The rapid advancement of generative Artificial Intelligence has opened new possibilities for improving speaking and debate instruction in higher education. This study aimed to investigate the implementation of “ChatGPT-based Debate Coursebook” in a second-semester Debate and Critical Thinking class, focusing on its strengths, challenges, and proposed improvements from both students and classroom observers. This research employed a qualitative case study design, involving 7 student participants and 2 observers from the English Education Department at a State Islamic University in Indonesia. Data were collected through semi-structured interviews, classroom observations, document analysis, and field notes. In addition, data analysis involved thematic coding, frequency analysis, and triangulation of data sources to ensure trustworthiness. The findings indicated that the coursebook provided clear structure, strengthened argumentation skills, and increased engagement, particularly through the integration of ChatGPT for independent practice and feedback. Nevertheless, several challenges emerged, including limited class time, initial difficulty understanding the Australasian Parliamentary Debate format, uneven or excessive reliance on ChatGPT, and limited live feedback. Students and observers recommended the addition of multimedia support, clearer scaffolding for ethical AI use, structured practice schedules, improved pronunciation and fluency activities, and more explicit assessment rubrics. Based on these insights, the study recommends developing AI-assisted materials that are multimodal, ethically guided, and supported by stronger lecturer scaffolding to balance AI use and foster deeper critical thinking.
Copyrights © 2026