Abstract

This study investigates the impact of Large Language Models (LLMs), such as ChatGPT, on student learning success (LS) and teaching efficacy within higher education. Using a systematic literature review (SLR) methodology, 130 high-quality, peer-reviewed sources were analysed to examine how LLMs influence pedagogical practices, student engagement, skill development, and ethical considerations. The findings identify both progressive and regressive factors associated with LLM integration. Positively, LLMs support personalised learning, foster critical thinking, enhance digital and AI literacy, and offer real-time feedback that promotes self-regulation and metacognitive growth. However, concerns include over-reliance on AI, ethical dilemmas, algorithmic bias, and a reduction in human-led discussions, all of which risk diminishing educational equity and intellectual depth. The study concludes that responsible, transparent, and inclusive deployment of LLMs is essential to maximise their educational benefits while mitigating risks. It contributes to the discourse on AI in education by offering a balanced framework for understanding how LLMs reshape teaching and learning dynamics. Recommendations are provided for educators, institutions, and policymakers to support ethical AI integration and foster student success in an increasingly AI-mediated academic environment.

Download

Share

COinS