Adoption, usage patterns, and perceptions of DeepSeek among medical interns: Evidence from a Chinese Teaching Hospital

Artificial intelligence (AI)-driven chatbots based on large language models are increasingly entering medical education, but evidence on the real-world use of domestic systems such as DeepSeek among Chinese medical interns is still scarce. This cross-sectional study investigated the prevalence, patterns and perceptions of DeepSeek use among medical interns at the Second Affiliated Hospital of Anhui Medical University. A total of 634 interns from clinical medicine, preventive medicine and nursing completed an anonymous questionnaire. Overall, about one-third of interns reported using DeepSeek. Among users, most engaged with the tool regularly, with over half using it several times per week. DeepSeek was primarily employed for completing assignments, searching for information, translation and polishing of written texts, while a smaller subset of users reported using it for data analysis. Attitudes toward DeepSeek were mainly neutral or optimistic, and across majors most users perceived its advantages and disadvantages as broadly balanced, while only a minority considered the disadvantages to outweigh the advantages. After adjustment, residence and major remained associated with DeepSeek use, whereas age and sex were not. Urban interns were more likely to report using DeepSeek than rural interns. Compared with clinical medicine interns, preventive medicine interns were less likely to report use, whereas nursing interns were more likely to do so. Among users, overall attitudes differed across majors, while usage frequency and the perceived balance of advantages and disadvantages did not. These findings provide descriptive evidence on the use and evaluation of DeepSeek among medical interns and may inform future institutional guidance on responsible use, verification of AI-generated information and academic integrity.