Interaction and dialogue: Integration and application of artificial intelligence in blended mode writing feedback

With the rapid advancement of Artificial Intelligence (AI), particularly conversational tools like Chat Generative Pre-trained Transformer (ChatGPT), their application in writing feedback has sparked widespread discussion. While AI writing feedback has proven effective, most research focuses on single feedback methods, with a limited exploration of integrating AI with other approaches. This study uses the Interactive, Constructive, Active, Passive (ICAP) framework to compare hybrid and single feedback in Chinese writing among 86 university students studying in Tibet. The study was conducted in two phases: Phase 1 involved a ten-week quasi-experiment, where the control group received human teacher feedback, and experimental groups received feedback from Kimi.AI (Passive or Active) or a combination of both. In Phase 2, data were collected through semistructured interviews and analyzed using AntConc software. Results showed that combined teacher and AI feedback improved writing skills significantly, highlighting the importance of Rich Interaction and AI's potential to promote educational equity.