Evaluating the Effectiveness of Legal Regulation of AI-Generated Content and Optimizing Regulatory Pathways

Abstract

The rapid expansion of generative artificial intelligence (AI) has transformed fields such as media, education, and politics, while simultaneously raising urgent legal and ethical challenges. Issues including disinformation, copyright disputes, and threats to democratic processes have intensified debates on how best to regulate AI-generated content. Yet existing research often focuses on single jurisdictions, lacks systematic evaluation criteria, and rarely explores pathways for globally adaptive governance, leaving both scholars and policymakers without a coherent framework for assessing effectiveness. This study addresses these gaps by developing a multidimensional evaluative framework grounded in technology-law co-evolution, risk-benefit balancing, and human rights-based governance. Employing comparative legal analysis, doctrinal review, and case studies from the European Union, the United States, and China, the research assesses regulatory effectiveness across three dimensions: enforceability, adaptability, and rights protection. Findings indicate that the EU ensures comprehensive safeguards but struggles with consistent enforcement, the U.S. emphasizes expressive freedom but remains fragmented, and China achieves strong compliance at the cost of transparency and rights. The paper contributes theoretically by integrating diverse perspectives into a coherent model and practically by proposing a hybrid regulatory pathway that combines hard law, soft law, and oversight mechanisms. This approach provides a roadmap for more balanced and adaptive governance of AI-generated content.