PUBLIC PERCEPTIONS OF AI HALLUCINATIONS: A CASE STUDY OF REDDIT DISCUSSIONS

This paper explores public perceptions of AI hallucinations by analyzing discussions on Reddit. As large language models (LLMs) become increasingly integrated into everyday life, concerns about the accuracy and reliability of AI-generated content have emerged. Using a text mining approach, we examine reddit discussions to identify major themes regarding AI hallucinations. The findings highlight a growing awareness of AI's limitations and the need to mitigate the impact of AI hallucinations. Additionally, the analysis of Reddit discussions led to the development of a set of research questions that could be valuable for researchers studying the phenomenon of AI hallucinations.