Regional forums representing 62% of the world's economy adopt a practical, flexible approach to achieving shared AI safety outcomes to suit their local needs.
This paper challenges the AI arms race narrative and explores evidence-based strategies in non-Western regional forums for international cooperation in AI development, and meaningful global collaboration.
The research is based on an analysis of 327 policy documents from APEC, ASEAN, the African Union, and the G20 to show that the global majority – representing 62% of global GDP and billions of citizens – is constructing a strategic “third path” operating beneath and beyond superpower rivalry: pragmatic pluralism.
Key insights for international cooperation include:
Functional equivalence: the path to interoperability is not unified laws, but recognition that different mechanisms can achieve the same security and accountability outcomes. For example, ASEAN’s voluntary standards and the EU’s legal compliance can be functionally equivalent for mitigating algorithmic bias.
Strategic power in convening: governance power is shifting to those who host the critical debates. ASEAN and the AU derive influence by asserting multi-alignment and co-design rights in global governance, without having to choose a single technological bloc.
Problem-driven governance: cooperation emerges from solving shared, concrete challenges (a “problem-up” approach) like climate adaptation, education, and integrating the informal economy. This builds trust and momentum that abstract principle-based negotiations often lack.
Capacity is enforcement: In many regions, capability-building IS the governance mechanism. Compliance is achieved through peer support and skills transfer, which is more effective than punitive sanctions where implementation capacity is low.
Report: Pragmatic pluralism: regional AI governance beyond great power competition
News brief: Global majority is building international AI governance through cooperation, not competition
This second publication expands on the framework of the first paper, Moving Beyond Competition, which analysed policy documents from Western-run regional forums – NATO, the UN, WTO and OECD – to propose domain-specific collaboration as the path forward to advance global AI governance while respecting national sovereignty.
