Deciphering AI-chatbots utilization in clinical practice, research and education among Chinese medical students in generative AI era: a cross-sectional study

Artificial intelligence chatbots (AI-chatbots) are increasingly being applied in the medical field; however, their specific utilization in Chinese medical education and practice remains poorly understood. This study aims to systematically analyze the perception and utilization of AI-chatbots among Chinese medical students, explore potential factors influencing their effectiveness, and provide recommendations for policy development and technological optimization. This cross-sectional study utilized an electronic questionnaire distributed through multiple channels to collect data from Chinese medical students, with most participants recruited from Guangdong Province, Southern China. The questionnaire assessed demographic characteristics, AI-chatbot usage patterns, attitudes towards their application in clinical, research, and educational domains, and perspectives on potential issues. Data were analyzed using descriptive statistics, independent sample t-tests, chi-square tests, Wilcoxon rank-sum tests, Logistic regressions and Pearson correlation analyses. Among 414 respondents, 53.6% had used AI-chatbots, primarily ChatGPT (58%). AI-chatbots were favored for research (36.5%) and education (40.1%), with limited use in clinical practice (4.5%). Users' evaluations of AI-chatbots in research and education were significantly more positive than those of non-users. Use frequency, time, and attention to updates demonstrated significant positive correlations with positive responses (P < 0.05). Information accuracy emerged as a top concern, with over 65% of respondents indicating the urgency for a solution. Our findings indicate that AI-chatbots are widely utilized particularly in research and educational-related domains among Chinese medical students, with our sample most recruited from Southern China. However, the implementation of AI-chatbots in clinical practice continues to face significant challenges. Medical students exhibit a cautiously optimistic attitude towards AI-chatbots, acknowledging their potential while remaining cognizant of associated risks. Future research should prioritize investigating the safety and efficacy of AI-chatbots in clinical practice, as well as exploring optimal strategies for their integration into the medical education curriculum.