FROM ADAPTIVE LEARNING TO VIRTUAL TUTORS: RECENT ADVANCES IN AI FOR TEACHING AND LEARNING

Abstract

Artificial Intelligence (AI) has changed education through the provision of personalization and adaptation through virtual tutors and adaptive learning systems. Prior research focused on the theories of benefits from using AI in education: better engagement of students and personalized feedback. However, there is little research on whether AI can be expanded and be more equitable in terms of access, opportunity, and more ethically sound. This review aims to explore these aspects by synthesizing qualitative insight and evidence of empirical observations to analyse the influence of AI on student outcomes and its implementation across different educational settings based on a systematic literature review on AI technologies in education, including AIsupported assessments, virtual tutors, and adaptive learning systems between 2014 and 2024. The findings highlight that AI tools significantly improve student performance and engagement by leveraging real-time data analytics and personalized feedback. However, challenges such as computational resource demands, algorithmic biases, data privacy concerns, and the digital divide in under resourced settings limit broader adoption. Ethical issues, including the potential for AI to reinforce educational disparities and reduce human interaction, were also explored. The review underscores the importance of balancing AI integration with the irreplaceable human elements of teaching, such as fostering creativity, emotional intelligence, and critical thinking. This comprehensive analysis offers actionable insights for policymakers, educators, and developers to promote sustainable, equitable, and ethical AI integration in education, ultimately enhancing both teaching practices and student learning outcomes.