Self-Efficacy and Perception in the Use of GAI Tools in the Translation Classroom

Abstract

The rise of generative artificial intelligence (GAI) tools in translation has sparked interest in their impact on translator training. This study examines the relationship between perception and self-efficacy in the use of GAI, specifically ChatGPT, within a specialized academic setting. The qualitative and exploratory research involved 30 final-year students from the Bachelor’s Degree in Translation at the Autonomous University of Baja California (UABC). Over the course of a semester, participants applied ChatGPT in pre-translation, translation, and post-editing tasks for specialized texts. To assess perception and self-efficacy in using these tools, a 15-item questionnaire with both closed and open-ended questions was used. The results indicate that a positive perception of ChatGPT is associated with greater confidence and performance in translation. However, challenges were identified, such as a lack of terminological accuracy and the need for specific training in post-editing. The findings suggest that the pedagogical integration of generative AI tools can strengthen translation competencies, reduce technological anxiety, and optimize the translation process. It is recommended to balance the use of these tools with teaching strategies that foster critical and creative skills. In conclusion, GAI presents both opportunities and challenges in translator training, highlighting the importance of proper implementation to maximize its benefits in professional practice.