AIM: to study modern approaches to the application of machine learning and deep learning technologies for the management of patients with colorectal cancer.MATERIALS AND METHODS: after screening 398 publications, 112 articles were selected and the full text of the works was studied. After studying the full texts of the articles, the works were selected, machine learning models in which showed an accuracy of more than 80%. The results of 41 original publications were used to write this review.RESULTS: several areas have been identified that are the most promising for the use of artificial intelligence technologies in the management of patients with colorectal cancer. They are predicting the response to neoadjuvant treatment, predicting the risks of metastasis and recurrence of the disease, predicting the toxicity of chemotherapy, assessing the risks of leakage of colorectal anastomoses. As the most promising factors that can be used to train models, researchers consider clinical parameters, the immune environment of the tumor, tumor RNA signatures, as well as visual pathomorphological characteristics. The models for predicting the risk of liver metastases in patients with stage T1 (AUC = 0.9631), as well as models aimed at assessing the risk of 30-day mortality during chemotherapy (AUC = 0.924), were characterized with the greatest accuracy. Most of the technologies discussed in this paper are software products trained on data sets of different quality and quantity, which are able to suggest a treatment scenario based on predictive models, and, in fact, can be used as a doctor’s assistant with very limited functionality.CONCLUSION: the current level of digital technologies in oncology and in the treatment of colorectal cancer does not allow us to talk about a strong AI capable of making decisions about the treatment of patients without medical supervision. Personalized treatment based on the microbiotic and mutation spectrum and, for example, personal pharmacokinetics, so far look fantastic, but certainly promising for future developments.