<![CDATA[Purpose: The study aims to develop and evaluate an AI-assisted animated video designed to improve elementary students’ understanding of cuboid and cube concepts. The research responds to the persistent difficulty young learners face in visualizing three-dimensional objects and seeks to provide a learning medium that is valid, practical, and instructionally effective. Method: The development process followed the ADDIE framework, involving needs analysis, media design, production, classroom implementation, and evaluation. Data were gathered through observations, interviews, expert reviews, student questionnaires, and learning outcome tests. Three experts assessed the media’s content accuracy, visual quality, and language clarity. The effectiveness of the media was examined using a pretest–posttest design with experimental and control groups, and the Mann–Whitney U test was applied due to non-normal score distributions. Findings: The expert validation procedures produced an average score of 92.33 percent, indicating a high level of validity across content, presentation, and linguistic dimensions. Student responses yielded a practicality rating of 89.22 percent, showing that the media was easy to use and engaging. The experimental group achieved a higher posttest mean score of 84.00, compared with 57.60 in the control group. Statistical analysis confirmed a significant difference between the groups (p < 0.05), demonstrating that the animated video supported clearer conceptual understanding of cuboid and cube topics. Significance: The findings highlight the pedagogical benefits of integrating AI-generated visual representations into early mathematics instruction. The media helped reduce abstraction barriers, matched the cognitive needs of young learners, and proved to be a promising digital resource for foundational geometry learning. The study contributes an empirically supported model for future AI-enhanced educational media development within mathematics education.]]>
