Speaker
Description
As AI tools are becoming part of everyday language learning, teachers need to understand how they can be used not only for faster feedback but also for more caring and meaningful assessment. This study explores the use of human–AI feedback in a Writing 2 course for first-year English majors at a Vietnamese university. The students begin the course at around B1 level and are expected to move toward B2 writing outcomes by the end of the semester. The study is designed as a mixed-methods classroom case study with one intact class of 36 students. During the semester, students complete several writing tasks. For each task, they write a first draft, receive AI-generated feedback, reflect on that feedback, revise their writing, and receive further guidance from the teacher. Their writing progress is tracked through an analytic rubric covering task achievement, coherence and cohesion, lexical resource, and grammatical accuracy. The study asks three main questions: how the teacher demonstrates AI language assessment literacy when designing and guiding AI-supported continuous assessment; how human–AI feedback helps track students’ B1-to-B2 writing progress; and how students perceive AI-supported feedback in relation to their progress, confidence, and wellbeing. Data are collected from students’ diagnostic and final writing tasks, drafts and revisions, AI feedback records, teacher notes, student reflections, questionnaires, and selected interviews. By focusing on both progress and care, the study aims to show how AI can support continuous assessment without replacing the teacher’s role. It contributes to CALL research by offering a classroom-based example of responsible, formative, and human-centred AI use in EFL writing.
