Evaluating the Effectiveness of AI-Powered Conversational Models in Construction Management Education: A Comparative Investigation of Student Performances and ChatGPT

Abstract

This study examines the impact of one of the AI conversational tools, ChatGPT-4, on learning outcomes in construction management courses. Focusing on the Site Management course, the same exam questions were administered to both students and ChatGPT-4. This evaluation was conducted in two phases—midterm and final exams—where the performances of both ChatGPT-4 and the students were compared and analyzed. The research reveals that while the AI outperformed students in the midterm exam, it showed a decline in performance during the final exam, indicating limitations in adapting to different exam environments. Neither ChatGPT-4 nor the students met the course passing criteria. The study emphasizes the need for further development of AI for more effective integration into education, with a focus on considering ethical dimensions in this process.