Exploring generative AI in the misinformation Era: Impacts as a misinformation source and fact-checker on belief in the information

Abstract

This research aims to explore the impact of generative AI on public belief in misinformation both as a source and as a fact-checker, and to examine how individual factors such as AI literacy and trust in AI influence these perceptions applying machine heuristics. The study conducted two online experiments focusing on health and nutrition issues. The first experiment assessed whether generative AI as a source of misinformation influences public belief differently compared to traditional internet sources. The second experiment evaluated the effectiveness of generative AI in the role of a fact-checker compared to human fact-checkers. The results reveal no significant differences in the belief in misinformation when it was presented by generative AI versus other internet sources. However, human fact-checkers were found to be perceived as more trustworthy than their AI counterparts especially health and nutrition issues, confirming negative machine heuristic. Notably, the level of trust in AI significantly affected participants’ acceptance of AI-generated fact-checking, particularly concerning claims about breastfeeding and child intelligence. This study contributes to the ongoing discourse on the role of AI in misinformation management, highlighting the complex interplay between technological innovations and human cognitive biases.