Key OECD AI governance initiatives
OECD AI Governance: Framework for International Development and M&E
OECD AI governance provides the primary intergovernmental framework for trustworthy artificial intelligence, structured around three pillars: normative standards (OECD AI Principles, harmonised AI system definition), global policy intelligence (OECD.AI Observatory), and implementation guidance (public sector frameworks, capacity building). For development and M&E professionals, these instruments shape how AI is regulated, adopted, and evaluated in partner countries and donor programmes.
Introduction
The Organisation for Economic Co-operation and Development (OECD) has established the most influential set of intergovernmental AI governance instruments, directly referenced by the G7, G20, and United Nations. Since 2019, its framework has evolved to address the specific needs of international development: inclusive growth, sustainable development goals (SDGs), and responsible AI adoption in low- and middle-income countries. This article maps the core OECD AI governance initiatives relevant to development cooperation and monitoring & evaluation (M&E).
What are the core OECD AI normative frameworks for development?
Two instruments form the normative backbone: the OECD Council Recommendation on AI (the AI Principles) and the OECD definition of an AI system. Together they provide a common language and value base for AI governance across jurisdictions.
OECD AI Principles (2019, revised 2024)
The first intergovernmental AI standard, adopted by all OECD members and other adherents. It sets five principles for trustworthy AI:
- Inclusive growth and sustainable development: AI should contribute to inclusive growth, reduce inequalities, and support the SDGs.
- Human-centred values and fairness: AI must respect human rights and democratic values.
- Transparency and explainability: AI actors should ensure meaningful information about AI systems is disclosed.
- Robustness, security and safety: AI systems should be robust throughout their lifetimes.
- Accountability: AI actors should be accountable for proper functioning and adherence to principles.
These principles are explicitly linked to advancing inclusive growth and the SDGs, making them a de facto governance reference for development policy.
OECD definition of an AI system (revised 2023)
"An AI system is a machine-based system that, for explicit or implicit objectives, infers, from the input it receives, how to generate outputs such as predictions, content, recommendations, or decisions that can influence physical or virtual environments." This harmonised definition is used by dozens of governments and international organisations as a legal anchor, supporting regulatory interoperability—critical for development programmes operating across multiple legal systems.
How does the OECD AI Policy Observatory support development actors?
The OECD.AI Policy Observatory (launched 2020) is the leading global repository for AI policies, strategies, and data. It aggregates over 850 AI policy initiatives from 60+ countries and the EU, plus indicators on R&D, skills, jobs, and AI deployment. Development agencies and M&E specialists use OECD.AI to benchmark national AI readiness, identify policy gaps, and inform capacity-building programmes.
- country readiness benchmarks
- policy trend analysis
- AI incidents monitor
- live policy database
- SDG alignment tools
In collaboration with the European Commission’s AI Watch, the Observatory improves data comparability that directly informs EU external action and development cooperation programming.
Which OECD governance bodies shape AI policy for development?
Intergovernmental body overseeing the AI Recommendation’s implementation. It produces practical guidance on risk management, public-sector AI, and accountability—increasingly used in humanitarian and development governance frameworks.
350+ experts across thematic groups (AI futures, risk, incidents, data privacy, compute, climate). Provides multi-stakeholder advice that feeds into guidance used by governments and international organisations.
What OECD frameworks guide AI in the public sector and development contexts?
The Governing with Artificial Intelligence in Government (2025 report) proposes a governance framework for public-sector AI built around three pillars: enablers (institutions, infrastructure, data, skills), guardrails (rules, accountability, transparency, oversight), and engagement (participation of citizens, civil society, business). This is highly relevant for ministries and agencies in partner countries using AI in social protection, taxation, health, and administrative systems financed by development cooperation.
Horizontal Project on “Thriving with AI”
A cross-cutting initiative providing evidence on how economies and societies adapt to AI. Its framing of inclusive growth, social protection, and labour-market transitions is directly used in development discourse—for example, on digital transformation and just transitions in low-income countries.
How does the OECD support the UN and global AI governance?
The OECD acts as host institution for the Global Partnership on AI (GPAI), providing analytical and secretariat support. GPAI working groups have produced guidance on AI in fragile contexts, data governance, and responsible innovation—resources increasingly used by development actors. In 2024, the OECD and UN announced enhanced collaboration to support the UN’s emerging global AI governance mechanisms under the Global Digital Compact. This positions the OECD as a key technical partner for UN-led AI-for-development and AI governance capacity-building initiatives.
Essential OECD AI milestones for development practitioners
If curating a development‑focused AI governance timeline, prioritise:
- 2019 / 2024: Adoption and revision of the OECD AI Principles (normative anchor).
- 2023: Revised OECD AI system definition (regulatory harmonisation).
- 2020 onward: Creation and scaling of OECD.AI Policy Observatory (evidence base).
- 2020: Establishment of Working Party on AI Governance and ONE AI network (expert guidance).
- 2025: “Governing with AI in Government” framework (public-sector implementation).
- 2024: Enhanced OECD–UN collaboration (global governance architecture).
Frequently asked questions about OECD AI governance
What is the OECD AI Principles' relevance to development?
The principles explicitly reference inclusive growth, reducing inequalities, and supporting the SDGs, making them the primary normative framework for AI-for-development policy.
How does the OECD define an AI system?
As a machine-based system that infers how to generate outputs (predictions, content, recommendations) influencing environments. This definition is used by many low- and middle-income countries adopting AI regulations.
What is OECD.AI and how can M&E professionals use it?
OECD.AI is an online policy observatory with 850+ AI policies and readiness indicators. M&E professionals use it to benchmark national AI capacity and inform programme design.
How does the OECD support AI governance in the public sector?
Through the 2025 report “Governing with AI in Government”, which provides a framework of enablers, guardrails, and engagement—directly applicable to AI use in social protection, health, and taxation systems in partner countries.
Authoritative resources
- OECD AI Policy Observatory – official portal (aggregated policies, data, and country readiness).
- OECD Council Recommendation on AI (full text) – the AI Principles (2019/2024).
- OECD definition of an AI system (2023) – explanatory notes and use cases.
- OECD Horizontal Project “Thriving with AI” – inclusive growth and adaptation.
- Global Partnership on AI (GPAI) – guidance on AI in fragile contexts and responsible innovation.
- OECD–UN collaboration on AI governance (2024) – joint statement and work programme.
Conclusion
OECD AI governance offers the most comprehensive intergovernmental framework for aligning artificial intelligence with development objectives. From the foundational AI Principles and harmonised definition to the OECD.AI Observatory and public-sector guidance, these instruments shape how governments, multilaterals, and M&E practitioners design, evaluate, and govern AI systems in development contexts. As the OECD deepens its collaboration with the UN under the Global Digital Compact, its frameworks will remain central to capacity building and policy coherence in AI-for-development.
