Japan's AI Promotion Act: No Fines, But Real Pressure

Japan has chosen the opposite path to the EU. It has a promotion-first framework law, supported by non-binding guidelines and existing sector laws. Understanding this "soft law" model is useful for AIGP candidates and for any organisation operating in Japan.

Japan's AI framework at a glance

How the Act works

1. Promotion first

The Act's stated aim is to promote AI research, development and use while ensuring transparency and appropriate handling of risks. It reflects Japan's concern that it has fallen behind other economies in AI.

2. A whole-of-government structure

The AI Strategic Headquarters, led by the Prime Minister, coordinates policy across ministries and drives the AI Basic Plan.

3. The AI Basic Plan

Adopted on 23 December 2025, the plan aims to make Japan the easiest country in the world to develop and use AI. It sets four basic policies: accelerating AI use, strategically strengthening AI development, improving AI reliability, and continuing to transform society with AI.

4. Investigation and guidance, not fines

The government can gather information, analyse cases where AI has infringed people's rights or interests, and give guidance or advice to businesses. Commentators expect serious cases may be made public. In Japan's business culture, being publicly identified as a bad actor carries real cost, which is why "no fines" doesn't mean "no consequences".

The guidelines layer

Japan's AI Guidelines for Business, issued by METI and MIC in 2024 and updated since, set out principles and practices for AI developers, providers and business users. They cover human-centricity, safety, fairness, privacy, security, transparency, accountability, education and innovation. The Act calls for guidelines aligned with international standards, and work on guidance for appropriate AI use continued through 2025–26.

Japan vs the EU and South Korea

See South Korea's AI Basic Act and EU AI Act risk tiers.

Other parts of Japan's AI governance landscape

The Hiroshima AI Process

Under Japan's G7 presidency in 2023, the G7 launched the Hiroshima AI Process, which produced International Guiding Principles and a Code of Conduct for organisations developing advanced AI systems. The OECD later launched a reporting framework so companies can report voluntarily on how they apply the Code. It's a good example of Japan's preference for international, voluntary standards.

Japan AI Safety Institute

Japan set up its AI Safety Institute (J-AISI) in 2024 to work on evaluation methods and safety standards, and to cooperate with similar institutes in other countries.

Personal data: the APPI

The Act on the Protection of Personal Information (APPI) applies to AI systems that process personal data. Japan's Personal Information Protection Commission has issued cautions to generative AI providers on handling personal data, and the APPI is subject to regular reviews that consider AI. Key APPI concepts for AI teams include purpose specification, restrictions on acquiring sensitive ("special care-required") personal information without consent, and rules on third-party provision and cross-border transfers.

Copyright: Article 30-4

Japan's Copyright Act includes Article 30-4, which allows the use of copyrighted works for information analysis, including machine learning, without permission, as long as the use doesn't involve enjoying the work's expression and doesn't unreasonably prejudice the rights holder's interests. The Agency for Cultural Affairs has published guidance on how this applies to generative AI, including limits where outputs reproduce protected expression.

Sector rules

Financial, medical and other regulators apply existing sector rules to AI. For example, AI-based medical software is regulated as a medical device where it meets the definition.

How Japan's approach affects foreign companies

• No AI-specific licence or registration is required to offer AI services in Japan.
• The government can investigate and publicise cases of inappropriate AI use under the AI Promotion Act, so reputational risk is real even without fines.
• The APPI applies to foreign businesses handling personal data of people in Japan in connection with supplying goods or services.
• Following the AI Business Operator Guidelines is the practical way to show responsible conduct.

Practical steps for companies in Japan

1. Map your AI uses against the AI Guidelines for Business.
2. Check APPI compliance for training data and AI outputs that contain personal information.
3. Prepare to respond to government information requests.
4. Monitor the AI Basic Plan and any sector-specific guidance.

Frequently asked questions

Sources

1. Future of Privacy Forum: Understanding Japan's AI Promotion Act
2. Bird & Bird: Japan's AI governance since the AI Act
3. Nippon.com / Jiji: Japan adopts first basic plan on AI
4. White & Case: Japan's first AI legislation

Last reviewed: October 2026.