The AI triad: divergent technological pathways and their global implications

Artificial intelligence (AI) is rapidly transforming the global landscape in science, technology, and governance. However, how the policy paradigms shape the respective AI ecosystems, technological capabilities, and the broader global governance landscape remains elusive. To answer this question, we conduct a mixed-methods approach combining qualitative policy analysis with quantitative technical assessment to connect high-level policy paradigms to the materials, technical capabilities, and innovation pathways of AI systems in China, the United States, and the European Union. By synthesizing policy documentation, technical performance data, and industrial analysis, we identify three fundamentally divergent development paradigms that are creating irreversible technological path dependencies. The United States maintains structural advantages in foundational model innovation and semiconductor design; China demonstrates exceptional capabilities in rapid commercialization and domain-specific applications; while the European Union adopts a distinctive path focused on trustworthy AI and regulatory standard-setting. Our findings reveal significant technological fragmentation across multiple dimensions: architectural approaches, computational infrastructures, talent distribution, and application ecosystems. This fragmentation poses substantial challenges to global standardization, international cooperation, and AI safety governance. The research concludes that current trajectories suggest increasing technological divergence rather than convergence, thereby necessitating new frameworks for managed coexistence.