The present study aimed to investigate the relationship between artificial intelligence and educational innovation from the perspective of elementary school principals in Tehran's District 1. The present study method was descriptive-correlational. The statistical population included all elementary school principals in Tehran's District 1 in the academic year 2023-2024, totaling 190 people. Using the Cochran table and simple random sampling method, a sample size of 127 people was determined and entered into the study. The research tools for collecting data included the Artificial Intelligence Questionnaire (Chen et al., 2022) and the Educational Innovation Questionnaire (Pazaki and Afzal Khani, 2016). Then, data analysis was performed using the structural equation modeling method and with the help of smart pls software. The findings showed that there is a significant positive relationship between artificial intelligence and the total score of educational innovation along with its components (p<0.01). There is a significant positive relationship between all components of artificial intelligence and educational innovation (p<0.01). Also, the coefficient of determination for the criterion variable of educational innovation was 0.343, which indicates that artificial intelligence explains a total of 34.3 percent of the variance of educational innovation. As a result, it can be said that in education and training and educational centers, special attention can be paid to artificial intelligence and its capabilities for educational innovation.