Japan’s first AI legislation becomes law – Focus is on promoting research and development; no monetary penalties
10 min read
On May 28, 2025, Japan's Parliament passed an Act to establish a national law to promote, and lightly regulate, AI, titled the "Act on Promotion of Research and Development and Utilization of Artificial Intelligence-Related Technologies" (the "AI Act").1 The AI Act is Japan's first law expressly directed at AI technology; however, its focus is on establishing basic policies and principles rather than creating detailed prescriptions and prohibitions like those, for example, in the EU AI Act. Now, one of the operating frameworks established by the AI Act, referred to as the Basic AI Plan, has been approved by the Cabinet Office.
Overview
The AI Act was enacted to promote innovation while establishing a framework to respond to risks associated with AI. Japan has heretofore taken a soft law approach as evidenced by the government's AI Guidelines for Business (last updated March 2025)2 and its role in helping to establish international guidelines, such as the Hiroshima AI Process International Guiding Principles.3 The provisions of the AI Act are thus not a major shift from its previous stance. Japan has stated that it is aiming to "establish a legal system [for AI] that serves as a global model" and that it is aiming to become "the world's most friendly country for developing and utilizing AI."4 The AI Act serves as a "fundamental law," which in the Japanese context means legislation that forms basic policies and creates broad rules for governing the subject matter. The AI Act also impels the government to take legislative and financial actions concerning AI (see Article 10), which will enable the government to enact more specific laws regulating AI should it become desirable.
Background
In July 2024, the Cabinet Office formed the AI Institutional Study Group with a mandate to determine, in consultation with various stakeholders, the next steps for AI regulation in Japan. On February 4, 2025, they published an interim summary of their findings.5 The guiding principle arising from the Group's findings was to promote AI innovation in Japan while mitigating its risks. The AI Act's goals and principles largely follow from the Group's recommendations. Specific directives relating to the AI Act and Japan’s approach toward AI are listed below. Most of these are directly reflected in the provisions of the AI Act.
- Strengthening the government's central position to oversee and formulate strategies
- Ensuring transparency and appropriateness throughout the AI development, use and deployment lifecycle
- Promotion of safety assessments and certification practices by domestic and international organizations
- Government surveys of and dissemination of information about serious incidents
- Use of AI by the government and development of government procurement guidelines
Key Provisions
Obligations of private businesses
While the majority of the provisions set forth a framework for future laws and policies relating to AI, there are several that impose duties on private business, but with no described penalty. For example, provisions in Article 7 outline the responsibilities of AI developers, AI providers and business users (each an "AI business actor").
Article 7 describes the duty of AI business actors to make reasonable efforts in utilizing AI to improve the efficiency and overall level of their business in line with the core principles of the Act. AI business actors must also comply with any policies that may be created by the national or local government bodies through the authority granted to them in Articles 4 and 5. Lastly, Article 25(2) gives the AI Strategic Headquarters the general ability to request cooperation from any entity when deemed especially necessary in performing its duties.
AI Strategic Headquarters
The AI Strategic Headquarters ("Headquarters"), as established in accordance with Article 19 of the AI Act, will be headed by the Prime Minister and will be comprised of all Cabinet members. Since the enactment of the AI Act, the Headquarters established two other managing bodies: the Artificial Intelligence Strategy Promotion Council and the Expert Investigation Team. The Promotion Council will be headed by the Strategic AI Minister and its members include around forty government officials with wide-ranging backgrounds, which reflects the AI Act's directive that AI initiatives should be undertaken using a whole-of-government approach among Japan's various ministries and agencies. The Expert Investigation Team's members are appointed by the Prime Minister and consist of people with an academic background in AI technology. The General Assembly of the Headquarters, which had its first meeting on September 12, 2025, will have three (3) Deputy Chief Cabinet Secretaries attend every meeting as observers, and may request any other relevant person to join a particular meeting. The General Assembly's meetings will be private, but in principle an overview of the meeting minutes and materials will be publicly released.
Several months later, the Headquarters published the Guidelines for Ensuring the Appropriateness of Research & Development and Utilization of AI-Related Technology ("AI Utilization Guidelines") on December 19, 2025, pursuant to Article 13 of the Act.6 The AI Utilization Guidelines recommend a general policy to ensure the appropriate use of AI for all subjects of the AI Act, which is comprised of (i) taking a risk-based approach, (ii) actively involving stakeholders in AI governance, (iii) establishing a holistic AI lifecycle governance framework to address risks present in the development and use of AI, and (iv) having an agile response to the risks related to AI. The AI Utilization Guidelines go on to recommend more specific practices for the different groups, including the government, AI business actors and the public.
Investment into AI
The government has also released a draft AI-related budget for fiscal year 2026, which provides insight on how the government plans to fund the four basic policies described in Chapter 2 of the Basic AI Plan (as further described below).7 In a 502.7 billion yen budget, nearly 90% of it (455.9 billion yen) will be spent on strategically strengthening Japan's AI development capabilities. The largest portion of this category is set to be over 387.3 billion yen going towards the development of a multimodal infrastructure model geared towards AI robots and physical AI. The next largest category is accelerating AI utilization (43.8 billion yen) which includes a broad range of items such as research on unmanned combat vehicle systems using AI and using AI to enhance the services provided by Hello Work, Japan's public employment security office. The other two categories, implementing AI governance and transformation toward an AI society have comparably smaller budgets of 1.1 and 1.9 billion yen respectively.
Additionally, on November 21, 2025, the government formulated a comprehensive economic package focused on a response to inflation, improving livelihoods, strengthening Japan's economy in the medium to long-term through strategic investment, and increasing Japan's defensive capabilities on the international stage.8 In this new package, AI and semiconductors were included as top strategic fields contributing to economic security.9
The Basic AI Plan
Article 18 sets forth the responsibility of the government to formulate a basic AI plan ("Basic AI Plan" or "Basic Plan") to promote R&D and the use of AI.10 The Basic AI Plan was approved by the Cabinet Office on December 23, 2025, and is organized into four chapters:
Chapter 1 sets out the background for Japan's implementation of AI-related policies and the goals it hopes to accomplish. Here, the government highlights its belief in the increasing importance of AI to industrial competitiveness, national defense, and for society as a whole. However, the Basic Plan notes that Japan lags behind peer countries in investment and development for AI and has been slow to adopt AI in everyday use and business. In view of this, the government notes that AI innovation can unlock Japan's potential and help solve longstanding challenges such as population decline, sluggish domestic investment, and stagnant wages. The government also states that Japan will focus on reproducing the value of "reliability" in AI innovation, which has been built up throughout Japanese society, and aims to unite diverse AI innovations from around the world under the concept of "trustworthy AI."
Chapter 2 sets forth the "three principles" and "four basic policies" for promoting the research, development and utilization of AI related technologies. Each of which are based on the basic policies described in Article 3 of the AI Act. The three principles are (i) promoting innovation while mitigating risks, (ii) agile response and (iii) promotion of integrated domestic and international policies. The four basic policies are (i) accelerating AI utilization, (ii) strategically strengthening Japan's AI developmental capabilities, (iii) leading in AI governance and (iv) a sustainable transformation toward an AI society. The measures the government is planning to take to implement each of the four basic policies is described in Chapter 3.
Chapter 3's measures are geared toward accomplishing the goals and solving the historic challenges that are laid out in Chapter 1 and are organized by the four basic policies in Chapter 2. Many of the measures have the government taking on a supporting or promotional role, but others describe specific actions for the government to take. Such actions include the creation of a government AI platform for internal use by its employees. Here, the government is aiming to lead by example in showing how to procure AI and utilize it in a safe and secure manner. Among its promotional activities, the Basic Plan also calls for building a data linkage infrastructure by leveraging high-quality data in quasi-public fields such as medical care, education and construction. As part of the transformation toward an AI society, the government will be reviewing and potentially amending current laws, such as the Act on the Protection of Personal Information.
Chapter 4 notes that the government will be setting benchmarks to measure the progress of the Basic Plan, and that the Basic Plan will be reviewed and changed regularly, with changes to be made at least annually.
Principle Code
The Intellectual Property Strategic Headquarters of the Cabinet Office also released a draft principle code relating to intellectual property and AI ("Principle Code").11 While there is no connection of the Principle Code to a specific Article in the AI Act, the draft does state that it is based on the intent of the AI Act. According to the draft, the Principle Code would apply to "Generative AI Businesses" (as defined in the Principle Code) and the government will expect such businesses to publish an acceptance statement on their website and describe how the company has implemented the three principles defined in the Principle Code or provide an explanation of why they have not implemented them, as applicable. Each principle contains detailed requirements, but includes, among other things, (i) that the company would disclose an overview of the type of data used to train AI models, (ii) that the company will create internal controls for protecting IP rights and post a summary of its policy, and (iii) that if requested by a third party that believes it may have a legal claim against the company regarding such person’s rights or other legally protected interests, the company will confirm whether certain webpages were used in the AI model. Similar to the AI Act, the draft Principle Code, as released on December 26, 2025, does not provide for monetary penalties for noncompliance.
Scope of Application
The AI Act does not clearly set out a territorial scope; however, Article 16 states that the government will be analyzing information regarding the development and use of AI domestically and abroad. Furthermore, neither the AI Act nor the Basic Plan distinguishes between on-shore and off-shore entities in its definition of research institutions or AI business actors, and the AI Utilization Guidelines state that overseas businesses are included among the “AI utilizing businesses” described within. It also appears that the Principle Code is intended to have extraterritorial effect.
There is no clearly defined sectoral scope, which is not surprising given the fundamental law role of the Act. Rather, the law applies generally to developers, providers and business users of AI.
For more information about AI regulation in Japan and in 31 other jurisdictions, please visit our global AI regulatory tracker here.
1 See the AI Act language here (only available in Japanese).
2 See the AI Guidelines for Business Version 1.1 here, issued by the Ministry of Internal Affairs and Communications and the Ministry of Economy, Trade and Industry.
3 See the Hiroshima Guiding Principles here.
4 From the overview of the AI Act published here.
5 See the interim summary published by the AI Regulation Research Group here.
6 See the AI Utilization Guidelines here.
7 See the AI budget for FY26 here (only in Japanese).
8 See materials released by the Cabinet Office describing the strategic growth and risk management investment categories here (only in Japanese).
9 See the economic package here (only in Japanese).
10 See the Basic AI Plan here.
11 See the draft Principle Code here.
White & Case means the international legal practice comprising White & Case LLP, a New York State registered limited liability partnership, White & Case LLP, a limited liability partnership incorporated under English law and all other affiliated partnerships, companies and entities.
This article is prepared for the general information of interested persons. It is not, and does not attempt to be, comprehensive in nature. Due to the general nature of its content, it should not be regarded as legal advice.
© 2026 White & Case LLP
