The Use and Perceptions of AI Chatbots in Medical Research: An International Cross-Sectional Survey

Abstract

Background Artificial Intelligence (AI) has undergone remarkable progress, leading to the development of advanced large language models (LLMs). Despite these LLMs' growing adoption, concerns persist regarding the scientific accuracy of AI-generated content, and their acceptance within academic publishing remains contentious. This study aimed to describe AI-chatbot use patterns and to assess medical researchers' perceptions of impact on research credibility, ethical concerns, guideline awareness, and disclosure of future intentions. Methodology A cross-sectional survey-based study spread into Saudi Arabia, Nigeria, Tunis, and England from 2023 to 2024 surveyed researchers, excluding non-medical and non-publishing researchers. Results We analyzed 434 respondents; 175 (40.3%) reported AI-chatbot use. Use varied by country (32.8%-45.9%), but neither gender nor country was significantly associated with use. Older age and more senior roles were associated with lower odds of use (odds ratio (OR): ages 41-50 years, 0.32; residents, 0.31; consultants, 0.17; <i>P</i> ≤ 0.009). Awareness strongly predicted use (OR 15.53), as did guideline awareness (OR 2.47), trust (<i>P</i> = 0.005), hypothesis formation (<i>P</i> = 0.001), willingness to cite (<i>P</i> = 0.003), and future use (<i>P</i> < 0.001); intention to declare use during submission did not differ (<i>P</i> = 0.468). Conclusions Our study shows that medical researchers have a positive attitude toward using AI chatbots, but with ethical and accuracy concerns requiring further interventions to create systematic unified rules.

Ähnliche Arbeiten