Self-reported artificial intelligence use, knowledge, and perceptions among undergraduate medical students at a Peruvian university

Artificial intelligence (AI) is increasingly used in medical education, yet evidence regarding its actual use, knowledge, and perceptions among undergraduate medical students in Peru remains limited. This study aimed to estimate the prevalence of self-reported AI use for academic activities and to describe patterns of use, self-reported knowledge, and perceptions among undergraduate medical students at a Peruvian university, with an exploratory assessment of associated characteristics. We conducted an observational, analytical, cross-sectional study in 2024 among students enrolled at Universidad Ricardo Palma. A total of 279 responses were received; one respondent did not provide informed consent and was excluded, leaving 278 participants for analysis. Data were collected using an adapted online questionnaire based on previously published instruments. AI use was assessed through a separate yes/no question regarding previous use of any AI tool. Overall, 257 of 278 participants (92.4%) reported previous AI use for academic activities. Most participants self-reported basic knowledge of AI principles (89.2%) and awareness of its use in medicine (95.3%). Among the predefined AI-tool categories, ChatGPT (69.1%) and Canva (19.8%) were the most frequently reported. The primary contribution of this study is descriptive, documenting the prevalence and patterns of AI use, self-reported knowledge, and perceptions in the surveyed population. Secondary exploratory regression analyses suggested associations with daily digital-device use and perceptions regarding treatment planning, but these estimates should be interpreted cautiously because only 21 participants were classified as non-users. These findings highlight the need for structured education addressing critical, ethical, and responsible AI use in undergraduate medical training.