Stroke is a cerebrovascular disease or brain injury that blocks blood vessels thereby limiting blood supply to the brain. Currently technology is growing. The medical community is greatly helped by the development of technology. One of them is a program that can be used to detect stroke with artificial intelligence. In this study, the data used came from the Kaggle.com website and the researchers used a machine learning method, namely random forest. Random forest is a combination of mutually independent classification trees that come from the same distribution through a voting process. Several stages were carried out in this study including the preprocessing, processing and evaluation stages. The results of this study are an accuracy of 99%.