America’s AI action plan and global governance: convergences, divergences, and theoretical implications

Abstract

This study critically evaluates America’s AI Action Plan (AAP) against best practices in artificial intelligence regulation and governance identified in recent scientific literature. Based on an integrative review of 100 peer-reviewed articles indexed in Scopus, the analysis contrasts the AAP’s three pillars, innovation, infrastructure, and diplomacy, with key governance dimensions, namely accountability, sustainability, and multilateralism. The findings indicate selective convergences, especially in cybersecurity, technological sovereignty, and standardization, alongside structural divergences concerning rights protection, environmental responsibility, and global cooperation. Drawing on these results, the paper proposes a theoretical framework that conceptualizes AI governance as a multidimensional and adaptive regime integrating innovation, regulation, ethics, sustainability, and epistemic co-production. The study offers practical value by informing policymakers and regulators on how to strengthen adaptive and risk-based regulatory design, and it contributes to the academic debate by framing AI governance as an emergent global normative system. The recommendations are intentionally delimited to public policy and governance arrangements, with implications for global coordination and the design of national AI strategies, rather than prescribing technical solutions for specific models or systems. Limitations include the exclusive use of Scopus and the focus on a single national strategy; future research should expand comparative analyses across regions and examine longitudinal transformations in AI governance regimes.