTok to court over videos he says used an artificial intelligence-generated imitation of his distinctive voice, in a case that could have implications for how performers’ vocal identities are protected as AI voice-cloning technology advances.
Kenjiro Tsuda, a prominent voice actor known for roles in major anime franchises including Jujutsu Kaisen and Yu-Gi-Oh!, is seeking the removal of videos that he alleges feature an AI-generated version of his recognisable voice.
A verdict in the case before the Tokyo District Court is due on Wednesday.
The lawsuit is reportedly the first in Japan to specifically seek protection of an individual’s vocal identity from AI-generated replicas.
Voice Identity at the Centre of Dispute
Tsuda’s legal team argues that the videos infringed his publicity rights — rights that allow individuals to control and derive commercial value from the use of aspects of their identity.
The videos were posted by an anonymous account whose profile image resembled a character from Jujutsu Kaisen voiced by Tsuda. The account published material covering subjects including urban legends, occult phenomena and conspiracy theories, with the narration allegedly generated using an AI imitation of Tsuda’s voice.
According to court records cited by AFP, the account had more than 200,000 subscribers at one point and was estimated in the legal complaint to have generated more than 500,000 yen ($3,200) per month.
Tsuda’s lawyers argue that the content benefited commercially from the perceived association with the actor’s distinctive voice.
TikTok, however, disputes the characterisation. According to court records, the platform argues that the voice is a generic male voice and that any resemblance to Tsuda’s voice is subjective.
The platform has also argued that it was providing a forum for user-generated content within the boundaries of free expression.
Japan Examines Protection for AI-Replicated Voices
The case comes as Japan considers how existing legal protections should apply to AI-generated replicas of people’s voices and likenesses.
Japan’s Justice Ministry recently issued legally non-binding guidelines addressing the protection of voices under existing rights frameworks.
The ministry stated that a person’s voice, as an expression of their personality, can fall under publicity rights and protections against unauthorised use of an individual’s portrait and likeness.
However, the extent to which those principles apply to AI-generated voice replicas remains an emerging legal issue.
The case is therefore being closely watched by performers and industry groups concerned about the commercial use of AI-generated replicas without consent.
Yuko Sasaki, executive director of the Japan Actors Union and a fellow voice actor, told AFP that professional voices are developed through years of training and experience.
Japanese voice actors have also launched a campaign known as “No More”, opposing the use of generative AI to reproduce performers’ voices without permission.
A Global AI Issue
The dispute reflects a wider international debate over AI’s ability to reproduce distinctive human characteristics.
The International Federation of Actors has previously warned that copyright and data-protection frameworks in many countries have not fully adapted to the emergence of AI-generated replicas of performers.
The issue has already reached major technology companies. In 2024, OpenAI apologised to actress Scarlett Johansson after she raised concerns about the similarity between her voice and a synthetic voice introduced by the company.
The Japanese case highlights a different but increasingly important question: whether a person’s voice itself can constitute a commercially protected element of identity when reproduced through artificial intelligence.
For the entertainment industry, the answer could affect voice actors, musicians, broadcasters and other performers whose careers depend partly on distinctive vocal characteristics.
It could also have implications for technology companies and social-media platforms, which may face increasing pressure to identify, moderate and remove unauthorised AI-generated replicas.
Waseda University intellectual-property professor Tatsuhiro Ueno told AFP that the case is likely to attract international attention because of Tsuda’s recognisable voice and his prominent anime roles.
As AI voice-generation technology becomes increasingly sophisticated, the dispute underscores a growing challenge for regulators and courts: determining how existing intellectual-property, publicity and privacy protections should apply when technology can reproduce a person’s voice without their direct participation.
