Iftikhar and Anjum present a critique of a recent study that used interpretable machine learning to predict 1-year inguinal hernia risk after robot-assisted radical prostatectomy by Yu et al. They begin by praising the innovativeness of the study, but then highlight certain limitations such as: use of a homogenous data set, omittance of surgical technique variables, under detection of subclinical hernias, use of a short follow up time. The authors then recommend possible solutions including but not limited to: diverse cohorts, wider range of follow up times, and more comprehensive biological and surgical predictors to improve the model’s accuracy and ability to generalize.