<![CDATA[The lack of a conducive bi'ah lughawiyyah (Arabic-speaking environment) remains the primary challenge in strengthening Arabic vocabulary (mufradat) among Madrasah Ibtidaiyah (MI) students, while the Kurikulum Merdeka demands learning that is differentiated, contextual, and student-centered. This article aims to formulate the concept of Generative AI-based digital bi'ah lughawiyyah as a new conceptual framework in Arabic mufradat learning at MI, map its implementation models relevant to the principles of Kurikulum Merdeka, and identify the opportunities, challenges, and integration strategies within the context of Islamic education at MI. This study employs a library research method with an integrative literature review design through critical synthesis of cross-disciplinary literature, educational technology, applied linguistics, second language acquisition, and curriculum policy to produce an original conceptual framework. The findings indicate that Generative AI such as ChatGPT, Google Gemini, and Large Language Model (LLM)-based applications has considerable potential to create an interactive, adaptive, and immersive digital bi'ah lughawiyyah through four implementation models: contextual conversation simulation, Arabic generative narratives, game-based adaptive mufradat practice, and teacher-driven instructional media creation. This article concludes that the integration of Generative AI as a digital bi'ah lughawiyyah is pedagogically relevant and aligned with the differentiated learning principles of Kurikulum Merdeka, provided that its implementation is carried out in a planned and systematic manner, grounded in Islamic values, so that AI serves as a complement, not a substitute for the role of teachers in Arabic language learning at MI.]]>
Copyrights © 2026
