Scilight Press

Large language models, such as ChatGPT, use vast textual data to mimic human-like responses and interact with users in a natural conversational way. This capability has introduced several opportunities and challenges in the field of language education and has sparked both excitement and concern among language educators. This study seeks to contribute to the ongoing debate surrounding the advantages and drawbacks of using ChatGPT in the language classroom by attempting to evaluate its effectiveness in Arabic language learning, with a particular focus on teaching case endings. To achieve this, this study analyzes and discusses data collected from ChatGPT learning sessions, quizzes, and surveys conducted with university students studying Arabic. The findings of this study revealed that while ChatGPT was efficient in providing detailed and personalized explanations, it was deficient in providing accurate and reliable information regarding Arabic case endings.