Japanese court rules human voices are protected in landmark AI case

A Tokyo court has sided with a Japanese voice actor who claimed a TikTok account had used AI to clone his "lustrous" baritone delivery, ruling that the human voice is protected by law in a landmark case.

Kenjiro Tsuda is a prominent anime voice actor who has starred in huge number of movies, series and video games across his more than 30-year career, with his turn as Seto Kaiba in Yu-Gi-Oh! arguably being his most iconic role. Tsuda asserted that the voice of the narrator of several TikTok videos uploaded to an anonymous account between July 2024 and the second half of last year was cloned from his own voice. While he was partially victorious in the case, his request to order TikTok to remove the videos in question was  dismissed, as the offending account has already been deleted.

In its decision, the court said that using the voice of an actor without permission should be deemed an infringement on their publicity rights, as reported by AFP. This was the first case of its kind in Japan.

Tsuda argued that the tell was the unmistakably deep and "lustrous" nature of the AI-generated vocal delivery in the narration. The actor's legal team said the account's popularity — at its height it had more than 200,000 followers — was due to the novelty of its "dubious, sordid content" being narrated by a prominent star. TikTok responded in court by arguing that a "generic male voice" whose similarity to Tsuda's was debatable was used in the narration, according to court records seen by AFP.

The Japanese government has been notably pro-AI in its national strategy, but there is growing concern among anime voice actors that more needs to be done to protect the creators whose work (or likeness) is often ingested by this technology. Last year, Japan's government asked OpenAI to ensure that anime and manga content it deemed to be "irreplaceable treasures" were not being used to train the company's Sora 2 generator in a way that violates copyright.