Governance efficiency and upgrade pathways of international generative AI policies and regulations

Abstract

Analyzing the governance efficiency of policies and regulations on generative artificial intelligence (GAI) not only facilitates the advancement of GAI technological innovation and theoretical research but also enhances the precision and efficiency of information governance across nations. First, based on governance theory, institutional theory, resource-based theory, and administrative ecology theory, this paper analyzes the factors influencing the governance efficiency of GAI policies & regulations from three dimensions: government governance, resource endowment, and technology environment. Second, this paper examines the policies and regulations on GAI from 24 countries as samples. Employing the fsQCA and NCA method, along with PMC index evaluation results, this paper explores potential pathways to enhance the governance efficiency of GAI policies and regulations. Third, the configurational pathway analysis of governance efficiency in GAI policies and regulations identifies six critical influencing factors: policy and regulatory quality, government actions, venture capital investment, AI governance capacities, public stakeholder engagement, and AI safety mechanisms. Finally, through necessity analysis, configurational analysis, and robustness testing of these six factors, the paper reveals that technology-resource driven, policy-actor coordinated, and government-resource mediated implementation configurations can effectively achieve high-level governance efficiency in GAI policies and regulations. Therefore, it provides a reference for optimizing the governance practice of GAI policies and regulations.