AI in the translation classroom: tool or threat? A review of current research and pedagogical implications

Abstract

The fast growth of artificial intelligence (AI) translation tools, such as neural machine translation (NMT) systems, such as DeepL, and large language models (LLMs) like ChatGPT, has clearly changed how translation is taught and learned at the university level within philology programs. This article looks at recent studies and theories from 2024 to 2025 to better understand how AI tools are being used in translation classes today, what practical and analytical skills students are developing in this new context, what challenges teachers face while teaching translation in the AI era, and how they can gradually adapt their classes or couses to use CATs more effectively in order to prepare skilled and flexible translators for the modern job market. Research based on various studies that use different methods, such as mixed-methods, quasi-experimental, and systematic reviews, shows that AI tools really help students work faster and make better translated drafts that can later undergo further human correction and revision. These tools also let learners take more control of their studies, get better access to learning materials, parallel texts, and examples of ready-made translations for analysis, and quickly find specific information connected with different realia, terminology, and cultural nuances, etc. At the same time, there are always issues with understanding different cultures in depth, adapting meanings for the target audience, being accurate in practical situations, and choosing the right way to speak in different contexts, which always show that human judgment is still needed and cannot be replaced by AI tools. The article suggests using a human-centered approach to teaching translation with AI, where such tools are seen as things to study and think about critically, not just quick ways to get finished translations without human revision. Concrete classroom strategies, such as post-editing tasks, comparing human and AI translations, teaching basic AI literacy, and discussing ethical aspects, are suggested as ways to use AI's strengths while also improving the skills needed for modern professional translators.