PURPOSE OF REVIEW: There is a growing interest in the applications of artificial intelligence in pediatric rheumatology. Although concerns with training datasets, ethical considerations, and the need for a major utilization of explainable artificial intelligence are still ongoing challenges, significant advancements have been made in recent years. In this review, we explore the most recent applications of artificial intelligence in pediatric rheumatology, with a special focus on machine learning models and their outcomes. RECENT FINDINGS: Supervised and unsupervised machine learning models have been largely employed to identify key biomarkers, predict treatment responses, and stratify patients based on disease presentation and progression. In addition, innovative artificial intelligence driven imaging tools and noninvasive diagnostic methods have improved diagnostic accuracy and emerged as encouraging solutions for identifying inflammation and disease activity. Large language models have been utilized for patient-based questions with promising results. Nevertheless, critical examination and human oversight are still crucial in interpreting artificial intelligence's outputs. SUMMARY: Artificial intelligence is revolutionizing pediatric rheumatology by improving diagnosis and disease classification, patient stratification and personalized treatment. However, we are only at the beginning, and the adventure has just begun.