Year: 2026
Type: article
Abstract: This paper studies practical paths for applying artificial intelligence in assessment in English language teaching. AI-based assessment is defined here as the supervised application of automated scoring, speech or text analysis, adaptive task delivery, and generative feedback to support evidence collection and pedagogical decisions. AI will be used to build the assessment system; it will not function independently as an examiner. It is a staged path that begins with the construction of definitions and low-stakes formative uses, continues to teacher moderation and learner dialogue, and develops into institutionally regulated implementation with validation, privacy protection, fairness review, and appeal mechanisms. Examples of writing, speaking, reading, listening and classroom feedback are provided. The paper is theoretical and practical, and there are no original learners, datasets, experiments, scores or statistical results. Therefore, it is not expected that AI will improve accuracy, achievement, efficiency and fairness. At its core, the educational goals are to learn by example, to think independently, and to be accountable.
Source: Academic Journal of Education
DOI: 10.54097/4nppc096
Author: Zhiwei Qiu
Institution: University of Leeds
Language: English
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
SDG: Quality education
Open Access status: hybrid
OpenAlex ID: W7225381250