Early 2026: the government steps in
In January 2026, Japan's Chief Cabinet Secretary Minoru Kihara said at a press conference that the government would assess the sexual-deepfake issue under the AI Promotion Act, which took effect in May 2025 (he initially declined to detail countermeasures, citing that it involved an individual company's service).
Note the nature of that law: it is built around investigation and guidance, with no mandatory penalties of its own. It lets the government investigate cases where AI infringes citizens' rights and issue guidance to the businesses involved.
The actual arrests
Tokyo's Metropolitan Police arrested a 31-year-old man who, over two years, used free generative-AI tools to create more than 520,000 deepfake images of about 300 entertainers, adult-film actresses and idols, uploading at least 2,200 to paid membership sites and earning roughly US$70,000.
A separate 2025 arrest involved a suspect accused of selling around 20,000 explicit deepfake images of 262 women. All of these were pursued under existing laws.
The current legal gap
Japan currently has no dedicated, comprehensive anti-deepfake law. Prosecutors mainly lean on the Penal Code's ban on distributing obscene material, plus defamation and portrait-rights provisions.
Some local governments have moved first — for example, Tottori Prefecture passed an ordinance carrying administrative penalties for creating or distributing such content.
What it means for viewers and the industry
This is about non-consensual fakes that paste a real person's face onto sexual imagery — a rights and harm issue, and something entirely separate from licensed adult work made with the performer's consent.
Expect more scrutiny ahead, possibly including labelling rules for AI content. If you want to know how to spot deepfakes, see our dedicated deepfake guide.
FAQ
Has Japan fully banned deepfakes?
Not with a dedicated comprehensive law yet. For now it uses the existing Penal Code (distributing obscene material), defamation and portrait rights, plus local ordinances such as Tottori's. The 2025 AI Promotion Act is investigation-and-guidance in nature, with no mandatory penalties.
What did the government actually do?
In early 2026, Chief Cabinet Secretary Minoru Kihara said the government would assess and investigate the AI sexual-deepfake issue under the AI Promotion Act that took effect in May 2025, and provide guidance to the businesses involved.
Is this about regular adult videos?
No. It concerns non-consensual fakes that use a real person's face, which is a rights and harm issue — separate from licensed adult work made with the performer's consent.
How do I spot an AI deepfake?
See our dedicated deepfake guide. Common tells include unnatural edges around the face, inconsistent lighting, and odd blinking or expressions.