To tackle the rampant drug trafficking among teenagers and students, the National Narcotics Agency (BNN) and related parties have developed the detection of drug users by involving experts directly. However, this method has two important obstacles (1) experts who are not always in place, (2) the limitations and complexity of test equipment, labor, time and large costs. To overcome this obstacle, several studies have developed an Expert System (ES) which is the application of Artificial Intelligence (AI) to detect drug users. However, various research records reveal that these drug ES models have not found a good level of accuracy for detecting drug users. This study aims to apply the Case Base Reasoning (CBR) algorithm to detect drug users among teenagers. This research produces a web-based drug ES application with the CBR algorithm. This research contributes in two ways, first on the scientific side by applying the CBR algorithm to detect drug users, second on the practical side by providing recommendations for BNN or related institutions in relation to preventing drug trafficking among adolescents and students.