United Nations and AI Governance

Global AI governance is the set of international agreements, standards and institutions through which countries attempt to coordinate the oversight of artificial intelligence. Unlike national laws, most of these instruments are soft law — non-binding declarations and principles that shape domestic regulation — with the exception of regional treaties such as the Council of Europe's Framework Convention.^[1]

Key Milestones

UNESCO Recommendation on the Ethics of Artificial Intelligence (2021). Adopted in November 2021 by all 193 member states, it was the first global standard-setting instrument on AI ethics, establishing principles of proportionality, fairness, transparency and environmental sustainability that national AI strategies continue to cite.^[2]

OECD AI Principles (2019, updated 2024). Endorsed by the G7 and later adopted by more than 40 countries, these became the reference text for trustworthy AI in trade and economic policy forums.^[3]

Bletchley Declaration (November 2023). Signed by 28 countries and the EU at the first AI Safety Summit in the United Kingdom, it acknowledged frontier-model risks without creating any compliance mechanism — a pattern typical of summit-level AI diplomacy.^[4]

Council of Europe Framework Convention on AI (2024). Adopted 17 May 2024 and opened for signature on 5 September 2024 in Vilnius, this is the first legally binding international treaty on AI, covering human rights, democracy and the rule of law in AI lifecycle governance; the United States, United Kingdom and European Union signed on the opening day.^[1]

UN General Assembly resolutions. In April 2024 the General Assembly adopted, by consensus, its first AI resolution — Seizing the opportunities of safe, secure and trustworthy artificial intelligence systems for sustainable development — followed by a second resolution in March 2025 focused on inclusive AI capacity building for developing countries.^[5] The Secretary-General's Independent International Scientific Advisory Panel on AI released Governing AI for Humanity in September 2024, recommending a standing international scientific panel, a global dialogue on AI governance and a dedicated UN AI office to coordinate the fragmented system.^[6]

Why It Matters for AI

AI development is concentrated in a handful of companies and countries, while its effects — labour displacement, election interference, autonomous weapons — cross every border. Without shared baselines, a company can shop for the weakest jurisdiction, and developing countries risk being written out of rule-setting entirely. The soft-law architecture lets countries adopt principles without surrendering sovereignty, but critics note it produces declarations faster than it produces accountability: only the Council of Europe treaty carries legal obligations, and its membership is European.

>See Also

• EU AI Act
• OECD AI Principles
• Bletchley Declaration
• ASEAN AI Governance
• Malaysia AI Governance Framework
• AI Safety