How Technology Is Changing the Arabic Language Forever | Prof. Nizar Habash

Language serves as the vital intersection of cultural identity and technological innovation, yet the rapid rise of AI reveals a significant representation gap for the nearly many million ways the Arab world communicates. Professor Nizar Habash, a computer science professor and director of the CAMEL Lab at NYU Abu Dhabi, explores the historical anxieties surrounding technological shifts, drawing a direct parallel between the 150-year delay of the Arabic printing press and contemporary concerns regarding data bias in large language models. The conversation navigates the inherent challenges of modeling a language characterized by immense dialectal variety and non-standardized orthography, shifting the focus from perceived linguistic complexity to the practical need for bespoke, regional data sets. As the field transitions from rigid, rule-based systems to sophisticated neural models capable of abstract meaning through embeddings, the dialogue underscores the urgency of building localized, open-source tools like JAIS and FAMAR to ensure the Arab world's history and diverse voices are accurately represented in the digital age.   00:00 Introduction 02:15 How AI Models Understand Diverse Arabic Dialects 05:48 Debunking Myths About Arabic’s Linguistic Complexity 08:21 The Historical Construction of Modern Standard Arabic 11:00 Impact of Technology on Arabic Script and Printing 15:19 The Journey into Computational Linguistics 17:35 The Evolution of Machine Translation and LLMs 23:51 Explaining Hallucinations in Statistical Language Models 31:14 Cultural Representation and Bias in Image Generation 38:13 Dialect Support and Limitations in Translation Tools 50:02 Camel Lab's Mission for Open-Source Arabic NLP 01:03:42 Descriptive vs. Prescriptive Views of Arabic   Nizar Habash is a Professor of Computer Science at New York University Abu Dhabi (NYUAD), and the director of the Computational Approaches to Modeling Language (CAMeL) Lab. Professor Habash specializes in natural language processing and computational linguistics.  He received his PhD in Computer Science from the University of Maryland College Park in 2003. He has two bachelors degrees, one in Computer Engineering and one in Linguistics and Languages from Old Dominion University. His research includes extensive work on machine translation, morphological analysis, and computational modeling of Arabic and its dialects. Professor Habash has been a principal investigator or co-investigator on over 25 research grants. And he has over 250 publications including a book entitled "Introduction to Arabic Natural Language Processing". Professor Habash is one of the recipients of the King Salman Academy for Arabic Language Award (2022); and he is the recipient of the Antonio Zampolli Prize (2024). His website is www.nizarhabash.com.   Connect with Nizar Habash 👉 https://www.instagram.com/nizarhabash/ Hosted by: Mikey Muhanna 👉 https://www.instagram.com/mikey_mu/?hl=en-gbFOLLOW & RATE THE AFIKRA PODCAST:» Apple Podcasts: https://podcasts.apple.com/lb/podcast/the-afikra-podcast/id1529437743» Spotify: https://open.spotify.com/show/5nafoF1Zs7F48mGZjlhrze» Anghami: https://play.anghami.com/podcast/1014643869THIS SERIES IS PART OF THE AFIKRA PODCAST NETWORK Explore all episodes in this series: https://youtube.com/playlist?list=PLfYG40bwRKl5xaTkBDrUKLCulvoCE8ubX&feature=sharedABOUT AFIKRAafikra | عفكرة is a movement to convert passive interest in the Arab world to active intellectual curiosity. We aim to collectively reframe the dominant narrative of the region by exploring the histories and cultures of the region – past, present and future – through conversations driven by curiosity.📍 Local events in 40+ locations worldwide http://afikra.com/chapters🎧 New podcasts + videos weekly http://afikra.com/podcasts⚡ Become a member: https://www.afikra.com/membership🔗 Instagram: https://www.instagram.com/afikra_🔗 Facebook: https://www.facebook.com/afikra.official🔗 Twitter: https://twitter.com/afikra Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.