<![CDATA[Purpose: This study evaluates the empirical impact of artificial intelligence-assisted writing tools on academic writing performance, drafting dynamics, and feedback interactions among undergraduate second-language learners in Indonesian higher education. Method: An explanatory sequential mixed-methods design was conducted during an instructional module at a public university in West Java. Data were gathered through pre- and post-intervention argumentative essays scored via the standardized ESL Composition Profile, screen-recorded writing logs, and semi-structured interviews. Findings: Results demonstrate statistically significant improvements in overall academic writing performance, with the most pronounced gains observed in linguistic accuracy and lexical richness. Qualitatively, participants exhibited recursive prompt refinement, strategically positioning generative technology as an interactive dialogue partner to scaffold composition rather than relying on passive text generation. Practical implications: The findings provide actionable frameworks for language educators to implement prompt-literacy instruction and hybrid feedback workshops, effectively reducing mechanical writing anxiety while encouraging higher-order rhetorical development. Originality/value: This research fills a critical empirical void in second-language writing scholarship by demonstrating how structured prompt engineering fosters metacognitive awareness, enhances learner autonomy, and safeguards authorial agency in digital learning environments.]]>
Copyrights © 2025
