On Wednesday, a Tokyo district court established that Japanese law safeguards the human voice as a protected aspect of personal identity. The case centred on anime performer Kenjiro Tsuda, who brought legal action against TikTok after discovering videos employing an AI-generated replica of his voice. Though Tsuda achieved a partial victory, the court declined his plea to have the videos removed, according to reporting from Agence France-Presse.

This lawsuit stands as Japan's inaugural legal challenge defending a performer's vocal identity against artificial intelligence reproductions. Tsuda, aged 55, is recognised for voicing Kento Nanami in Jujutsu Kaisen and has contributed to the Yu-Gi-Oh! series. The court's decision carries significant implications for how Japanese jurisprudence addresses emerging AI technologies and performer rights.

The unauthorised use of a performer's voice … can be considered an infringement on publicity rights

Judge Aya Takahashi

The court's determination

Publicity rights grant individuals authority over and financial benefit from elements of their personal identity. The Tokyo district court concluded that these protections extend to vocal characteristics when deployed exclusively for commercial gain. Judge Takahashi drew a parallel between a person's voice and a portrait, characterising both as emblematic representations of individual personality, according to AFP.

The judge determined that TikTok would not be compelled to delete the disputed videos, noting that the account operator had already voluntarily removed them.

The ruling "ended up making both sides look victorious,"

Takashi Hirano, Tsuda's lawyer

The disputed content

Between July 2024 and September 2025, the account in question disseminated 188 videos, as documented by the Guardian. These posts combined visual imagery with synthetic narration discussing urban legends, paranormal phenomena and conspiracy-related subject matter. The account's avatar depicted a character from Jujutsu Kaisen that Tsuda performs.

The profile accumulated more than 200,000 followers at its peak. Tsuda's legal team contended the account would have generated approximately ¥500,000 ($3,200) monthly, arguing the channel's success stemmed directly from the recognisable voice providing narration.

TikTok's defence centred on the assertion that the videos featured a "generic male voice". The platform maintained that any resemblance to Tsuda's actual voice remained subjective, according to court documents reviewed by AFP. Following the judgment, TikTok released a statement indicating it would examine the ruling before determining subsequent actions.

We will continue to sincerely work on the balance between technology development and rights protection

TikTok, in a statement to AFP

Industry response and broader concerns

During 2024, Japanese performers initiated an advocacy movement titled "No More," which seeks to prevent unauthorised replication of their voices through generative AI technology. Bin Shimada, participating in the campaign, warned AFP that unrestricted AI voice synthesis on digital platforms threatens the viability of the acting profession.

A voice "is the result of years of rigorous training and apprenticeship,"

Yuko Sasaki of the Japan Actors Union

The technical capacity for voice replication continues advancing rapidly. Google's current Gemini text-to-speech technology can synthesise a voice using merely 30 seconds of audio samples. Public backlash erupted this week when Netflix incorporated an AI-synthesised Gene Wilder voice in its Wonka production. ElevenLabs, a provider of AI voice generation technology, achieved a valuation increase to $22 billion on Wednesday.

Source: The Next Web