When AI Translates: Skill Devaluation and Learner Boredom in Translation Classrooms for Application-Oriented EFL Majors

Abstract

The integration of generative artificial intelligence (AI) into translation education has predominantly been celebrated for its capacity to enhance learner engagement and motivation. This study challenges that dominant narrative by investigating a countervailing possibility: that generative AI, by rendering translation skills seemingly redundant, may hollow out learners’ perceived value of those skills and thereby breed boredom. Drawing on the Meaning and Attention Components (MAC) model of boredom (Westgate & Wilson, 2018), we propose and empirically test a pathway in which perceived AI-induced skill devaluation predicts translation classroom boredom through meaning deficits, with the attention pathway (challenge-skill imbalance) serving as a competing explanation. Using an explanatory sequential mixed-methods design with 326 application-oriented EFL translation majors in China, we found that: (a) perceived AI skill devaluation significantly predicted translation classroom boredom; (b) meaning deficits fully mediated this relationship; and (c) the meaning pathway exerted significantly stronger effects than the attention pathway. Qualitative interviews with 18 participants corroborated these findings, revealing narratives of “why bother learning what AI can do.” This study introduces the construct of meaning-deficit boredom in AI-era translation classrooms, extends the MAC model to foreign language education, and offers a critical boundary condition to the engagement-dominant AI narrative.