A Study On The Washback Effect Of Generative AI Text Detection On English Writing Teaching: A Two-Way Investigation Based On Teacher Cognition And Student Strategies

Abstract

The rapid adoption of generative AI tools in education has revolutionized writing practices while simultaneously introducing concerns over academic authenticity. In response, AI text detection technologies have been increasingly deployed, especially in English writing instruction. This article offers a comparative analysis of the washback effect of such tools on two primary educational actors: teachers and students. Drawing upon literature, pedagogical reports, and synthesized classroom models, this study contrasts how detection technologies shape teacher cognition (beliefs, attitudes, instructional decisions) versus student strategies (adaptive behaviors, AI-use workarounds, and writing habits). The analysis reveals diverging adaptations: teachers often revise assessment methods to emphasize control and monitoring, while students employ avoidance, obfuscation, or ethical use strategies. All conceptual models synthesized from multiple reputable sources. These models are intended to illustrate dominant behavioral and pedagogical trends rather than offer statistically precise findings. These trends suggest that AI detection tools may be reconfiguring the English writing landscape in ways that demand balanced policy and pedagogical reflection.