<![CDATA[This study investigates the implementation of Artificial Intelligence (AI) in the Arabic Language Content Development Curriculum at a higher education institution, grounded in a constructivist theoretical framework. The primary aim of the study is to analyze changes in students’ learning outcomes who took a course before and after the AI implementation, and to examine the influence of confounding variables such as age, gender, origin school/region, and core subject on improvements in Arabic language learning outcomes. The research design is a true experimental design with a single experimental group applied to students in the Arabic Language and Literature program. Participants met inclusion criteria with a minimum sample size of 32 respondents. Data were collected through pretests and posttests, questionnaires, observations, and focused interviews. The instruments included sentence-construction items with difficulty levels from easy to hard (11 items) and evaluations of the Istimak Baseline, Istimak Advanced, and Ta’bir Shafawi Baseline courses. Statistical analysis encompassed pretest–posttest comparisons and significance tests on several item pairs (p 0.05). Overall, there was an increase in the average score of learning outcomes after the AI intervention, indicating improved understanding of Arabic material. Significant improvement overall was achieved in some item pairs; several items showed pattern variation that warrants further contextual analysis. Confounding variables exhibited different trends across categories (age, gender, origin region, core subject), with AI benefits more evident for certain groups. Participants’ attitudes toward AI showed a positive perception regarding fairness, transparency, and relevance of the material, though concerns about data privacy, plagiarism, and potential over-reliance emerged. The findings support a constructivist framework combined with AI to provide more interactive, personalized, and contextual Arabic language learning. AI tools such as LingQ contribute to personalization, increased engagement, and ongoing feedback, while requiring ethical guidelines for AI use, privacy protection, and efforts to maintain academic integrity and reduce bias. Limitations include the single-group experimental design and potential Hawthorne effects. Recommendations include developing AI ethics guidelines, SAP training for instructors, and risk mitigation strategies such as human–AI balance and fair access for all students. ]]>
