OECD / Global · principles

OECD Principles on Artificial Intelligence

Intergovernmental principles for innovative and trustworthy AI, updated as AI capabilities and policy needs evolve.

voluntaryOECDVerified 23 Aug 2026

Applicability

Policy makers and AI actors using the principles as a responsible-AI reference.

Enforcement context

Non-binding principles, often reflected in governance policy.

Regulatory timeline

Material dates and status changes.

22 May 2019

OECD AI Principles

Intergovernmental principles for innovative and trustworthy AI, updated as AI capabilities and policy needs evolve.

Source for this event →

CRG implementation mapping

Operational obligations and evidence expectations.

OECD-ACC

Accountability

Maintain accountability for proper functioning and respect for applicable principles.

Actor: AI actors · Domain: governance

Evidence: Named accountability, documentation and auditability.

Mapped CRG assurance controls: GOV-01GOV-03GOV-05
OECD-HUM

Human-centred values

Design and use AI consistent with human rights, democratic values and appropriate safeguards.

Actor: AI actors · Domain: human impact

Evidence: Impact analysis, safeguards, oversight and grievance mechanisms.

Mapped CRG assurance controls: FAIR-01FAIR-02FAIR-03
OECD-ROB

Robustness, security and safety

Maintain robustness, security and safety throughout the AI system lifecycle.

Actor: AI actors · Domain: security

Evidence: Testing, resilience, monitoring and incident response evidence.

Mapped CRG assurance controls: SEC-01SEC-03SEC-05
Primary source

Verify the underlying instrument.

Cybatar's mapping is designed for AI governance and assurance work. Legal interpretation remains anchored to the current primary text and competent authority guidance.

Open OECD source