The Application of Artificial Intelligence Automated Evaluation Systems in English Writing Instruction: Scoring Accuracy, Feedback Effectiveness, and User Experience

With the deepening application of artificial intelligence (AI) in education, automated evaluation systems have become important aids in English writing instruction. However, disputes persist regarding their scoring accuracy, feedback effectiveness, and user experience. This paper reviews the application of AI automated evaluation systems in English writing instruction, focusing on scoring accuracy, feedback effectiveness, and user experience. Findings show that AI systems perform well in evaluating linguistic forms, demonstrating high consistency with human raters, especially in grammar and vocabulary. Nevertheless, they remain inadequate in assessing content depth, logical structure, and creativity. Regarding feedback, AI systems can enhance students’ writing proficiency if combined with scaffolded designs (e.g., model essays, checklists, multiple rounds of revision). In terms of user experience, students value the immediacy and interactivity of AI feedback but trust it less than teacher feedback on content-related aspects. The conclusion points out that AI systems should be positioned as playing an auxiliary role in formative evaluation, leverage their advantages in efficiency and personalization through human-machine collaboration, and pay attention to teacher training, system optimization, and ethical norms, so as to promote English writing instruction toward a more efficient, inclusive, and human-centered direction.