Transformasi Berbasis AI dalam Pembelajaran Bahasa Arab Diferensiasi: Solusi Perencanaan Pembelajaran di SMA Bustanul Ulum Jember

Abstract

The aim of this research is to create and test a differentiated Arabic language learning model based on artificial intelligence (AI) to meet the diverse learning needs of students. At Bustanul Ulum Mlokorejo Jember High School, this research was conducted using a qualitative case study approach. Data were collected through interviews with 8 informants, participatory observations with 12 meetings, and document analysis (lesson plans, AI system logs, student portfolios). The research results show that machine learning analysis of certain linguistic errors (such as diacritics and syntax) enhances material personalization, and 65% of students reported being more motivated to learn. However, implementation faces challenges due to inadequate network infrastructure (40% of sessions experience latency) and a lack of teacher instruction to interpret algorithmic data. Cultural contextualization also emerged as an important finding; for example, incorporating examples from pesantren contexts increased learning relevance by 25%. Research shows that, when combined with data literacy-based teacher training, inclusive infrastructure policies, and cultural adaptation, artificial intelligence can become a transformative supporting tool in differentiated learning. This study suggests expanding the model to broader contexts and exploring artificial intelligence to reduce dialect bias in Arabic language learning.