Background: Acute kidney injury (AKI) is a significant and prevalent complication of coronary artery bypass graft (CABG) surgery. Advanced age is an independent predictor of AKI; however, the existing research on AKI in elderly patients after CABG is limited. This study sought to employ machine-learning techniques to predict patients at high risk of developing AKI following CABG, using preoperative and intraoperative variables. Methods: Patients were retrospectively enrolled in this study between January 2019 and December 2020. The following nine machine-learning algorithms were used to predict postoperative AKI events: logistic regression (LR), simple decision tree (DT), random forest (RF), support vector machine (SVM), extreme gradient boosting (XGBoost), adaptive boosting (AdaBoost), gradient bosting, light gradient boosting machine (lightGBM), and K-nearest neighbor (KNN). SHapley Additive exPlanations (SHAP) values were employed to determine the contribution of each feature to the models and to assess feature importance. Receiver operating characteristic (ROC) curves were plotted, and the areas under the curves (AUCs) of the ROC curves were calculated to evaluate the predictive performance of the various machine-learning models for AKI. Results: A total of 2,155 participants were included in the study. The RF model had the highest AUC [0.737, 95% confidence interval (CI): 0.687–0.784], while the KNN model had the lowest AUC (0.644, 95% CI: 0.581–0.704). Certain variables, including age, the estimated glomerular filtration rate (eGFR), uric acid (UA), alanine aminotransferase (ALT), and B-type natriuretic peptide (BNP) at the baseline, as well as surgery duration and the intraoperative use of an intra-aortic balloon pump (IABP), were identified as significant risk factors for postoperative AKI. Conclusions: Machine-learning models can effectively predict the risk of AKI in elderly patients after CABG surgery. Among all the machine-learning models examined, the RF model showed the best performance.