Speaker
Description
This systematic review examines the impact of generative artificial intelligence (GenAI), particularly tools like ChatGPT, on English Language Learners’ (ELLs) writing quality and writing efficiency. Drawing from 24 peer-reviewed empirical studies published between 2022 and 2025, the review synthesises findings in two areas: (1) writing efficiency, which is measured by planning, drafting, editing, and revising processes; and (2) writing quality, which is defined in terms of lexical sophistication, syntactic complexity, and cohesion. The systematic review's findings showed a consistent improvement in surface-level writing characteristics, such as vocabulary usage, grammar correctness, and organizational coherence. Additionally, learners who use AI-assisted writing platforms report feeling more confident and motivated. Besides writing quality, GenAI improves writing efficiency by expediting drafting, simplifying idea generation, and offering real-time feedback during revision. Reduced student voice, an excessive dependence on AI-generated recommendations, and decreased syntactic complexity are still issues, though. Although GenAI has shown promise in the short term, especially when it comes to scaffolding lower-proficiency learners, its long-term effects on the development of independent writing are still unclear. The review offers pedagogical recommendations for the balanced integration of GenAI in writing instruction.
Keywords: generative artificial intelligence, writing quality, writing efficiency, lexical sophistication, syntactic complexity, text cohesion, AI writing support, second language writing
