التّرجمة-الآليّة-للشّعر-من-وإلى-العربيّة-في-ظلّ-الذّكاء-الاصطناعي_-تحليل-نقدي-لمخرجات-chatgpt-في-ضوء-التّحدّيات-الجماليّة-والثقافيّة.pdf

Automated Translation of Poetry to and from Arabic in the Era of Artificial Intelligence: A Critical Analysis of ChatGPT Outputs in Light of Aesthetic and Cultural Challenges

Abstract

The contemporary era is witnessing a qualitative shift in translation tools, driven by the rapid advancement of artificial intelligence—raising new questions about the extent to which machine models, particularly ChatGPT, can engage with poetry as a highly complex linguistic and artistic text. The significance of this study lies in its attempt to explore the limitations of machine translation when applied to poetic texts between Arabic and the English and French languages, given the dense symbolism and stylistic specificity inherent in such texts. Accordingly, this paper aims to critically examine the model’s performance in conveying aesthetic dimensions and cultural imagination, whether translating from Arabic or into it. The analysis has led to findings that reveal semantic and aesthetic gaps that are difficult to bridge without conscious human intervention. In light of this, the study underscores the necessity of developing translation mechanisms that are more attuned to the structure and intent of poetry, in order to ensure a more faithful rendering of the texts’ artistic identity