AI-enhanced collaborative story writing in the EFL classroom

Artificial intelligence (AI) is opening new possibilities for educators to explore innovative solutions to classroom challenges. While the benefits of AI in enhancing individual learning are well established, its potential in collaborative learning environments remains underexplored. This paper reports on the first cycle of a design-based research project that involved the development of Collabowrite, a collaborative story-writing application designed for compulsory English-as-a-foreign-language (EFL) classes at a Japanese university. During a 14-week semester, 52 students engaged with the application in 160 separate usage instances, using AI features such as grammar checks, virtual group members, artwork generation, and story narration. The grammar checks were evaluated for accuracy at 90.4% across 892 instances, with students adhering to elements of AI-provided guidance 91.4% of the time. Students rated their enjoyment of the activities at an average of 4.63 out of 5. Pre- and post-intervention surveys involving 27 participants and using single-item measures indicated significant positive shifts in perceptions toward English writing, collaborative work, and the role of technology in language learning. Reflexive thematic analysis of qualitative data generated three salient themes that capture the diverse experiences of participants. These findings suggest the need to revisit sociocultural learning theories to incorporate AI into classroom dynamics. This initial research cycle led to the formulation of six design principles for integrating AI-enhanced collaborative writing tools into EFL classrooms. Future cycles will further evaluate the efficacy of Collabowrite as an educational tool and its impact on students’ foreign language writing abilities.