Exploring the Impact of AI on Academic Performance of University Students

Abstract

The utilization of artificial intelligence (AI) in schools has drawn a lot of awareness due to its possible contribution to students' performance. The current study investigates the association between the utilization of AI tools and university students' academic performance. With an application of a database of 2,000 student responses, the research explores how various determinants contribute to how one performs academically, such as knowledge about AI, use, and self-estimated impact of AI tools on productivity, psychological well-being, and memorability of material. Sentiment analysis measures the extent to which each characteristic contributes. Machine learning models such as Random Forest and Decision Tree regressors are employed to predict academic performance based on selected features. The models deliver good performance, where Random Forest delivers an R-squared of 0.8839 and a mean squared error (MSE) of 0.1220, while that of Decision Tree is an R-squared of 0.8283 and an MSE of 0.1797. These results suggest that experience and use of AI are strong predictors of productivity and retention and that AI tools also positively affect mental well-being and focus. The study highlights the necessity of ethical AI use in schools and identifies ethical concerns regarding AI implementation in school environments. The findings add to the body of knowledge of effective application of AI to enhance student performance and guide AI-powered education for future studies.