Teacher practices and challenges in AI-integrated EFL instruction: A case study from an underdeveloped region
TL;DR
Findings reveal that the teacher's engagement is non-linear and front-loaded: while Behavioral Engagement peaked during the Design and Development phases through active material generation, it significantly declined during the Implementation phase due to infrastructural anxiety and instability.
Abstract
Despite the rapid proliferation of artificial intelligence (AI) in education, its implementation in underdeveloped regions remains under-explored, creating a significant knowledge gap regarding digital equity. This qualitative case study investigates how an EFL teacher in Shigatse, Tibet, engages with AI throughout the instructional process. Adopting a unified analytical framework that integrates the multidimensional construct of engagement (behavioral, emotional, cognitive) with the ADDIE instructional design model (Analysis, Design, Development, Implementation, Evaluation), this study offers a process-oriented examination of teacher practices under resource constraints. Findings reveal that the teacher's engagement is non-linear and front-loaded: while Behavioral Engagement peaked during the Design and Development phases through active material generation, it significantly declined during the Implementation phase due to infrastructural anxiety and instability. Furthermore, the teacher's Emotional Engagement evolved from blind optimism to rational pragmatism, shaped by the paradoxical burden of AI-induced supervision. These results challenge the assumption of seamless AI integration, highlighting that in underdeveloped contexts, teachers must navigate a dichotomy between high preparation hopes and low implementation realities.