Bridging Efficiency and Emotion: Evaluating User Engagement with AI-Generated and Human-Generated TikTok Videos Through Media Richness and Uses and Gratifications Theories

Abstract

This study examines the impact of AI-generated and human-generated TikTok videos on user engagement, measured through likes, comments, and shares, by applying the Media Richness Theory (MRT) and the Uses and Gratifications Theory (UGT). The data consists of 400 TikTok videos, and 255,541 comments from these videos are used for analysis. The research consists of 02 studies, in which study 1 uses regression analysis, study 2 uses sentiment analysis and topic modeling, to identify key factors affecting engagement, including image, audio, and content quality. The results show that AI-generated content is perceived as effective and informative but lacks emotional depth, resulting in lower likes, comments and shares, and lack of richness in comment words. Meanwhile, human-generated content drives stronger emotional connections and authenticity, leading to higher engagement across all metrics. The study contributes to MRT by demonstrating that media richness is influenced not only by technical quality but also by perceived authenticity, and refines UGT by highlighting the distinct satisfactions found in AI- and human-generated content. The practical implications suggest that content creators should combine AI effectiveness with human emotional appeal to optimize engagement. Future research should explore hybrid content strategies to enhance user experience and platform effectiveness