AI laws in Japan focus on promotion rather than penalties. Japan has chosen a different path from Europe. Its first dedicated AI legislation, the AI Promotion Act, is designed to encourage AI research and development and coordinate government policy rather than to impose duties backed by fines. Existing laws on privacy, copyright and consumer protection still apply to AI, government guidelines set out what is expected of businesses, and some local governments have started to act on deepfakes.
This guide explains how that system works and what it means for you. See our Japan AI law tracker for the current status of each rule.
AI laws in Japan: the short answer
- Businesses: there are no new AI-specific compliance duties with penalties, but you are expected to cooperate with government measures and follow official guidelines.
- Everyone: laws on privacy, defamation and copyright still apply when you use AI.
- Deepfakes: there is no national AI deepfake law, but at least one prefecture, Tottori, now punishes AI-made sexual images of minors.
Why Japan’s approach to AI regulation relies on soft law
Japan wants to catch up in AI development and has long preferred guidance and cooperation between government and industry to hard regulation. Its approach to AI governance reflects that. Rather than regulating AI by sorting systems by risk and banning some uses, as the EU AI Act does, it sets national goals to promote AI innovation and the use of AI technologies, creates a central body to coordinate policy, and relies on current law and guidance to deal with AI risk and harm.
This light-touch AI regulatory approach is often described as soft law: non-binding guidelines and shared AI principles do most of the work, while binding rules come from laws that already exist.
The AI Promotion Act: Japan’s first AI legislation
The Act on Promotion of Research and Development, and Utilization of AI-related Technology (Act No. 53 of 2025) passed the Diet on 28 May 2025, was promulgated and partly in force on 4 June 2025, and came fully into force on 1 September 2025.
It does three main things:
- Creates the AI Strategy Headquarters (sometimes translated as the AI Strategic Headquarters), which the Prime Minister chairs with every cabinet minister as a member, to set and coordinate the government’s work on AI.
- Requires an AI Basic Plan, adopted by the Cabinet and reviewed, in principle, every year.
- Gives the government a role in harms: where AI misuse damages people’s rights, the government can investigate and give businesses guidance, advice and information.
Businesses must cooperate with government measures, but there are no fines for failing to do so. Read the tracker entry.
The AI Basic Plan
The Cabinet adopted Japan’s first AI Basic Plan on 23 December 2025 and approved a revised version on 14 July 2026. Because it is reviewed, in principle, every year, the plan is the best guide to where Japanese AI policy is heading, including areas where the government may consider firmer rules in future.
AI regulatory guidance businesses are expected to follow
Alongside the Act, Japan relies on official guidance. Under the Act itself, the AI Strategy Headquarters issued guidelines on ensuring the appropriate development and use of AI on 19 December 2025. The government’s AI Guidelines for Business, first issued in 2024 by the economy and communications ministries, set out principles for AI developers, AI providers and AI business users, such as safety, fairness, privacy, transparency and accountability. They are voluntary, but they shape what regulators and business partners expect of AI use.
Japan is also active in global AI governance and responsible AI development. It set up an AI Safety Institute in 2024 to study how to evaluate the safety of AI systems, and it led the G7’s Hiroshima AI Process, which produced the International Guiding Principles and the International Code of Conduct for Organizations Developing Advanced AI Systems.
Data protection laws and other existing rules
- Privacy: the Act on the Protection of Personal Information applies to businesses using AI systems to process personal data. The privacy regulator has warned them to be careful about entering personal information into generative AI services.
- Copyright: Article 30-4 of Japan’s Copyright Act allows works to be used for data analysis, including to train AI models, with limits where use would unreasonably harm the rights holder. Copyright in what AI produces, and infringement by outputs that copy existing works, follow the ordinary rules.
- Defamation and fraud: AI-generated content that damages someone’s reputation or deceives people can break existing criminal and civil law.
Deepfakes and local rules
Japan has no national AI-specific deepfake law. One prefecture has stepped in. Tottori Prefecture amended its youth ordinance so that it has banned AI-altered sexual images of minors since 1 April 2025. Since 4 August 2025 the prefecture can order deletion, publicly name people who ignore a deletion order, and impose fines of up to ¥50,000.
What Japanese AI rules mean for you
Companies
- Follow the AI Basic Plan and the business guidelines: they set expectations even without penalties.
- Cooperate if the government investigates AI misuse involving your products.
- Check how each AI system you use handles personal information.
Creators and everyday users
- Privacy, defamation and copyright rules still apply when using generative AI.
- Never create or share sexual images of minors, AI-made or not; other prefectures may follow Tottori’s lead.
Key dates
- 1 September 2025: the AI Promotion Act is fully in force.
- 23 December 2025: the first AI Basic Plan is adopted.
- 14 July 2026: the AI Basic Plan is revised; the next revision is due within a year.
Common questions regarding AI in Japan
Is AI regulated in Japan?
Yes, but lightly, and there is no comprehensive AI regime with fines. The AI Promotion Act sets up a national framework for AI and lets the government investigate AI misuse and give guidance. Most of the practical rules related to AI come from existing laws on privacy, copyright and consumer protection, plus the government’s voluntary AI Guidelines for Business.
How is Japan’s approach different from the EU AI Act?
The EU AI Act sorts AI systems by risk, bans some uses outright and imposes detailed duties on high-risk systems, with large fines. Japan’s law does none of that. It focuses on promoting AI and coordinating government action, and relies on cooperation rather than enforcement. A company working in both places should plan around the EU’s AI Act, which is stricter.
Can AI companies train on copyrighted works in Japan?
Japanese copyright law allows works to be used for data analysis, which covers AI training in many cases, without the rights holder’s permission. The exception has limits: it does not apply where the use would unreasonably harm the rights holder’s interests, and AI outputs that copy protected works can still infringe.
Are deepfakes illegal in Japan?
There is no national law aimed specifically at AI deepfakes. Depending on how they are used, deepfakes can break laws on defamation, fraud or obscenity, and Tottori Prefecture has banned AI-made sexual images of real minors, with fines.
Who regulates AI in Japan?
There is no single artificial intelligence regulator. The AI Strategy Headquarters, led by the Prime Minister, coordinates policy, and existing bodies handle their own areas: the Personal Information Protection Commission for privacy, the Agency for Cultural Affairs for copyright, and the economy and communications ministries for the business guidelines.
Will Japan bring in stricter AI rules?
The Act itself has no penalties. Because the AI Basic Plan is reviewed every year, it is the place to watch for any move towards firmer rules on particular AI applications.
Last checked: 5 November 2026. This guide explains the law in plain English; it is not legal advice. Laws change quickly, so check the tracker entries and their official sources before you rely on them. Spotted an error? Tell us, and see how we check facts.
