Japan’s Courts Say Your Voice Is Yours. The TikTok Clone Still Got Away

Kenjiro Tsuda won the argument and lost the case. On Sept. 30, the Tokyo District Court ruled that a person’s voice can be protected under Japan’s publicity rights, the first time a Japanese court has said so, according to Music Business Worldwide, which drew on AFP’s reporting. Then it dismissed his demand that TikTok take down the videos that cloned him, because there was nothing left to take down.

Tsuda, 55, is one of anime’s most recognizable voices: Kento Nanami in Jujutsu Kaisen, Seto Kaiba in Yu-Gi-Oh! Duel Monsters. His lawsuit targeted TikTok’s operator over at least 188 videos, posted between July 2024 and September 2025, that used AI-generated narration resembling his voice over clips about urban legends and conspiracy theories, The Next Web reports. The account had more than 210,000 followers and averaged 1.47 million views per video. MBW puts its estimated earnings at ¥500,000 to ¥750,000 a month, about US$3,200 to US$4,800 (roughly CA$4,450 to CA$6,700).

That is a small business. It was built on someone else’s throat.

What the judge actually said

Presiding Judge Aya Takahashi held that publicity rights can cover a voice when it is used for its commercial appeal. In AFP’s translation, as quoted by Music Business Worldwide: “The unauthorized use of a performer’s voice … can be considered an infringement on publicity rights.” Inven Global reports the court reasoned that a person’s voice, like their portrait, symbolizes an individual’s personality.

Tsuda had argued under both Japan’s Unfair Competition Prevention Act and publicity rights that the narration was meant to draw viewers by confusing them with his voice, per Inven Global. TikTok’s defence was that this was a “generic male voice” and that any resemblance was “subjective”; the platform also said the account holder claimed the voice model was “trained on a friend’s voice”.

The court never had to decide whether that was true. The anonymous uploader deleted the account while the suit was under way, so the takedown claim had no target. Tsuda’s side, according to Inven Global, said its “original objective was fully achieved” anyway.

A right without a remedy, for now

For creators, that outcome cuts two ways. The principle is now on the books: in Japan, a recognizable voice is something you can own commercially, not just a sound anyone can model. That is the hard part, and it has been decided in performers’ favour.

The enforcement problem is untouched. An anonymous account monetized a clone for more than a year, a famous performer had to sue the platform to get anywhere, and the uploader made the whole thing moot by hitting delete. A working voice actor without Tsuda’s profile, or a mid-size streamer whose voice gets cloned for a scam channel, faces the same anonymity with fewer resources.

The ruling did not arrive in a vacuum. Japan’s publicity rights are judge-made, recognized by the Supreme Court in 2012, and no statute spelled out voice protection. This summer the Justice Ministry moved first: an expert committee concluded on July 27 that voices should fall under publicity rights and that people can demand compensation or removal of posts where infringement is confirmed. The final report, published Aug. 7, was issued as an interpretive guideline, MBW reports, citing Jiji Press. It singled out AI “cover” tracks as a particular harm.

Performers have been organizing too. Japanese voice actors launched a campaign in 2024 against unauthorized AI voice cloning, The Next Web notes. Yuko Sasaki, executive director of the Japan Actors Union, has described a voice as “the result of years of rigorous training and apprenticeship”.

How it compares

In the US, protection is patchy. Tennessee’s ELVIS Act, signed in March 2024, was the first US state law to add voice to existing publicity protections. The federal NO FAKES Act would create an intellectual property right in a person’s voice and visual likeness but has drawn objections over carve-outs for satire and political commentary, MBW notes. Canada has no voice-specific statute.

Our read: Japan now has something closer to a national standard than the US does, delivered by a district court and a ministry guideline rather than a legislature. The open question for creators is whether platforms will treat it as a reason to act on voice-clone complaints before the lawsuit, not after the account vanishes.

Tsuda’s clone told stories about urban legends. The legend now has a citation: 188 videos, a deleted account, and a judgment that says the voice was his.