To investigate the effectiveness of generative AI-supported error-corrective case-based teaching in clinical nephrology clerkship.

Medical students rotating through the Department of Nephrology at the Army Specialized Medical Center from June 2024 to September 2025 were selected and randomly assigned to an experimental group and a control group using a random number table. The control group received traditional case-based teaching, while the experimental group underwent generative AI-supported error-corrective case-based teaching. Both groups received a 2-week clinical training period. The pre- and post-teaching examination scores, learning engagement and critical thinking level were compared between the two groups, and their comprehensive learning effect and teaching satisfaction were evaluated after teaching.

A total of 64 medical students were enrolled, with 32 in each group. Post-teaching assessment results showed that the case-analysis score [(42.50±4.56) vs (38.47±4.75)] and total examination score [(82.06±8.09) vs (75.97±7.37)] in the experimental group were higher than those in the control group (all P<0.05). The scores of dedication dimension [(30.34±5.95) vs (27.63±4.84)], absorption dimension [(34.53±7.31) vs (25.38±5.66)] and total learning-engagement score [(92.19±9.09) vs (79.19±8.10)] were higher in the experimental group (all P<0.05). The experimental group also obtained higher scores in learning effect [(67.41±4.31) vs (60.25±6.61)], learning experience [(38.09±3.21) vs (34.19±5.37)], classroom interaction [(18.41±1.88) vs (16.50±2.11)], and academic load [(10.91±1.82) vs (9.56±2.69)] (all P<0.05). The total critical-thinking score was higher in the experimental group [(266.72±16.57) vs (252.06±13.86)], together with higher scores in the sub-dimensions of truth-seeking, open-mindedness, systematicity, critical-thinking self-confidence, and inquisitiveness (all P<0.05). The total teaching-satisfaction score of the experimental group was higher than that of the control group [(83.38±3.08) vs (79.16±4.15), P<0.05].

Generative AI-supported error-corrective case-based teaching can improve medical students′ case-analysis ability, learning engagement, learning effect, and critical thinking, and enhance teaching satisfaction.