The Impact of AI-Assisted Learning Tools on Students’ Academic Writing Performance

Abstract

This study explores how students' perceptions of AI-supported learning tools are related to their academic writing outcomes, and how responsible and ethical use of AI can directly predict these outcomes. The research design used was quantitative correlational research design. A sample of 204 undergraduate and postgraduate students from public and private universities in Pakistan was selected for collecting data using a structured questionnaire which was rated on a five-point Likert scale. The instrument assessed three constructs: AI-Assisted Writing Perceptions (AIWPS), Academic Writing Performance (AWPS), and Ethical and Responsible AI Writing Use (ERAIWUS). Pearson correlation and multiple regression of statistical analysis were performed. The findings showed that students' perceptions of the AI-assisted writing tools were positively correlated with the academic writing performance (r = .800, p < .001). There was also a strong positive correlation between ethical and responsible use of AI and academic writing performance (r = .503, p < .001). Multiple regression analysis was performed to determine if both variables explained a significant portion of the variance in academic writing performance. Both variables were found to be significant and together accounted for 64.9% of the academic writing performance variance (R² = .649, F(2, 201) = 185.82, p < .001). The study findings suggest that AI-powered learning tools can be successfully used to support students' academic writing abilities in an ethical and responsible way. The results underscore the need to foster responsible use of AI in the educational context to enhance the potential value of AI technologies and preserve academic integrity.