Japan's Ministry of Justice expert panel has largely approved a draft guideline that protects a person's voice from unauthorized AI-generated use [1].
This move establishes a legal precedent by treating a person's voice similarly to their portrait and publicity rights. It aims to protect voice actors and public figures from economic harm and identity theft as generative AI technology makes high-fidelity voice cloning more accessible.
The expert panel met July 13, 2026 [2] to review the draft. The Ministry of Justice plans to publish the final guideline in August 2026 [3]. A spokesperson for the Ministry said voices should be protected targets in the same manner as portraits [4].
The guidelines come amid rising reports of AI-generated imitations appearing on social media. For example, more than 180 videos featuring AI-generated imitations of voice actor Tsuda Kenjirou were posted on TikTok [5]. This trend has prompted industry professionals to call for stronger legal safeguards to protect their livelihoods.
Voice actor Nakao Ryusei highlighted the personal and professional stakes of the issue. "Our voices are our tools of trade and our lives themselves," Nakao said. "Please listen to the feelings of voice actors whose voices are being used by generative AI without permission" [6].
The Minister of Justice said that unauthorized AI use has become widespread and the August guidelines are intended to address this trend [7]. By explicitly including voices under civil liability, the government provides a clearer path for individuals to seek damages, or request the removal of unauthorized AI clones.
The new framework seeks to balance technological innovation with the intellectual and economic rights of creators. The Ministry of Justice said that the protections will apply to civil liability, allowing affected parties to hold unauthorized users accountable in court [1].
“Our voices are our tools of trade and our lives themselves.”
This regulatory shift signals Japan's transition from general AI ethics guidelines to enforceable civil protections. By categorizing voice as a protected attribute similar to a physical likeness, the government is creating a legal mechanism to combat 'deepfake' audio. This may serve as a model for other nations facing similar challenges in the entertainment and voice-over industries as AI cloning threatens traditional employment models.
