This study is critically examines the integration of Artificial Intelligence (AI) into Adaptive Learning Systems (ALS), specifically within the context of Elementary School research. The study aims to map the theoretical foundation, identify thematic trends, and reveal significant design gaps in this research field. Bibliometric Analysis was applied to 77 documents indexed in the Scopus database, covering the period from 2015 to 2025. The analysis applied included publication performance analysis, author co-citation analysis to uncover the intellectual structure, and keyword co-occurrence analysis for thematic trend mapping. The results show a significant surge in publications post-2021, reaching 23 documents in 2024, with Uzbekistan as the largest volume contributor. Co-citation analysis confirms that the intellectual structure of this field is shaped by authors from East Asia. The thematic cluster analysis reveals a dominant focus on AI technology, adaptive pedagogy, and learner performance. This study concludes that despite the field having a stable theoretical foundation, there is a mismatch in geographical and thematic focus, leading to fundamental design gaps. These crucial gaps include a lack of attention to children's data ethics, the integration of the teacher's role, and pedagogical design models that are less suited to the elementary education context