Factors Influencing the Adoption of Artificial Intelligence Tools and Their Association with Employability Skills among Undergraduate and Postgraduate Student

Abstract: Artificial Intelligence (AI) tools such as ChatGPT, Gemini, Microsoft Copilot, and Claude have rapidly permeated higher education, fundamentally altering the ways in which students learn, conduct research, and prepare for careers. Despite their widespread availability, the extent to which students adopt these tools varies considerably, influenced by factors such as perceived usefulness, ease of use, AI literacy, trust, and ethical awareness. This study investigates the determinants of AI tool adoption among 300 undergraduate and postgraduate students and examines the association between AI tool adoption and perceived employability skills. Employing a quantitative, survey-based research design with a structured five-point Likert scale questionnaire, data were analysed using multiple linear regression, simple linear regression, and independent samples t-test. The multiple regression model explained 34.1% of the variance in AI adoption (R² = 0.341, F(5, 294) = 30.4, p < .001), with AI literacy emerging as the strongest predictor (β = 0.304, p < .001), followed by perceived ease of use (β = 0.269), trust in AI (β = 0.269), perceived usefulness (β = 0.263), and ethical awareness (β = 0.164). The simple regression demonstrated that AI adoption significantly predicted employability skills (R² = 0.349, β = 0.591, p < .001). Furthermore, postgraduate students exhibited significantly higher AI adoption levels than undergraduate students, t(298) = 2.74, p = .007, with a small effect size (Cohen’s d = 0.334). These findings underscore the necessity for educational institutions to integrate AI literacy training and responsible AI usage policies into curricula, thereby equipping students with the digital competencies demanded by AI-driven workplaces. The study contributes to the emerging literature on AI adoption in higher education by integrating multiple adoption determinants within a single empirical framework and linking them to employability outcomes.