Artificial Intelligence and the Commentary of the Diwan al-Hamasah: A Comparative Study in Light of Arabic Linguistics (With a Focus on Tarmah ibn Hakim al-Ta’i as a Model)

Abstract

The aim of this study is to conduct a comparative analysis between the commentary of the Diwan al-Hamasah by al-Marzouqi and the commentary generated by the AI model ChatGPT 5 ،  supported by Natural Language Processing (NLP) technologies ،  specifically applied to four verses of satire and pride by Tarmah ibn Hakim al-Ta’i. This research employs a qualitative comparative methodology to identify similarities and discrepancies within the four verses based on the analysis of Arabic linguistics ،  alongside a quantitative approach using the AraBERT AI model ،  specifically designed for Arabic language processing. This AI model analyzes the text extracted from both commentaries ،  applying modern mathematical techniques to measure congruence and divergence in meaning ،  allowing for an accurate and objective evaluation of AI's ability to simulate human commentary. The results of the study indicate similarities between the two commentaries in expressing the general meanings and fundamental concepts of the text. However ،  there are also discrepancies in the outcomes of the integrated analysis of the text ،  particularly in terms of poetic vocabulary ،  morphological structure ،  syntactic structure ،  rhetorical dimension ،  and cultural and social context. These differences arise due to issues such as AI’s errors in understanding ،  analysis ،  and processing ،  as well as AI’s consideration of dimensions not addressed by al-Marzouqi. The study concludes that artificial intelligence serves as a complementary tool that enhances the understanding of Arabic literary texts ،  rather than being the primary or sole means of interpretation.