Measuring the Impact of Large Language Models on Academic Success and Quality of Life Among Students with Visual Disability: An Assistive Technology Perspective.

In the rapid digital era, artificial intelligence (AI) tools have progressively arisen to shape the education environment. In this context, large language models (LLMs) (i.e., ChatGPT vs. 4.0 and Gemini vs. 2.5) have emerged as powerful applications for academic inclusion. This paper investigated how using and trusting LLMs can impact the academic success and quality of life (QoL) of visually impaired university students. Quantitative research was conducted, obtaining data from 385 visually impaired university students through a structured survey design. Partial Least Squares Structural Equation Modelling (PLS-SEM) was implemented to test the study hypotheses. The findings revealed that trust in LLMs can significantly predict LLM usage, which in turn can improve QoL. While LLM usage failed to directly support the academic success of disabled students, but its impact was mediated through QoL, suggesting that enhancements in well-being can contribute to higher academic success. The results highlighted the importance of promoting trust in AI applications, along with developing an accessible, inclusive, and student-centred digital environment. The study offers practical contributions for educators and policymakers, shedding light on the importance of LLM applications for both the QoL and academic success of visually impaired university students.