What is it about?
Detecting whether an Arabic poem is written by a human or generated by AI is challenging, because Arabic poetry follows strict rules of meter (prosody) and often reuses familiar patterns and structures. In this research, we developed a method that analyzes both the reusability of poetic patterns and the prosodic features to distinguish human-authored poems from AI-generated ones. Our approach helps preserve Arabic poetic heritage and can be used in systems for detecting machine-generated content in literature.
Featured Image
Photo by Nahrizul Kadri on Unsplash
Why is it important?
Arabic poetry is a cornerstone of cultural heritage, but with the rise of advanced AI that can generate convincing poetry, it's becoming harder to tell human creativity apart from machine output. This matters for preserving authenticity in literature, education, and publishing. Our method helps researchers, educators, and content moderators detect AI-generated poems, supporting efforts to protect original human expression and maintain trust in creative works.
Perspectives
Working on this research has deepened my commitment to protecting Arabic poetry from AI-generated imitation, and I am proud that this work can be generalized to other types of content and other languages.
Yazan Ghunaim
Read the Original
This page is a summary of: Arabic AI Poem Detection Based on Reusability and Prosody, ACM Transactions on Asian and Low-Resource Language Information Processing, June 2026, ACM (Association for Computing Machinery),
DOI: 10.1145/3820659.
You can read the full text:
Resources
Contributors
The following have contributed to this page
