Artificial Intelligence Myths: Prevalence Among Turkish University Students and Comparative Analysis of ChatGPT Responses

Abstract

Artificial intelligence (AI) has rapidly emerged as a transformative technology across various disciplines. However, its widespread adoption is accompanied by numerous myths, which are fueled by limited public understanding and can significantly shape how individuals perceive and interact with AI, often leading to negative consequences such as misunderstanding, fear, or resistance. Despite the importance of addressing these myths, research on the prevalence of such beliefs remains insufficient, particularly in the Turkish context. This study aims to determine the prevalence of AI myths among Turkish university students, investigate the factors influencing the adoption of these myths, and compare student perceptions with ChatGPT's responses to the same myths. The study analyzed survey data from 288 students (102 males, 35.4%, and 186 females, 64.6%) using an AI-myth survey consisting of 18 items. Both descriptive and inferential analyses were conducted to determine the prevalence of AI myths and investigate how factors such as academic background, gender, AI-related training, and media consumption influence the adoption of these myths. A comparative analysis was also performed between student responses and ChatGPT’s reactions to these myths. Analysis showed that certain AI myths are particularly prevalent among students. Students from technical disciplines demonstrated a greater ability to identify these myths, while prior AI training and media consumption had minimal impact. ChatGPT’s responses highlighted areas where better communication about AI is needed. The findings suggest that improving AI literacy and dispelling myths are essential for preparing students for more informed engagement with AI technologies.

Ähnliche Arbeiten