The rapid integration of generative artificial intelligence (GenAI) tools into language education is reshaping pedagogical practices, learner engagement and institutional responsibilities. While these technologies offer novel opportunities for personalised learning, linguistic support and broader access, they also give rise to significant and often underexplored ethical challenges. This chapter presents a critical review of empirical research on the ethical dimensions of ChatGPT use in language learning, focusing on ChatGPT as a widely adopted form of GenAI. Drawing on peer-reviewed journal articles, the review identifies and examines recurring ethical concerns, including academic integrity, privacy, equity and governance and ethical pedagogy. In addition, the chapter engages with these challenges by revisiting existing AI ethical frameworks. It also outlines implications at learner, teacher and institutional levels. This review contributes to a more nuanced understanding of how GenAI transforms not only the mechanics of language learning but also its ethical foundations, hoping to serve as a stepping stone for researchers, educators and policymakers alike to explore, reflect on and shape the responsible integration of GenAI in language education.