Use of generative AI and copyright protection

empirical evidence from digital content creators in China

Abstract

This study examines how the intensity of Generative Artificial Intelligence (GenAI) use influences copyright protection behavior among digital content creators in China. Data were collected through a survey conducted between [data collection period] among registered creators on major Chinese content-creation platforms, covering formats such as text and images, short videos, and podcasts. The final sample consists of 8,342 respondents. Empirically, Logit models are employed to analyze copyright registration behavior, while OLS models are used to estimate the intensity of copyright enforcement actions. Mediation mechanisms are examined using bootstrap procedures, and 2SLS estimation is used to address potential endogeneity concerns. The results show that greater GenAI usage intensity is significantly associated with both a higher likelihood of copyright registration and greater enforcement intensity. Perceived infringement risk partially mediates the relationship between GenAI use and copyright registration, whereas AI literacy partially mediates its relationship with enforcement intensity. In addition, stricter platform copyright policies strengthen these effects. The effects are particularly pronounced among full-time professional creators and creators producing new-media content. This study provides empirical evidence on the relationship between GenAI adoption and copyright protection behavior and contributes to the literature on platform governance and the development of policies for creator protection.