<p>Despite growing interest in AI-enhanced instruction, the specific contribution of AI to the development of critical thinking within flipped learning environments remains underexplored. This mixed-methods study investigates the impact of an integrated AI-assisted flipped learning model on university students’ critical thinking skills, dispositions, and motivation. Over 10 weeks, 70 students were assigned to either an experimental group using the Socrat.ai platform within a flipped-classroom framework or a control group receiving traditional flipped learning without AI support. Quantitative data were collected using validated scales for critical thinking skills, dispositions, and motivation; pre-test equivalence checks confirmed comparability between groups, and post-test comparisons revealed greater improvements in the experimental group across all three dimensions. Qualitative data from semi-structured interviews with nine experimental-group students yielded four themes: AI-supported questioning as a catalyst for critical thinking, enhanced motivation through personalized feedback, improved preparedness for in-class discussion, and perceived challenges of AI use. These qualitative insights contextualize the quantitative gains by identifying Socrat.ai’s adaptive Socratic questioning, longitudinal memory, and criterion-referenced feedback as plausible mechanisms of change. Because both groups followed the same flipped classroom structure, the observed gains are tentatively attributed to the AI-assisted component. The findings offer one of the first controlled comparisons that isolate AI’s incremental contribution within an active pedagogical framework, with important implications for educators seeking to amplify critical-thinking outcomes through AI-assisted flipped learning.</p>