The persuasive effects of AI-generated digital human avatar videos in public legal education

This study examines the persuasive effects of artificial intelligence (AI)-generated digital human avatar (DHA) videos on public legal education and explores the moderating roles of technology transparency and demographic characteristics in this context. In collaboration with the largest chain law firm in mainland China, we invited 808 clients who were visiting the firm for a consultation to participate in an experiment. The results revealed that compared with reading textual materials or receiving no materials, the use of DHA videos, regardless of whether their AI-generated nature was disclosed, significantly increased participants’ perception that the use of pirated software is unethical and illegal. Moreover, the persuasive effect of the DHA videos was more pronounced among older individuals and among those with lower education levels, whereas other demographic factors, such as gender, income, and city of residence, had no significant effects. These findings were replicated in a supplementary study (n = 548) that featured a pretest–posttest design (see Web Appendix B). These findings contribute to the growing body of literature on AI-driven legal education and offer practical insights into the use of digital human technology for public legal advocacy.