Algorithm Aversion to Generative Artificial Intelligence in Creative Content

Despite the widespread penetration of generative artificial intelligence (AI) in creative content production, disdain, aversion, and resistance among users have become increasingly apparent, challenging the acceptance and trustworthiness of AI-generated content. This study employs a mixed-methods approach to explore these issues comprehensively. It begins by collecting user submission data from an anti-AI generation platform to examine focus areas, issues, and emotions, using BERTopic modeling, sentiment analysis, and temporal evolution analysis. Furthermore, in-depth interviews with 22 users who strongly oppose AI-generated content provide deeper insights. The findings reveal that user aversion is driven by multiple factors and extends beyond the technology itself to encompass AI-created content, associated products, and the entities utilizing this technology. These cognitions and emotions further translate into specific resistance behaviors. This study offers crucial insights into user behavior and attitudes towards algorithms, enhancing understanding of societal acceptance and guiding the development of ethically aligned, user-friendly AI applications.