|
Getting your Trinity Audio player ready...
|
A Tokyo court has delivered Japan’s first legal ruling declaring that a person’s voice is protected under publicity rights, a landmark decision that emerged from a lawsuit filed by one of the country’s most recognizable anime voice actors against TikTok over AI-generated content he says mimics his distinctive baritone. Wednesday’s verdict from the Tokyo District Court marks a genuine milestone for how Japanese law treats vocal identity in the AI era, even though the actor who brought the case didn’t get everything he asked for.
Kenjiro Tsuda, widely known for voicing Kento Nanami in the hit anime series “Jujutsu Kaisen” and for roles in “Yu-Gi-Oh!,” sued the operator of TikTok seeking removal of videos he said used an AI-generated voice closely resembling his own. Presiding judge Aya Takahashi ruled that the unauthorized use of a performer’s voice can constitute an infringement of publicity rights, establishing for the first time in Japan that a voice deserves legal protection comparable to portrait rights, the existing legal framework that already shields a person’s image and likeness from unauthorized commercial use. Takahashi described a voice as symbolic of one’s personality in the same way a portrait is, and clarified that protection applies specifically when the purpose of using a given voice is exclusively to take advantage of its commercial appeal.
Despite establishing that legal principle, the court dismissed Tsuda’s actual request to have the disputed videos taken down, making Wednesday’s outcome a genuinely partial victory. The precedent he was seeking got established, but the specific remedy he wanted, removal of the content itself, was denied. TikTok’s operator had argued during the case that the narration in question used what it described as a standard male voice, one that wouldn’t reasonably be mistaken for Tsuda’s, while the individual who posted the videos claimed the AI had been trained using a friend’s voice rather than any deliberate attempt to clone Tsuda’s specifically. Despite those defenses, an analysis presented during the case reportedly found strong similarities between Tsuda’s actual voice and the AI-generated narration, similarities Tsuda argued were significant enough to mislead viewers into believing they were hearing him.
The case lands at a moment when Japan’s animation industry, one of the country’s most culturally significant and commercially valuable creative sectors, has grown genuinely alarmed about what unrestricted AI voice cloning could mean for the profession. Yuko Sasaki, executive director of the Japan Actors Union, described voice acting as the product of years of rigorous training and apprenticeship, a skill built through sustained effort rather than something that should be casually replicable by AI software trained on a handful of audio samples. Fellow actor Bin Shimada, involved in an industry campaign group pushing back against unauthorized AI voice replication, warned that the entire industry could be crushed if unbridled voice cloning continues spreading unchecked across social media platforms. That concern isn’t abstract in Japan, where voice actors frequently achieve a level of independent celebrity and fan devotion that rivals or exceeds the characters they perform, making their individual vocal identities genuinely valuable commercial assets in their own right.
Parallel to Tsuda’s lawsuit, Sasaki and other prominent Japanese voice actors have organized a campaign called “No More,” specifically targeting the use of generative AI to mimic performers’ voices without consent. That kind of organized industry pushback mirrors concerns raised by performers in other countries as voice cloning technology has become both more accessible and more convincingly realistic. TikTok responded to the ruling by saying it will respect the court’s judgment while considering what steps to take after fully scrutinizing the decision’s details, language that leaves open exactly how the platform intends to handle the underlying content going forward, or whether it plans broader policy changes in response to the precedent this ruling now establishes.
What makes this case particularly significant from a legal standpoint is the gap it steps into. Japan currently lacks any law explicitly defining or protecting voice rights, leaving courts to work out how existing legal concepts like publicity rights should extend to a category of personal identity that technology has only recently made vulnerable to convincing, low-cost replication at scale. Japan’s Justice Ministry has issued guidelines stating that uploading videos to social media using voices similar to real people, and generating revenue from that content, may constitute a rights violation, guidance that predates Wednesday’s ruling but that the court’s decision now gives considerably more concrete legal weight behind.
Japan is far from alone in grappling with exactly this question, and the Tokyo ruling fits into a broader international pattern of courts and legislators scrambling to catch up with voice cloning technology. In India, a Bombay High Court case brought by a veteran Bollywood performer resulted in judges ruling that a person’s voice, vocal style, and even mannerisms are protected under personality rights and the right to publicity, finding that unauthorized commercial use of a celebrity’s persona violates publicity rights and amounts to passing off. In China, the Beijing Internet Court ruled in a case later designated a “typical case” by the country’s Supreme People’s Court that an AI-generated voice infringed a dubbing performer’s personality rights. In the United States, voice actors Paul Lehrman and Linnea Sage cleared an early legal hurdle in their own case against an AI voice cloning company, with a federal judge allowing breach of contract and right of publicity claims to proceed, while Congress has separately reintroduced the bipartisan NO FAKES Act, aimed at establishing a federal framework protecting individuals’ right of publicity against exactly this kind of unauthorized AI replication.
Taken together, these rulings across Japan, India, China, and the United States suggest courts in multiple major jurisdictions are converging on a similar basic conclusion, that a person’s voice carries legally protectable value tied to their individual identity, even as each country works out the specific mechanics of how that protection actually gets enforced against a technology capable of replicating a voice from remarkably little source material. For Japan’s voice acting industry, Wednesday’s ruling offers a genuine legal foothold, even if Tsuda’s own case shows that establishing the underlying right and actually winning a specific remedy remain two very different battles that don’t automatically follow from one another.
Further detail on the ruling is available through Kyodo News’s official coverage. For more coverage of AI regulation and intellectual property law, visit Business Tech.
