Japan has just given AI voice clones a legal problem of their own: your voice may be protected even when the words were never yours.
In a Sept. 30 ruling involving Japanese actor Kenjiro Tsuda, the Tokyo District Court recognized that a person’s voice can receive protection under publicity rights. The protection applies when the purpose of using the voice is exclusively to exploit its commercial appeal.
The dispute centered on TikTok videos allegedly created with AI-generated speech resembling Tsuda’s voice, rather than recordings of Tsuda himself. The distinction matters for businesses using synthetic voices in advertising, training, or customer communications: generating new speech does not remove the need to consider consent and rights to a person’s identity.
The court recognized voice protection but dismissed the takedown request
According to court documents reviewed by The Associated Press, a faceless TikTok account published several videos featuring a voice that Tsuda claimed was an AI-generated imitation of his own. Tsuda, who has provided voices for popular anime series, pointed to viewers’ recognition of the voice as evidence that the videos were designed to resemble him.
He took the case to court, seeking to have the videos deleted. The judge agreed with the central legal argument, ruling that “the human voice is symbolic of individual personality, just like one’s portrait,” and that a voice can therefore receive protection under publicity rights.
However, the court dismissed Tsuda’s deletion request because the account had already been closed. It did not decide whether the disputed videos infringed his publicity rights.
More must-read AI coverage
- SS&C Intralinks DealCentre AI vs. Datasite: Which platform is built for the future of dealmaking?
- SS&C Intralinks FundCentre AI vs. Juniper Square: Which platform better supports modern private markets fund managers?
- Why Data, Not Models, Determines AI Success
- The Rise of the AI-Native Factory: How Physical AI Is Transforming Manufacturing
Other disputes highlight questions about consent and compensation
Tsuda’s case adds to a wider debate over consent and compensation, although disputes about vocal identity and copyrighted training material involve different legal questions.
In May 2024, OpenAI paused ChatGPT’s Sky voice after Scarlett Johansson said it sounded like her. OpenAI denied that Sky was her voice or was intended to resemble it.
Publishers have also questioned how AI companies use and compensate them for content. Reddit reportedly considered revisiting its Google AI licensing agreement amid concerns about AI search reducing website traffic.
What does this mean for everyone else?
Japan’s case isn’t all that unique. Several regions have already dealt with AI-generated impersonation across entertainment, advertising, politics, and fraud. The ruling’s significance is its recognition that vocal identity can receive protection under publicity rights, subject to the conditions the court described.
If you suspect your voice has been imitated without permission, preserve the disputed audio, URLs, account details, screenshots, and dates alongside your original recordings. These materials can help document a platform report or legal consultation. Businesses using synthetic voices should review consent, permitted uses, and vendor licensing terms before publishing.
For everyone else, the bigger takeaway is to check what protections exist where you live rather than assuming a ruling in Japan automatically applies elsewhere. Voice-cloning disputes can involve publicity rights, privacy law, copyright, consumer protection, or other rules depending on the jurisdiction, so the legal route may differ even when the underlying AI technology is the same.
Read more: China’s emerging AI face-licensing market shows how consent-based digital identities can be commercialized—and why control over their use still matters.