The integration of Artificial Intelligence (AI) in education has revolutionized both learning and teaching processes, with Virtual Assistants (VAs) emerging as one of its most impactful applications. This systematic review synthesizes empirical studies published between 2014 and 2025 that examine the role of AI- powered VAs in primary, secondary, and tertiary education. Four research questions are addressed: (1) the educational uses and implications of VAs, (2) the most frequently applied models and algorithms, (3) their contribution to learner motivation and self-determination, and (4) the benefits and challenges of instructional design supported by VAs. A total of sixty-two studies were analyzed following PRISMA guidelines. The findings indicate that VAs are most often implemented in activity-based, emotion-aware, and classroom management contexts, primarily through algorithms such as natural language processing models, convolutional neural networks, and decision trees. While VAs enhance personalization, interaction, and inclusiveness, persistent concerns remain regarding cultural diversity in emotion recognition, data privacy, and pedagogical alignment. One major conclusion of this study is that a persistent gap remains between technological innovation and educational theory, highlighting the need for Virtual Assistants to be pedagogically embedded within authentic classroom environments.
Papachristou et al. (Thu,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: