https://jjcit.org/paper/203 10.5455/jjcit.71-1685381801 Yassine Saoudi,Mohamed Mohsen Gammoudi Chatbot,LLMs,Arabic conversational AI challenges,Generative artificial intelligence,Arabic question answering systems,Taxonomy,Performance evaluation 25 6254 2431 29-May-2023 27-Jul.-2023 and 19-Aug.-2023 20-Aug.-2023 Conversational systems have recently garnered increased attention due to advancements in Large Language Models (LLMs) and Language Models for Dialogue Applications (LaMDA). However, conversational Artificial Intelligence (AI) research focuses primarily on English. Despite Arabic being one of the most widely used languages on the Internet, only a few studies have concentrated on Arabic conversational dialogue systems thus far. This study presents a comprehensive qualitative analysis of critical research works in this domain to examine the strengths and limitations of existing approaches. The analysis begins with an overview of chatbot history and classification, then explores the language challenges encountered when developing Generative Arabic Conversational AI. Rule-based/Retrieval-based and deep learning-based approaches for Arabic chatbots are also examined. Furthermore, the study investigates the evolution of Generative Conversational AI with the advancements in deep-learning techniques. It also comprehensively reviews various metrics used to assess conversational systems.