OECD AI

Stefan Coetzeecreated 2026-05-12updated 2026-06-082 min readMOC
Vault note, not reviewed against the source. Written in the knowledge vault on 2026-05-12 by models working with Stefan Coetzee and published as it stands, with private addresses, e-mail addresses and an employer name redacted. Check claims against the primary source before relying on them.

OECD AI Principles Cluster

Map of OECD AI Principles. Adopted 22 May 2019; revised May 2024. International voluntary principles for trustworthy AI. First intergovernmental AI standard. Influenced ISO 42001, NIST AI RMF, EU AI Act, multiple national AI strategies.

Anchors

Provenance

  • OECD AI Expert Group (AIGO) — established 2018.
  • OECD AI Principles (Recommendation of the Council on Artificial Intelligence) — adopted 22 May 2019.
  • G20 AI Principles (June 2019) — drew on OECD.
  • 2024 update — revised May 2024 in response to generative AI maturation. Addresses misinformation, mis/disinformation, sustainability.

What OECD AI Principles are

International voluntary principles for trustworthy AI. Two parts:

  • Five values-based principles for responsible AI stewardship.
  • Five recommendations for national policies and international cooperation.

OECD member countries (38 countries) plus non-member adherents committed to implementing.

Five values-based principles

  1. Inclusive growth, sustainable development, well-being.
  2. Human-centred values and fairness (incl. respect for rule of law, human rights, democratic values).
  3. Transparency and explainability.
  4. Robustness, security and safety.
  5. Accountability.

Detail in OECD AI Values-Based Principles.

Five recommendations for national policies

  1. Investing in AI research and development.
  2. Fostering a digital ecosystem for AI.
  3. Shaping an enabling policy environment for AI.
  4. Building human capacity and preparing for labour market transition.
  5. International co-operation for trustworthy AI.

2024 update changes

  • Mis/disinformation addressed.
  • Information integrity strengthened.
  • Sustainability dimension expanded.
  • GenAI considerations incorporated.

Adherents

OECD members (38 countries) + non-member adherents (Argentina, Brazil, Costa Rica, Egypt, Malta, Peru, Romania, Singapore, Ukraine, and growing).

Total ~50 jurisdictions formally adherent.

Why this matters

  • First international AI standard influences subsequent frameworks (ISO 42001, NIST AI RMF, EU AI Act).
  • Diplomatic baseline for cross-jurisdictional AI governance.
  • National strategies reference.
  • Procurement-readable for international vendors.
  • NIST AI RMF — US implementation consistent with OECD principles.
  • ISO 42001 — international standard consistent with OECD.
  • EU AI Act — regulatory implementation aligned.

See also

OECD AI Cluster (pillars MOC) · position · anchors · NIST AI RMF Cluster · ISO 42001 Cluster