BEYOND REPLACEMENT: HUMAN–AI COMPLEMENTARITY IN EFL WRITING AND SPEAKING

Abstract

The increasing use of artificial intelligence (AI) in English language education has introduced new possibilities for providing feedback on learners’ productive skills. This thesis examines AI-assisted feedback in EFL writing and speaking through the concept of human–AI complementarity. Rather than considering AI as a replacement for teacher feedback, the thesis explores how the different strengths of AI and teachers can be integrated into the feedback process. Recent empirical evidence indicates that AI can provide immediate, accessible, and language-focused support, contributing to improvements in areas such as writing accuracy, learner autonomy, speaking fluency, and syntactic complexity. At the same time, teacher feedback remains important for contextual interpretation, deeper-level language development, pedagogical decision-making, and interpersonal support. Evidence from teacher-mediated and hybrid feedback models further suggests that combining AI capabilities with teacher expertise can provide broader support for learners’ productive skills. The thesis argues that the pedagogical value of AI-assisted feedback lies not in replacing human feedback but in extending and complementing it. A human–AI feedback model may therefore offer a more flexible and comprehensive approach to the development of EFL writing and speaking skills.