Record STMT-W7EC837Soft law

Recommendation on the Ethics of Artificial Intelligence

Recommandation sur l’éthique de l’intelligence artificielle

Recommendation International lawGlobaladopted

The UNESCO Recommendation on the Ethics of Artificial Intelligence is a globally adopted normative instrument that establishes ethical values and principles to guide the development and deployment of AI technologies. It provides a comprehensive framework centered on human dignity, human rights, gender equality, and the protection of the environment. Furthermore, it outlines specific policy areas for member states to implement, ranging from ethical impact assessment and data governance to education, culture, and labor.

Summary and content fields are drawn from the document text and are pending editorial review.

Who issued it

UNESCOIntergovernmental

Lead issuer

Legal forceSoft law

Under whose authorityInternational law

Judged fromfull text

Who it governs

National governmentState or provincial governmentRegulatorsPublic sector bodiesCompaniesAI developers and providersAI deployers and usersResearchers and academics

This document is for member states, policymakers, AI actors, and stakeholders across public and private sectors seeking guidance on the responsible and ethical governance of AI.

Sectors

Cross-sector

What the document contains

What's covered

The document addresses ethical questions raised by artificial intelligence across its entire lifecycle, including research, design, development, deployment, and end-of-life. It sets out fundamental values and principles, and details strategic policy areas such as data policy, international cooperation, environmental protection, gender equality, culture, education, and health.

What's missing

It does not provide a single rigid definition of AI, nor does it establish binding international enforcement mechanisms or penalties, relying instead on voluntary implementation by member states.

Key takeaways

The recommendation sets a global normative standard for ethical AI governance centered on human dignity and human rights.
It prohibits practices such as social scoring and mass surveillance systems.
It emphasizes the necessity of human oversight, ultimate human responsibility, and explainability in AI applications.
Member states are encouraged to implement ethical impact assessments and regulatory sandboxes.
It stresses the importance of international cooperation to bridge the digital and technological divide for developing nations.

Principles it states

Respect, protection and promotion of human rights, fundamental freedoms and human dignityEnvironment and ecosystem flourishingEnsuring diversity and inclusionLiving in peaceful, just and interconnected societiesProportionality and do no harmSafety and securityFairness and non-discriminationSustainabilityRight to privacy and data protectionHuman oversight and determinationTransparency and explainabilityResponsibility and accountabilityAwareness and educationMulti-stakeholder and adaptive governance and collaboration

Outline

· PRÉAMBULE
· I. CHAMP D’APPLICATION
· II. BUTS ET OBJECTIFS
· III. VALEURS ET PRINCIPES
· IV. DOMAINES D’ACTION STRATÉGIQUE
· V. SUIVI ET ÉVALUATION
· VI. UTILISATION DE LA PRÉSENTE RECOMMANDATION
· VII. PROMOTION DE LA PRÉSENTE RECOMMANDATION
· VIII. DISPOSITIONS FINALES

Issued23 November 2021

Length18,224 words

FormatNot recorded

Scopeglobal

How it scores

Pending review

Each score is backed by a quote verified against the document text. Editorial review has not yet confirmed them.

Accountability

4/5

Clear responsibility for AI outcomes

La responsabilité éthique des décisions et actions fondées d’une quelconque manière sur un système d’IA devrait toujours incomber en dernier ressort aux acteurs de l’IA

III.2 PRINCIPES - Responsabilité et redevabilité, Paragraph 42

screening score · 2026-10-23

Fairness

4/5

Non-discrimination and equitable treatment

Les acteurs de l’IA doivent promouvoir la justice sociale, garantir l’équité et lutter contre les discriminations en tous genres

III.2 PRINCIPES - Équité et non-discrimination, Paragraph 28

screening score · 2026-10-23

Human Dignity

5/5

Respect for the inherent worth of persons

La dignité inviolable et intrinsèque de chaque être humain est le fondement du système universel, indivisible, inaliénable, interdépendant et indissociable des droits de l’homme

III.1 VALEURS - Respect, protection et promotion des droits de l’homme, Paragraph 13

screening score · 2026-10-23

Human Oversight

4/5

Humans remain in control of consequential decisions

les systèmes d’IA ne devraient pas être utilisés à des fins de notation sociale ou de surveillance de masse.

III.2 PRINCIPES - Principes de proportionnalité et d’innocuité, Paragraph 26

screening score · 2026-10-23

Privacy

4/5

Protection of personal data and private life

Des cadres de protection des données et des mécanismes de gouvernance adéquats doivent être mis en place selon une approche multipartite

III.2 PRINCIPES - Droit au respect de la vie privée et protection des données, Paragraph 33

screening score · 2026-10-23

Transparency

4/5

Openness about how AI systems work and are used

Les États membres devraient définir des exigences claires en matière de transparence et d’explicabilité des systèmes d’IA

DOMAINE STRATÉGIQUE 2: GOUVERNANCE ET GESTION ÉTHIQUES, Paragraph 70

screening score · 2026-10-23

Impact Assessment

4/5

Required assessment of an AI system's risks/impacts before or during use

Les États membres devraient mettre en place des cadres relatifs aux évaluations de l’impact tels que l’évaluation de l’impact éthique

DOMAINE STRATÉGIQUE 1: ÉVALUATIONS DE L’IMPACT ÉTHIQUE, Paragraph 50

screening score · 2026-10-23

Topics

ai_ethics human_rights transparency governance data_protection environmental_sustainability