ChatGPT & AI Tutors: Ethical and Pedagogical Impacts of Large Language Models on Teaching Practices

Abstract

Recent advances in large language models (LLMs) such as ChatGPT have rapidly transformed the landscape of educational technology. Positioned as AI tutors, these systems offer personalized feedback, on-demand explanation, and adaptive learning support, leading to both excitement and concern among educators. This paper critically examines the ethical and pedagogical impacts of LLM-powered AI tutors on teaching practices in diverse learning contexts. Leveraging a theoretical synthesis of constructivist and socio-cultural pedagogies with contemporary research on artificial intelligence in education, this study analyses how ChatGPT influences teacher roles, learner autonomy, assessment practices, equity, data privacy, and instructional design.
The analysis reveals that while AI tutors enhance access to individualized learning support and can augment instructional capacity, they simultaneously raise ethical risks around algorithmic bias, data privacy, transparency of decision-making, and the potential erosion of professional agency. Pedagogically, LLMs reconfigure teacher-learner interactions, shifting the locus of some instructional tasks to AI systems; this shift has implications for deep learning, critical thinking, and the cultivation of socio-emotional skills. Through a mixed approach of literature synthesis and exploratory qualitative insights from educators, the paper proposes a set of principles for responsible integration of AI tutors, grounded in equity, accountability, and pedagogical alignment.
Ultimately, this research advocates for an educational ecosystem where AI tutors complement —rather than replace— human teaching practice, guided by ethical frameworks that foreground learner wellbeing and professional integrity. The paper concludes with recommendations for policy, future research, and practical guidelines for educators navigating the adoption of AI tutors.