The red chili variety "inko hot" is a type of red chili that has a high economic value. Unfortunately, these red chili plants are often infected with anthracnose disease, which results in significant losses for farmers. Anthracnose is one of the major diseases infecting chili plants, potentially resulting in crop failure and losses of up to 80%. The purpose of this study is to develop a classification system to identify anthracnose disease in red chili fruit, using Convolutional Neural Network (CNN) method. In this experiment, 1500 data were used, of which 80% were used as training data and 20% as validation data. The best results of this experiment produced a model with an accuracy of 97% and a loss rate of 6.45%, by applying the Nadam optimization algorithm and going through 50 iterations (epochs). The model showed good performance with a prediction accuracy rate of 83.33%. The development of this classification system has significant potential in providing efficient solutions to recognize diseases in chili plants. Through continuous development, this system can be a valuable tool for farmers to increase crop productivity and reduce the negative impact of disease attacks on red chili peppers and other crops.