This article aims to develop and validate an Autism Classification system using the Histogram of Oriented Gradient (HOG) feature extraction method and Support Vector Machine (SVM). The research problem focuses on the difficulties in identifying early signs of autism in children based on their facial features and patterns. To approach this issue, we use theoretical references from the field of artificial intelligence and digital image processing. This system is developed based on data collected through empirical research, which is then qualitatively and quantitatively analyzed using the SVM method. The results show that the system achieves an accuracy rate of 88%, precision of 86.5%, recall of 87%, f1-Score of 86.5% and f2-Score of 86 demonstrating the system's effectiveness in identifying children who may have autism. The article concludes by stating that the developed Autism Classification system shows significant potential in the early detection of autism in children. This suggests that AI-based facial analysis can be a valuable tool in helping parents and healthcare professionals identify early signs of autism. Recommendations for further research are to validate the system's performance on larger and more diverse datasets to increase its reliability and generalizability.