A study on college writing teaching methods using ChatGPT : Focusing on topic selection

This study aims to create a college writing class that critically selects topics using ChatGPT. In the academic writing process, topic selection generally comprises the following steps: topic search, research scope limitation, reference search, and topic selection. This study proposes using ChatGPT to access the process of exploring and specifying topics easily. The existing class related to topic selection comprised “topic selection - data collection - group practice,” however, it was revised to “pro visional topic ① - data search - provisional topic ② - data search - group practice.” It aimed to strengthen self-directed learning. To examine student perceptions, they were administered questionnaires and reports. The survey results revealed that most students responded negatively regarding the usage of ChatGPT. However, 30% of the students who used ChatGPT demon strated positive efficacy. Considering that the use of ChatGPT leads to a sense of efficacy, support from schools is required so that most students can use it. As AI-based education has become significant, student learning contexts related to departments, schools, and learning achievements must be discussed in detail. These considerations make individualized learning possible. AI-based education can be used as an auxiliary tool in classes because it is cus tomized for students. Moreover, this is positive, because it alleviates learners’ edu cational alienation. However, based on the results of this survey, it is questionable whether students’ educational alienation will be completely resolved with artificial intelligence technology during the process of realizing customized education. Rath er, the fundamental premise of using artificial intelligence technology is self-direc tion to complete discussions critically. Thus, AI-based education should proceed in a direction that presupposes fundamental literacy and self-direction rather than the process of applying technology. To enable individual learning through AI-based education, students’ detailed learning contexts must be considered.