Abstract
This project develops and evaluates an inquiry-based AI Teaching Assistant (AI TA) designed to foster deeper cognitive engagement in management education. As generative AI becomes increasingly embedded in students’ learning practices, it often encourages answer-seeking behavior and reduces the need for independent reasoning. This creates a central pedagogical challenge: how to integrate AI into teaching while preserving, and ideally strengthening, students’ critical thinking. Supported by a Teaching Development Grant (TDG), this project addresses this challenge by redefining the role of AI from an answer provider to an inquiry facilitator.
The AI TA was implemented in an undergraduate strategic management course at the University of Hong Kong. The system was trained on course-specific materials, including lecture slides, teaching notes, case materials, and transcripts, allowing it to provide contextually relevant support aligned with the instructor’s pedagogy. Unlike conventional AI tools, the chatbot was designed to guide students through structured questioning rather than provide direct answers. Its primary objective was to prompt students to engage more actively with course concepts, articulate their reasoning, and develop independent problem-solving skills.
The project was implemented in two stages. The initial deployment revealed important limitations: student usage was relatively low, and interactions were largely driven by answer-seeking behavior. These patterns highlighted a key tension in AI-supported learning—while students value efficiency, this often comes at the expense of deeper cognitive engagement. In response, the AI TA was redesigned to adopt a more inquiry-driven interaction style, emphasizing guided questioning, iterative dialogue, and integration into course assignments.
The revised implementation led to a substantial increase in student engagement and a qualitative shift in how students interacted with the tool. Rather than using the AI as a shortcut for answers, students engaged in more sustained exchanges that required them to think through problems, evaluate alternatives, and refine their understanding. Analysis of interaction data and student feedback suggests that inquiry-based AI can effectively encourage active learning when its design aligns with pedagogical objectives and when its use is embedded within course structure.
Overall, this project demonstrates that the educational impact of AI depends less on the technology itself and more on how it is designed and integrated into teaching. By shifting AI from a source of answers to a mechanism for inquiry, the AI TA provides a scalable approach to leveraging generative AI in ways that enhance, rather than undermine, critical thinking in higher education.