Social Principles of Human-Centric AI
Japan · Guideline · Council for Social Principles of Human-centric AI
Our summary
A set of principles for developing and deploying AI in ways that respect human dignity, advance diversity, and promote sustainability. The framework outlines social principles such as human-centrism, privacy protection, fairness and transparency, alongside practical guidelines for AI research and implementation. It is designed to support the transformation toward Society 5.0, where AI benefits society while mitigating risks through changes to human capabilities, social systems, industry structures, governance and innovation environments.
Written by the Index in our own words. It is not legal advice; the official text is the authority.
Inside this policy
Every section of this policy in order. Our notes summarise each one in plain language and do not quote the official text, which is the authority.
Drafted by the Index An automated summary of the official text, not yet reviewed by a person.Reviewed Checked by a person against the official text.
8 sections
Jump to a chapter
Introduction
Japan aims to build an AI-Ready Society to address social challenges such as aging population and labor shortage while achieving the UN's Sustainable Development Goals, requiring transformation across workforce capabilities, organizational frameworks, business models, technological innovation systems, and governance structures. AI technologies are treated as part of complex information systems rather than standalone tools, and the document presents social principles for AI while identifying issues for research, development, and social implementation requiring stakeholder cooperation to realize Society 5.0 effectively.
Basic Philosophy
Human-centric AI is founded on three values: respecting human dignity by using AI to enhance rather than control human abilities, enabling diversity and inclusion through appropriate AI development that allows people of different backgrounds to pursue well-being while generating social value, and creating sustainability through AI-driven solutions to resolve disparities and support environmental and climate goals. Realizing these values requires transformation of individuals, organizations and social innovation environments, including developing AI literacy and digitizing information for AI analysis. Japan is obliged to contribute to global sustainability objectives through AI.
Social Changes for AI-Ready Society
In an AI-Ready Society, people should understand AI's benefits and risks including how biases can arise from statistics, social conditions, or deliberate misuse, and education and social systems must support diverse individuals in creative work with AI while developing sufficient professionals skilled in implementing and designing AI systems. Social systems including healthcare, finance, insurance, transportation and energy must flexibly evolve to accommodate AI advancement, capturing benefits from improved efficiency and individualization while addressing negative outcomes such as inequality, disparity and social exclusion through flexible architecture designs with appropriate mechanisms.
Foundational infrastructure and innovation support
Societies need flexible labor markets, startup investment, and shared data infrastructure to enable diverse objectives and system interoperability, with universities, research institutions, industries and the public jointly developing and evaluating AI through shared data access protected by privacy and security safeguards. Governance systems must bring together government, business, universities, research institutions, and the public to identify issues, evaluate impacts, and decide on rules, standards and conduct codes, remaining flexible and technology-driven with international cooperation for cross-border challenges. Social Principles of AI apply to society and governments, while R&D and Utilization Principles apply to AI developers and business operators.
Human-centric and education principles
AI development and use must not violate fundamental constitutional or international human rights, must expand human abilities rather than replace humans, and must respect people's own decision-making. Education systems should provide all people with literacy about AI basics, mathematics, data science, and the limitations and risks of AI technology including bias and fairness concerns, moving from uniform teaching to personalized approaches using individual interests and AI technologies. Schools and administration should partner with private enterprises and citizens to develop this educational environment, ensuring burdens do not fall solely on schools or teachers.
Privacy protection and security principles
AI systems using personal data require careful handling with technical and non-technical protections proportional to the data's sensitivity, from highly influential information like medical history to semi-public information, ensuring AI does not infringe on individual freedom, dignity or equality while avoiding over-dependence on single AI systems or types. Societies must maintain fair competition preventing concentrated data collection or monopolistic practices regardless of whether resources concentrate in specific countries or companies, and must prevent AI use from creating unfair wealth or influence concentration. Decision-making using AI must be fair, transparent and accountable to prevent undue discrimination against individuals.
Fairness, dialogue, and innovation principles
All people should be treated fairly without unjustified discrimination based on diverse backgrounds including race, gender, nationality, age, political beliefs and religion, and AI applications should provide appropriate case-by-case explanations regarding when and how AI is used, data collection and usage, and measures to ensure appropriate results, with open dialogue facilitated on AI use, adoption and operation to enable public understanding and judgment. Japan should integrate these principles into government policies and international discussions to establish an AI-Ready Society globally, presenting a vision aligned with the Sustainable Development Goals, and the principles must adapt over time in response to technological advances, social shifts, and international developments.
Council establishment and procedures
The Council, established under the AI Strategy Expert Meeting, examines and proposes Social Principles of Human-centric AI to advance their integration into national strategy. The Council operates as a public body with publicly available meeting minutes, unless the Chairperson determines disclosure inappropriate, and its operations are managed by the Cabinet Office.
Controls
What this policy requires, forbids, grants or penalises, one control at a time and in our own words. Each control links to the official text, which is the authority. This is not legal advice.
Drafted by the Index Drafted automatically from the official text, not yet reviewed by a person.Reviewed Checked by a person against the official text.
Themes covered 14 of 35
Choose a theme to see only its controls. About the themes
Risk and scope 2 of 4
- Risk classification, not covered
- Impact assessment, not covered
Data and privacy 3 of 4
- Copyright and training data, not covered
Transparency and records 1 of 4
- Labelling of synthetic content, not covered
- Technical documentation, not covered
- Record keeping and logging, not covered
Technical safeguards 4 of 4
Assurance and lifecycle 0 of 4
- Conformity assessment and certification, not covered
- Registration and notification, not covered
- Incident and breach reporting, not covered
- Post-market monitoring and corrective action, not covered
Rights and protected groups 0 of 4
- Automated decision-making rights, not covered
- Children and vulnerable groups, not covered
- Biometric identification and surveillance, not covered
- Whistleblowing, complaints and redress, not covered
Roles and organisation 4 of 6
- General purpose and foundation models, not covered
- Public sector use and procurement, not covered
Enforcement and sectors 0 of 5
- Enforcement, penalties and regulators, not covered
- Sector rules: health, not covered
- Sector rules: finance and insurance, not covered
- Sector rules: employment, not covered
- Sector rules: education, not covered
The full list of controls could not be loaded, so only the first 24 are shown.
Introduction 1 control
- Applies to: Public authorityDrafted by the Index
Policymakers and society should treat AI broadly to encompass various advanced technological systems when addressing complex information systems for social challenges.
Basic Philosophy and Structure 1 control
- ObligationApplies to: AnyoneDrafted by the Index
All stakeholders should cooperate in developing AI systems and policies based on three foundational pillars: respect for human dignity, diversity and inclusion, and sustainability.
Sustainability and Human Potential 2 controls
- ObligationApplies to: AnyoneDrafted by the Index
Society should develop understanding of AI's impacts including algorithmic bias, creative capacities working with AI, and expertise in AI implementation across multiple disciplines.
Sustainability and Human PotentialOfficial text of Sustainability and Human Potential (opens in a new tab) - ObligationApplies to: AnyoneDrafted by the Index
AI systems should be used to create new solutions, address environmental challenges, and strengthen scientific and technological capacity.
Sustainability and Human PotentialOfficial text of Sustainability and Human Potential (opens in a new tab)
Industrial Structures, Innovation Systems and Governance 3 controls
- ObligationApplies to: Public authorityDrafted by the Index
Governments should maintain flexible, fair labor markets and international openness to enable diverse people to pursue varied goals in the context of AI development.
Industrial Structures, Innovation Systems and GovernanceOfficial text of Industrial Structures, Innovation Systems and Governance (opens in a new tab) - ObligationApplies to: AnyoneDrafted by the Index
Universities, industry, and public institutions should collaborate on AI research and utilization with secure access to analyzable data.
Industrial Structures, Innovation Systems and GovernanceOfficial text of Industrial Structures, Innovation Systems and Governance (opens in a new tab) - ObligationApplies to: Public authorityDrafted by the Index
A governance framework should enable continuous coordination among all stakeholders to update policies and systems as technology and society evolve.
Industrial Structures, Innovation Systems and GovernanceOfficial text of Industrial Structures, Innovation Systems and Governance (opens in a new tab)
Human-Centric Principle and Principle of Education/Literacy 4 controls
- ProhibitionApplies to: ProviderDrafted by the Index
AI systems must not violate constitutional or international human rights.
Human-Centric Principle and Principle of Education/LiteracyOfficial text of Human-Centric Principle and Principle of Education/Literacy (opens in a new tab) - ObligationApplies to: ProviderDrafted by the Index
AI systems should expand human capabilities while preventing over-dependence and misuse.
Human-Centric Principle and Principle of Education/LiteracyOfficial text of Human-Centric Principle and Principle of Education/Literacy (opens in a new tab) - ObligationApplies to: AnyoneDrafted by the Index
Appropriate stakeholders must bear responsibility for consequences of users' decisions about AI use.
Human-Centric Principle and Principle of Education/LiteracyOfficial text of Human-Centric Principle and Principle of Education/Literacy (opens in a new tab) - ObligationApplies to: ProviderDrafted by the Index
All people deserve access to AI benefits through user-friendly design that prevents digital divides.
Human-Centric Principle and Principle of Education/LiteracyOfficial text of Human-Centric Principle and Principle of Education/Literacy (opens in a new tab)
Principle of Privacy Protection and Principle of Ensuring Security 4 controls
- ObligationApplies to: ProviderDrafted by the Index
AI systems using personal data must protect individual freedom, dignity and equality through technical and non-technical safeguards.
Principle of Privacy Protection and Principle of Ensuring SecurityOfficial text of Principle of Privacy Protection and Principle of Ensuring Security (opens in a new tab) - RightApplies to: AnyoneDrafted by the Index
Individuals must have meaningful involvement in managing their personal data.
Principle of Privacy Protection and Principle of Ensuring SecurityOfficial text of Principle of Privacy Protection and Principle of Ensuring Security (opens in a new tab) - ObligationApplies to: ProviderDrafted by the Index
Personal data protection must account for varying degrees of sensitivity, balancing protection needs with legitimate uses based on cultural context.
Principle of Privacy Protection and Principle of Ensuring SecurityOfficial text of Principle of Privacy Protection and Principle of Ensuring Security (opens in a new tab) - ObligationApplies to: Public authorityDrafted by the Index
Society should promote research and development to assess and mitigate security risks from AI systems.
Principle of Privacy Protection and Principle of Ensuring SecurityOfficial text of Principle of Privacy Protection and Principle of Ensuring Security (opens in a new tab)
Principle of Fairness and Principle of Innovation 6 controls
- ProhibitionApplies to: ProviderDrafted by the Index
AI systems must treat all people fairly without unjustified discrimination based on race, gender, nationality, age, or other characteristics.
Principle of Fairness and Principle of InnovationOfficial text of Principle of Fairness and Principle of Innovation (opens in a new tab) - ObligationApplies to: DeployerDrafted by the Index
Deployers should provide case-by-case explanations of AI use and data handling to users through open dialogue.
Principle of Fairness and Principle of InnovationOfficial text of Principle of Fairness and Principle of Innovation (opens in a new tab) - ObligationApplies to: Public authorityDrafted by the Index
Society should promote AI innovation by transcending boundaries of nationality, industry, and academia through global cooperation.
Principle of Fairness and Principle of InnovationOfficial text of Principle of Fairness and Principle of Innovation (opens in a new tab) - ObligationApplies to: Public authorityDrafted by the Index
Methods for assessing AI quality and reliability must be established.
Principle of Fairness and Principle of InnovationOfficial text of Principle of Fairness and Principle of Innovation (opens in a new tab) - ObligationApplies to: Public authorityDrafted by the Index
Infrastructure enabling cross-border data utilization without monopolization should be created.
Principle of Fairness and Principle of InnovationOfficial text of Principle of Fairness and Principle of Innovation (opens in a new tab) - ObligationApplies to: Public authorityDrafted by the Index
Governments should implement regulatory reforms to facilitate efficient AI deployment.
Principle of Fairness and Principle of InnovationOfficial text of Principle of Fairness and Principle of Innovation (opens in a new tab)
Council for Social Principles of Human-centric AI 3 controls
- ObligationApplies to: Public authorityDrafted by the Index
A Council is established to examine and propose basic principles for implementing AI in society, reporting to the Integrated Innovation Strategy Promotion Council.
Council for Social Principles of Human-centric AIOfficial text of Council for Social Principles of Human-centric AI (opens in a new tab) - ObligationApplies to: Public authorityDrafted by the Index
The Council operates under public governance with open meetings and published minutes, except where the Chairperson determines otherwise.
Council for Social Principles of Human-centric AIOfficial text of Council for Social Principles of Human-centric AI (opens in a new tab) - ObligationApplies to: Public authorityDrafted by the Index
The Cabinet Office shall provide administrative support to the Council with cooperation from relevant agencies.
Council for Social Principles of Human-centric AIOfficial text of Council for Social Principles of Human-centric AI (opens in a new tab)
Change history
No changes recorded since we began monitoring this policy.