Reliability of Artificial Intelligence in Translating Arabic Language and English Language Legal Texts رسالة ماجستير

Abstract

This study investigated AI reliability in Arabic-English legal translation from students' perceptions at Arab American University. The researcher used the mixed method approach and collected data through questionnaires and interviews. The study sample included 63 students (23 English majors and 40 Law majors) who are enrolled in legal translation and terminology courses. Additionally, the semi-structured interviews were conducted with four instructors from the English department. The results showed that using AI as a reliable alternative to human translators came with an arithmetic mean of 2.73 and a low score. In addition, the limitations for AI in translating Arabic language and English came with an arithmetic mean of 3.60 and a medium score. Regarding the benefits of AI, it came with an arithmetic mean of 3.77 and a medium score. Furthermore, the four instructors agreed that AI could not completely replace a human translator to ensure the accuracy and integrity of legal translation. Some of the participants believed that AI could not translate cultural differences and struggled to handle tone. Other participants stressed the importance of not fully relying on AI as technical text and the need to adhere to ethical decisions and comprehensive understanding. Based on the results of the study, the researcher suggests some recommendations, such as the necessity of human intervention besides AI tools to achieve accuracy and avoid ethical plagiarism and creating training workshops and educational curricula for target audiences, such as students, translators, researchers, and corporations.