Generative Artificial Intelligence (GenAI) chatbots in education presents both opportunities and ethical challenges. While these tools can enhance inclusivity through personalized learning and assistive support, they also raise concerns. Without ethical oversight, GenAI risks reinforcing the very inequalities it seeks to address. This study conducted a Systematic Literature Review (SLR) of 43 peer-reviewed papers, guided by PRISMA methodology. Thematic analysis identified five key challenges in GenAI-supported education. These were interpreted using the Ethics of Care and Critical Pedagogy frameworks, which provided a lens to assess AI’s impact on student well-being, relational ethics, and educational equity. Findings show that GenAI chatbots often reproduce systemic bias, offer limited transparency, compromise data protection, and pose threats to academic integrity and the educator’s role. To address these issues, the study recommends bias-aware design, explainable AI models, AI-integrated curricula, robust data governance, and hybrid learning models that preserve human interaction.