Human Centred Values

AI systems should be designed and operated in ways that respect the rule of law, human rights, democratic values and diversity, with appropriate safeguards to ensure human determination and control over consequential decisions (human agency), whilst incorporating mechanisms to protect fairness an…

Semantic Classification

Content

• AI systems should be designed and operated in ways that respect the rule of law, human rights, democratic values and diversity, with appropriate safeguards to ensure human determination and control over consequential decisions (human agency), whilst incorporating mechanisms to protect fairness and human dignity.

Fast vs. Slow Takeoff Scenarios

• The possibility of fast or slow takeoff scenarios can exist within both interpretations, depending on the rate at which technological progress accelerates and the degree to which humans can maintain agency and control over the process.

• How tractable are proposals for making advanced AI systems safe and aligned with human values?

• To what extent, if any, should governments regulate or restrict AI development? Can it be left to industry?

• Will AI liberate humanity or lead to greater inequality, less human agency and new dangers?

• What moral philosophy should guide these decisions - utilitarianism, human-centric values, or an AI-centric ethic?

  Source

• Primary: OECD AI Principles 2024 revision (Principle 1.2)

• Related: Universal Declaration of Human Rights, EU Charter of Fundamental Rights

  Context

  Human-centred values constitute OECD’s second core AI principle, establishing that technology must serve humanity’s fundamental rights and values rather than optimising for narrow technical or economic objectives. This principle provides the normative foundation for trustworthy AI development and deployment.

  Key Characteristics

• Human rights foundation: Grounding in internationally recognised rights frameworks

  • Democratic alignment: Compatibility with democratic governance and participation

  • Diversity respect: Acknowledging and accommodating human diversity

  • Human agency: Preserving meaningful human control over consequential decisions

  • Rule of law: Operating within legal frameworks and due process

    Relationships

  • Parent Concept: OECD AI Principle 2 (Human-Centred Values and Fairness)

  • Related Terms:

    • Fairness (OECD) (AI-0160)
    • Human Rights (AI-0166)
    • Democratic Values (AI-0167)
    • Accountability (AI-0165)
    • Well-Being (AI-0158)

  • Operationalised Through: Human oversight, transparency, fairness mechanisms

    Core Components

    Human Rights Respect

  • Protection of fundamental rights enumerated in international instruments

  • Non-discrimination and equality before the law

  • Privacy and data protection

  • Freedom of expression and assembly

  • Right to effective remedy

    Rule of Law

  • Legal certainty and predictability

  • Due process and procedural fairness

  • Judicial oversight and redress mechanisms

  • Proportionality of restrictions

  • Legitimate purposes for rights limitations

    Democratic Values

  • Participatory governance

  • Transparent decision-making processes

  • Pluralism and diversity of opinion

  • Electoral integrity protection

  • Free formation and expression of political will

    Diversity Recognition

  • Cultural diversity accommodation

  • Linguistic plurality

  • Accessibility for persons with disabilities

  • Age-appropriate design

  • Socio-economic context sensitivity

    Human Agency

  • Meaningful human control over consequential decisions

  • Ability to override automated systems

  • Understanding of AI system role in decisions

  • Preservation of human judgment and discretion

  • Avoidance of undue automation bias

    Implementation Considerations

    1. Rights-based design: Incorporating human rights by design from inception
    2. Human oversight: Ensuring meaningful human involvement in high-stakes decisions
    3. Contestability: Enabling challenge and review of AI-influenced decisions
    4. Cultural adaptation: Respecting diverse values whilst maintaining core rights
    5. Democratic participation: Including stakeholders in AI governance decisions

    OECD Framework Alignment

  • Dimension: People and Planet Context

  • Principle Number: P2 (part 1 of 2)

  • Actor Responsibility: All AI actors, with particular emphasis on providers and deployers

    Regulatory Context

    Human-centred values underpin:

  • EU AI Act fundamental rights framework (Articles 1, 27)

  • Prohibited AI practices (Article 5) preventing manipulation and exploitation

  • High-risk AI system requirements (Articles 8-15) ensuring human oversight

  • Transparency obligations enabling informed decision-making

    Challenges in Implementation

  • Cultural variation: Balancing universal rights with cultural diversity

  • Economic pressures: Resisting optimisation for profit over human values

  • Technical complexity: Translating values into technical requirements

  • Trade-offs: Managing tensions between competing values (privacy vs security)

  • Enforcement: Ensuring accountability for values violations

    Human Agency in Practice

    Human agency requires:

    1. Information: Understanding what AI system does and its limitations
    2. Override capability: Ability to intervene in automated processes
    3. Alternative pathways: Non-automated options for consequential decisions
    4. Meaningful control: Genuine influence over outcomes, not mere notification
    5. Cognitive support: Tools enabling informed judgment without overwhelming

    2024 Revision Updates

    The 2024 OECD revision strengthened human-centred values by:

  • Explicitly naming human agency as core component

  • Emphasising diversity alongside rights and democracy

  • Connecting to fairness as integrated principle

  • Clarifying human determination over consequential decisions

    Assessment Dimensions

    Human-centred values can be evaluated through:

  • Fundamental rights impact assessments

  • Democratic participation in AI governance

  • Accessibility and diversity metrics

  • Human agency preservation in decision processes

  • Cultural appropriateness evaluations

    Related Standards

  • ISO/IEC 42001:2023 - Artificial intelligence — Management system

  • IEEE 7000-2021 - Model Process for Addressing Ethical Concerns During System Design

  • ISO/IEC 23894:2023 - AI guidance on risk management

    See Also

  • Fairness (OECD) (AI-0160)

  • Human Rights (AI-0166)

  • Democratic Values (AI-0167)

  • Accountability (AI-0165)

  • Human Oversight (EU AI Act)

    ---

    Part of AI Grounded Ontology - OECD AI Principles Framework Aligned with OECD AI Principles 2024, UDHR, and EU Charter of Fundamental Rights

    Academic Context & Current Developments

    Human Centred Values in Artificial Intelligence: Current State (2024–2025)

    Human centred values in AI refer to the principle that AI systems should be designed and operated in ways that respect the rule of law, human rights, democratic values, and diversity, with appropriate safeguards to ensure human determination and control over consequential decisions (human agency), whilst incorporating mechanisms to protect fairness and human dignity. This approach is increasingly central to global AI policy, industry practice, and academic research.

    ---

    1. Current Developments

    The period 2024–2025 has seen a marked shift towards embedding human centred values into the core of AI development and deployment. Key trends include:

• Increased regulatory activity: Governments worldwide have intensified efforts to regulate AI in line with human rights and democratic principles. In 2024, U.S. federal agencies introduced 59 AI-related regulations—more than double the number in 2023—and issued by twice as many agencies. Similar regulatory momentum is evident in the EU, UK, and Asia-Pacific regions (Stanford HAI, 2025 AI Index Report).

• Advancements in value alignment: There is growing emphasis on AI value alignment—ensuring that AI systems act in accordance with shared human values and ethical principles. This includes tailoring AI to specific cultural, legal, and societal contexts, and engaging stakeholders from government, business, and civil society in the design process (WEF, AI Value Alignment).

• New evaluation benchmarks: The development of new benchmarks such as HELM Safety, AIR-Bench, and FACTS is enabling more rigorous assessment of AI systems for factuality, safety, and alignment with human values. These tools are helping to close the gap between recognising responsible AI (RAI) risks and taking meaningful action (Stanford HAI, 2025 AI Index Report).

• Global cooperation on AI governance: International organisations such as the OECD, EU, UN, and African Union have released frameworks focused on transparency, trustworthiness, and human centred values in AI. These frameworks are being adopted by governments and industry to guide responsible AI development (OECD AI Principles).

  ---

  2. Key Applications

  Human centred values are being applied across a range of sectors, with notable use cases in:

• Healthcare: AI systems are being designed to support, rather than replace, healthcare professionals. For example, platforms like Doctronic and Surgery Hero use AI to provide contextual insights and personalised advice, empowering both physicians and patients while maintaining human agency and oversight (Tredence, Human Centered AI).

• Education: Human centred AI is transforming education by creating personalised learning experiences that cater to individual student needs, while ensuring ethical use of technology and respect for student privacy (Lindenwood University, Human-Centered AI).

• Business: Companies are using AI to personalise marketing, streamline operations, and enhance customer satisfaction, all while ensuring transparency, fairness, and ethical decision-making. Human centred AI is helping businesses navigate the complex ethical landscape of AI adoption and build trust with customers and stakeholders (Lindenwood University, Human-Centered AI).

  ---

  3. Industry Standards

  Several international standards and frameworks have been developed to guide the implementation of human centred values in AI:

• OECD AI Principles: These principles promote the use of AI that is innovative and trustworthy, respects human rights and democratic values, and includes mechanisms for human agency and oversight. They cover non-discrimination, equality, freedom, dignity, autonomy, privacy, diversity, fairness, social justice, and internationally recognised labour rights (OECD AI Principles).

• UNESCO Recommendation on the Ethics of AI: This recommendation provides a global framework for the ethical development and use of AI, with a focus on human rights, diversity, and the protection of human dignity (UNESCO Ethics of AI).

• EU AI Act: The EU AI Act establishes a legal framework for AI that prioritises human centred values, including requirements for transparency, human oversight, and the protection of fundamental rights (European Commission, AI Act).

  ---

  4. Academic Literature

  Recent academic research has focused on the challenges and opportunities of aligning AI with human centred values:

• Dignum, V. (2024). AI Value Alignment: Guiding Artificial Intelligence Towards Shared Human Goals. Global Future Council White Paper. This paper explores the technical and societal challenges of aligning AI with human values and provides recommendations for policymakers and practitioners (WEF, AI Value Alignment).

• Russell, S. (2024). Humanity AI: Human-Centric Solutions for 2025. AI Tool Report. This article discusses the shift from AI-first to human-first thinking and the importance of embedding ethical frameworks, transparency, and inclusivity in AI system design (AI Tool Report, Humanity AI).

• Brookings Institution. (2024). Hype and harm: Why we must ask harder questions about AI and its alignment with human values. This article examines the ethical implications of AI and the need for robust mechanisms to ensure alignment with human values (Brookings, Hype and Harm).

  ---

  5. Notable Examples

  Several products, platforms, and organisations are leading the way in implementing human centred values in AI:

• Doctronic: An AI platform that provides contextual insights to healthcare professionals, supporting rather than replacing human decision-making (Tredence, Human Centered AI).

• Surgery Hero by C2-Ai: An AI system that empowers patients to prepare for surgery with personalised, actionable advice, enhancing patient agency and control (Tredence, Human Centered AI).

• OECD AI Principles: Adopted by over 50 countries, these principles provide a global standard for the ethical development and use of AI (OECD AI Principles).

• UNESCO Recommendation on the Ethics of AI: This global framework is being implemented by governments and organisations worldwide to ensure that AI respects human rights and democratic values (UNESCO Ethics of AI).

  ---

  6. Future Outlook

  The future of human centred values in AI is likely to be shaped by several emerging trends:

• Greater emphasis on cultural and contextual adaptation: As AI systems are deployed in diverse cultural and legal contexts, there will be a growing need to tailor AI to local values and norms.

• Enhanced stakeholder engagement: Continuous engagement with governments, businesses, civil society, and the public will be essential to ensure that AI systems reflect and uphold shared human values.

• Development of new evaluation tools: The creation of more sophisticated benchmarks and evaluation tools will enable more rigorous assessment of AI systems for alignment with human centred values.

• Increased global cooperation: International collaboration on AI governance and standard-setting will continue to grow, with a focus on transparency, trustworthiness, and the protection of human rights and democratic values.

• Growing public awareness and demand for ethical AI: As public awareness of AI’s ethical implications increases, there will be greater demand for AI systems that respect human centred values and protect human dignity.

  ---

  References

• OECD. (2019). OECD AI Principles. https://www.oecd.org/en/topics/sub-issues/ai-principles.html

• UNESCO. (2021). Recommendation on the Ethics of Artificial Intelligence. https://www.unesco.org/en/artificial-intelligence/recommendation-ethics

• World Economic Forum. (2024). AI Value Alignment: Aligning AI with Human Values. https://www.weforum.org/stories/2024/10/ai-value-alignment-how-we-can-align-artificial-intelligence-with-human-values/

• Stanford HAI. (2025). AI Index Report 2025. https://hai.stanford.edu/ai-index/2025-ai-index-report

• Lindenwood University. (2024). Human-Centered AI: What Is Human-Centric Artificial Intelligence? https://online.lindenwood.edu/blog/human-centered-ai-what-is-human-centric-artificial-intelligence/

• Tredence. (2024). Human Centered AI: Principles, Benefits, Challenges, and Industry. https://www.tredence.com/blog/human-centered-ai

• Brookings Institution. (2024). Hype and harm: Why we must ask harder questions about AI and its alignment with human values. https://www.brookings.edu/articles/hype-and-harm-why-we-must-ask-harder-questions-about-ai-and-its-alignment-with-human-values/

  Related Resources

  For more information, see:

• Metaverse

• MetaverseDomain

  Citations & Sources

  1. https://www.theaireport.ai/articles/humanity-ai-human-centric-solutions-for-2025

  2. https://www.weforum.org/stories/2024/10/ai-value-alignment-how-we-can-align-artificial-intelligence-with-human-values/

  3. https://online.lindenwood.edu/blog/human-centered-ai-what-is-human-centric-artificial-intelligence/

  4. https://www.tredence.com/blog/human-centered-ai

  5. https://www.oecd.org/en/topics/sub-issues/ai-principles.html

  6. https://hai.stanford.edu/ai-index/2025-ai-index-report

  7. https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai

  8. https://www.unesco.org/en/artificial-intelligence/recommendation-ethics

  9. https://www.brookings.edu/articles/hype-and-harm-why-we-must-ask-harder-questions-about-ai-and-its-alignment-with-human-values/

  Metadata

• Last Enriched: 2025-11-11

• Enrichment Source: Perplexity AI (Sonar)

• Verification Status: Automated enrichment - human review recommended

  Source

• Primary: OECD AI Principles 2024 revision (Principle 1.2)

• Related: Universal Declaration of Human Rights, EU Charter of Fundamental Rights

  Context

  Human-centred values constitute OECD’s second core AI principle, establishing that technology must serve humanity’s fundamental rights and values rather than optimising for narrow technical or economic objectives. This principle provides the normative foundation for trustworthy AI development and deployment.

Provenance