Adopted by all 193 UNESCO member states at the General Conference in November 2021, the Recommendation is the first global normative instrument on AI ethics. AIPIA integrates its four values, ten principles, and eleven policy areas into its Code of Ethics and training curriculum.
The UNESCO Recommendation on the Ethics of Artificial Intelligence is the first global instrument on AI ethics, adopted unanimously at the 41st session of the General Conference in November 2021. All 193 member states of UNESCO endorsed it — making it, by membership reach, the most universally accepted document on AI ethics in existence.
Where the OECD Principles bind 47 mostly higher-income countries and the Council of Europe Convention focuses on Europe and partner democracies, the UNESCO Recommendation extends to the Global South: low- and middle-income countries that are net importers of AI systems and whose populations bear the consequences of design choices made elsewhere. This breadth gives the Recommendation a distinctive moral weight in international discussions on AI governance.
The Recommendation is built around four values, ten principles, and eleven policy action areas. The values establish the ethical ground; the principles operationalise the values; the policy action areas direct concrete government action. Member states report periodically to UNESCO on implementation.
A distinguishing feature of the Recommendation is its explicit attention to environmental impact, cultural diversity, and the protection of vulnerable groups — dimensions less central to other AI frameworks. The 2024 update following the General Conference's review reinforced these dimensions in light of generative AI's environmental and cultural implications.
The Recommendation begins with four values that frame everything that follows. They are not optional preferences; they are the bedrock against which AI policy choices are evaluated.
AI systems must respect, protect, and promote human rights and human dignity throughout their lifecycle. No AI application may infringe these in pursuit of efficiency or commercial gain.
AI must be assessed against its environmental impact. Energy use, materials, water consumption, and broader ecological footprint are part of responsible AI development — not externalities.
AI systems must respect, promote, and protect cultural diversity, gender equality, and the participation of historically excluded groups. Inclusion is built into design, not retrofitted afterwards.
AI should enable peaceful coexistence, avoid weaponisation that undermines international law, and foster cooperation across borders. AI applications inconsistent with peaceful interconnection are incompatible with the Recommendation.
The Recommendation's ten principles translate values into operational requirements: proportionality and do-no-harm; safety and security; fairness and non-discrimination; sustainability; right to privacy and data protection; human oversight and determination; transparency and explainability; responsibility and accountability; awareness and literacy; multi-stakeholder and adaptive governance and collaboration. AIPIA's Code of Ethics maps each principle to specific professional commitments — see code-of-ethics for the full mapping.
Beyond values and principles, the Recommendation directs member states to take concrete policy action across eleven areas. Most national AI strategies adopted since 2021 are structured against these areas.
AIPIA members work increasingly outside the EU: consulting clients in the UAE, the UK, the US, and Latin America. The UNESCO Recommendation is the single most widely endorsed AI ethics framework, which makes it valuable wherever an EU-only reference would not suffice.
Practical applications include cross-border consulting (citing UNESCO principles when working with governments and NGOs), capacity-building programmes in countries implementing the Recommendation, alignment of corporate AI policies with a universally recognised standard, and contributing to AIPIA training material that serves international audiences.
AIPIA aligns its credentials, working groups, and member briefings with UNESCO's Readiness Assessment Methodology and Ethical Impact Assessment — instruments developed by UNESCO to help countries and organisations operationalise the Recommendation.
AIPIA's Code of Ethics maps to the UNESCO Recommendation. Members operating internationally gain a coherent professional anchor endorsed by 193 member states.