Researchers have developed a machine learning model to predict gastrointestinal bleeding in critically ill cirrhotic patients hospitalized in the intensive care unit (ICU). The study used data from the MIMIC-IV database with a training group of 2,528 patients and a test group of 632 patients, while the model was validated on an additional 523 patients from the EICU database. Random Forest performed best with an accuracy of 0.86 in the training cohort and 0.74 in the test cohort, with a sensitivity of 0.68 and a specificity of 0.71. Key predictive factors were red blood cell count, hemoglobin level, platelet count, and anticoagulation therapy. The analysis showed that patients taking anticoagulants had an independently lower risk of gastrointestinal bleeding (odds ratio: 0.29). The model provides early risk stratification based on readily available clinical data and can support clinical decision-making in the prevention of complications in critically ill patients with cirrhosis.