Examining the Acceptance and Use of AI-Based Assistive Technology Among University Students with Visual Disability: The Moderating Role of Physical Self-Esteem.

AI-based assistive technologies (AIATs) are increasingly recognised as essential tools to enhance accessibility, independence, and inclusion for visually impaired students in higher education. However, limited evidence exists regarding the determinants of their acceptance and use, particularly in terms of psychosocial factors. This study aimed to extend the Unified Theory of Acceptance and Use of Technology (UTAUT) by incorporating physical self-esteem (PSE) as a moderator and behavioural intention (BI) as a mediator within a single model. Data were collected through a validated questionnaire administered to 395 visually impaired undergraduates across five Saudi universities. Constructs included effort expectancy (EE), performance expectancy (PE), facilitating conditions (FCs), social influence (SI), BI, and PSE. Partial Least Squares Structural Equation Modelling (PLS-SEM) was used for analysis. Results showed that PE and SI significantly predicted both BI and adoption, while EE strongly predicted BI but not AIAT adoption; FC had no significant influence on either outcome. BI positively affected AIAT adoption and mediated the effects of PE, EE, and SI, but not FC. Moderation analysis indicated that PSE strengthened the influence of PE, EE, and SI on BI and adoption. These findings underscore the significance of psychological factors, particularly self-esteem, in promoting the adoption of AIAT and offer guidance for developing inclusive educational strategies.