Is it time to introduce an artificial intelligence curriculum in undergraduate medical education? medical students’ perspectives: a systematic review and meta-analysis

BACKGROUND: Although artificial intelligence (AI) is increasingly transforming healthcare systems, its systematic integration into undergraduate medical education (UME) remains limited. Despite widespread recognition of AI's potential to enhance clinical practice, AI-related competencies are still inadequately embedded in most medical curricula, even though medical students generally express positive attitudes toward AI. OBJECTIVES: This systematic review and meta-analysis synthesizes the current literature on medical students' attitudes toward teaching about AI and clarifies the challenges, opportunities, and potential approaches for incorporating AI into the UME curriculum. These findings underscore the need for competency-based AI integration in UME worldwide. METHODS: Following Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines, a systematic literature review was conducted using the PubMed, Web of Science, Scopus, and Cochrane Library databases between January 2000 and February 2025, revealing 26 studies (n = 20,963 students) assessing attitudes and challenges related to learning AI. Meta-analysis calculated the pooled proportions of student support. A random-effects model with Freeman-Tukey transformation was used; heterogeneity and bias were assessed via standard statistical methods. RESULTS: The systematic review included 26 studies (n = 20,963 medical students). Meta-analysis revealed that 16,278 participants (78.0%) held positive attitudes towards AI curriculum integration. Despite this strong support, a significant 'optimism-competence gap' was identified: while 17,420 participants (83.1%) agreed on the necessity for AI training, only 7,630 participants (36.4%) felt confident in applying AI in clinical practice. Subgroup analyses confirmed consistent support across geographic regions. The analysis indicated extreme heterogeneity (I² = 98.5%). The primary implementation challenges, derived from the systematic review, were curricular overcrowding (reported in 13 out of 19 studies, 68%), lack of faculty expertise (in 12 out of 23 studies, 52%), and ethical concerns (in 9 out of 22 studies, 41%). CONCLUSION: A standardized AI curriculum is urgently needed to bridge the optimism-competence gap. Prioritizing faculty training, interdisciplinary collaboration, and ethical frameworks will ensure that future physicians can harness AI's potential while mitigating risks. Our findings, particularly the quantified optimism-competence gap and the influence of unmeasured institutional factors, provide a new evidence base for developing these prioritized interventions.