AI-based learning systems are transforming education in academic and corporate settings, with global spending on AI-enabled training expected to exceed $200 billion by 2025. This study examines how these systems enhance learning outcomes for college students and employees, focusing on system functionality, self-efficacy, familiarity, and social influence. Using structural equation modeling with data from 598 participants (258 students, 340 employees), findings reveal that familiarity positively impacts self-efficacy and learning outcomes across both groups, while social influence varies. Students benefit from media-rich environments, while employees gain from job-relevant content and supervisor influence. System functionality enhances self-efficacy and participation motivation, but self-efficacy's direct effect on learning outcomes is significant only for employees. These results highlight the need to tailor AI-based systems to user profiles. Limitations include reliance on self-reported data and specific learning contexts. Future research should incorporate objective measures of learning outcomes and explore additional factors like teaching strategies.
Read full abstract