Global AI Governance Frameworks Show Signs of Convergence
Recent analyses suggest that national AI governance strategies in China, the EU, and the US are beginning to converge around risk-based, sectoral, and interoperable frameworks. The trend reflects ongoing harmonization efforts between international standards bodies like ISO, IEEE, and the ITU to establish common principles for model risk management and auditability.
The European Union's AI Act, formally approved on May 21, 2024, establishes the world's first comprehensive, risk-based legal framework for AI. Its extraterritorial reach means non-EU companies serving the EU market must comply, effectively setting a global regulatory baseline. While the EU's approach is horizontal and rights-focused, the U.S. pursues a sector-specific model, relying on the National Institute of Standards and Technology's (NIST) AI Risk Management Framework as a voluntary standard for agencies and industry. In contrast, China's state-driven approach has produced specific regulations for algorithmic recommendations, deep synthesis, and generative AI to advance its goal of global AI leadership by 2030. At the core of international technical harmonization is the joint committee ISO/IEC JTC 1/SC 42, where over 60 countries collaborate on foundational standards. This body has published key documents like ISO/IEC 42001, which provides a certifiable framework for an AI management system, and ISO/IEC 23894 for AI risk management. High-level consensus is building through initiatives like the December 2025 "Seoul Statement," where the ITU, ISO, and IEC pledged to advance human-centric AI standards. This followed the May 2024 Seoul Declaration, where the EU and ten other nations committed to fostering interoperable AI governance frameworks. This standards development is unfolding amid intense geopolitical competition, as nations increasingly pursue "sovereign AI" strategies to bolster domestic economies and national security. The push
