Self-Reported Knowledge, Attitudes, Perceptions, and Readiness Regarding AI Among Obstetrics and Gynecology Trainees: Cross-Sectional Study

Abstract Background AI technologies refer to computer-based systems designed to perform tasks that typically require human intelligence and have been increasingly used in obstetrics and gynecology (O&G). Objective This study aimed to assess O&G trainees’ self-reported knowledge of AI, attitude toward its introduction into clinical practice, perception of its clinical importance, and their readiness for its introduction. Methods A cross-sectional study was conducted from December 1, 2024, to December 31, 2024, among O&G trainees in Jordan. Data were collected on participants’ characteristics, self-reported knowledge of AI in O&G, their attitudes toward its introduction, and perception of its importance. The scores of the 3 domains and the study-specific knowledge, attitude, and perception (KAP)–based readiness were converted into percentages of their maximum attainable scores and were grouped into low, moderate, and high categories. Multivariable linear regression analysis was used to identify variables associated with KAP-based readiness. Results A total of 218 trainees were recruited; the median age was 28 (IQR 24-38) years, 180 (82%) participants were female, 117 (53.7%) were junior trainees, 148 (67.9%) were working in public hospitals, 183 (83.9%) reported “average or better” knowledge of IT, and 196 (89.9%) had never received formal training on the medical applications of AI. The highest median percentage score was for self-reported knowledge (71.1%, IQR 62.2%‐75.6%). Additionally, KAP-based readiness was moderate in 188 (86.2%) participants. In multivariable linear regression, none of the examined trainee characteristics were independently associated with the KAP-based readiness (all P values >.05). Conclusions O&G trainees in Jordan demonstrated moderate self-reported knowledge of AI, generally positive attitudes and perceptions of its importance, and moderate study-specific KAP-based readiness. Moreover, formal training on AI medical applications was uncommon, and most trainees supported the integration of AI training into medical education. These findings support the need for structured AI education, and future research should evaluate broader and objectively measured individual and organizational determinants of AI-related readiness.