UNESCO Recommendation on AI Ethics (2021)
The UNESCO Recommendation on the Ethics of Artificial Intelligence was adopted in November 2021 by all 194 member states. This instrument was the world's first global normative framework for ethical AI development and use. It emphasizes on the importance of centring human rights, dignity, and sustainability as core pillars, aiming to guide policymakers, developers, and users in harnessing AI's potential while taking care to mitigate risks.
Core Values
The document establishes four foundational values to anchor AI ethics. Human rights and dignity must remain paramount, ensuring that all AI respects, protects, and fulfils these obligations throughout its lifecycle from design to use. Diversity and inclusiveness promote equity, and can counter biases that could exacerbate inequalities. The guidelines also call for environmental sustainability, which requires AI actors to align their systems with ecological protection and restoration goals. Finally, creating a culture of integrity and awareness is essential in order to foster ethical responsibility among all stakeholders.
Key Principles
The UNESCO Guidelines comprise ten principles in all, to operationalize these values into actionable norms. These principles are:
Proportionality and "do no harm," which call for AI use to be necessary, legitimate, and risk-assessed to avoid undue impacts.
Safety and security prioritize robust systems that are resistant to errors, attacks, and/or misuse.
Fairness and non-discrimination call for the prevention of bias and promotion of justice.
Privacy protection safeguards data across all stages of the AI lifecycle.
Human oversight is essential to ensure meaningful control, instead of complete autonomy in critical decisions that involve life-and-death scenarios.
Transparency and explainability call for decisions that can be understood and meaningfully contested.
Responsibility and accountability assign clear human liability.
Multi-stakeholder governance and sustainability embed these principles in broader ecosystems.
Policy Action Areas
The Recommendation translates the principles into 11 sectoral policies for practical implementation. These are as follows:
Ethical Impact Assessment: Member states should carry out ethical impact assessments to identify and assess the benefits, concerns, and risks of AI systems and appropriate risk prevention, mitigation, and monitoring measures among other assurance mechanisms.
Ethical Governance and Stewardship: Member States should ensure that AI governance mechanisms are inclusive, transparent, multidisciplinary, multilateral (this includes the possibility of mitigation and redress of harm across borders) and multi-stakeholder. In particular, governance should include aspects of anticipation, and effective protection, monitoring of impact, enforcement and redress.
Data Policy: Member States should work to develop data governance strategies that ensure the continual evaluation of the quality of training data for AI systems including the adequacy of the data collection and selection processes, proper data security and protection measures, as well as feedback mechanisms to learn from mistakes and share best practices among all AI actors.
Development and International Cooperation: Member States and transnational corporations should prioritize AI ethics by including discussions of AI-related ethical issues into relevant international, intergovernmental and multi-stakeholder fora.
Environment and Ecoystems: Member States and business enterprises should assess the direct and indirect environmental impact throughout the AI system life cycle, including, but not limited to, its carbon footprint, energy consumption and the environmental impact of raw material extraction for supporting the manufacturing of AI technologies, and reduce the environmental impact of AI systems and data infrastructures. Member States should ensure compliance of all AI actors with environmental law, policies and practices.
Gender: Member States should ensure that the potential for digital technologies and artificial intelligence to contribute to achieving gender equality is fully maximized, and must ensure that the human rights and fundamental freedoms of girls and women, and their safety and integrity are not violated at any stage of the AI system life cycle. Moreover, Ethical Impact Assessment should include a transversal gender perspective.
Culture: Member States are encouraged to incorporate AI systems, where appropriate, in the preservation, enrichment, understanding, promotion, management and accessibility of tangible, documentary and intangible cultural heritage, including endangered languages as well as indigenous languages and knowledges, for example by introducing or updating educational programmes related to the application of AI systems in these areas, where appropriate, and by ensuring a participatory approach, targeted at institutions and the public.
Education and Research: Member States should work with international organizations, educational institutions and private and non-governmental entities to provide adequate AI literacy education to the public on all levels in all countries in order to empower people and reduce the digital divides and digital access inequalities resulting from the wide adoption of AI systems.
Communication and Information: Member States should use AI systems to improve access to information and knowledge. This can include support to researchers, academia, journalists, the general public and developers, to enhance freedom of expression, academic and scientific freedoms, access to information, and increased proactive disclosure of official data and information.
Economy and Labour: Member States should assess and address the impact of AI systems on labour markets and its implications for education requirements, in all countries and with special emphasis on countries where the economy is labour-intensive. This can include the introduction of a wider range of “core” and interdisciplinary skills at all education levels to provide current workers and new generations a fair chance of finding jobs in a rapidly changing market, and to ensure their awareness of the ethical aspects of AI systems. Skills such as “learning how to learn”, communication, critical thinking, teamwork, empathy, and the ability to transfer one’s knowledge across domains, should be taught alongside specialist, technical skills, as well as low-skilled tasks. Being transparent about what skills are in demand and updating curricula around these are key.
Health and Social Well-Being: Member States should endeavour to employ effective AI systems for improving human health and protecting the right to life, including mitigating disease outbreaks, while building and maintaining international solidarity to tackle global health risks and uncertainties, and ensure that their deployment of AI systems in health care be consistent with international law and their human rights law obligations. Member States should ensure that actors involved in health care AI systems take into consideration the importance of a patient’s relationships with their family and with health care staff.
Implementation, Monitoring, and Evaluation Mechanisms
Member states commit to readiness assessments to gauge national AI ethics capacity, followed by tailored action plans. AI actors, namely governments, firms, and researchers must conduct ethics impact assessments before deployment, evaluating risks to rights, biases, and sustainability. Monitoring involves periodic reporting to UNESCO, with a global observatory tracking progress. Public awareness campaigns build literacy, empowering citizens to engage with AI.
Risk Management and Oversight
Risks are classified by impact levels, requiring mitigation strategies like audits and redress mechanisms. Human determination overrides AI in high-stakes areas; explainability supports democratic accountability. Liability remains attributable to humans, not machines.
