Novice Translators’ Perceptions of AI-Assisted Translation

Abstract

This study investigates how novice translators perceive and engage with AI-assisted translation tools, emphasizing their emerging AI literacy and its impact on translation learning. Twelve undergraduate students enrolled in an introductory translation course participated in semi-structured interviews, sharing their experiences with tools such as Google Translate, DeepL, ChatGPT, and Gemini. Data were analyzed thematically to identify patterns in tool usage, perceived benefits and concerns, and attitudes toward AI’s role in learning. The findings indicate that novice translators actively experiment with multiple AI tools, often combining them strategically to support vocabulary, grammar, phrasing, and text naturalness. Participants appreciated AI’s efficiency and scaffolding potential but expressed concerns about overreliance, skill erosion, cultural inaccuracies, and ethical or privacy issues. Despite these concerns, they consistently viewed AI as a supportive resource rather than a replacement for manual translation, emphasizing the importance of human judgment, critical evaluation, and post-editing. Instructor guidance also played a role in shaping responsible AI use, with mixed stances encouraging careful, reflective engagement. The study underscores the need for translation curricula to incorporate AI literacy, enabling learners to leverage AI’s benefits while maintaining foundational skills, autonomy, and ethical awareness. These insights highlight the potential for human-centered, hybrid approaches in AI-integrated translation education.