Demo versions of the datasets and the code that are used for the analyses of the study "Estimating diagnostic uncertainty in artificial intelligence assisted pathology using conformal prediction". The repo can be used to replicate the core tables and figures in the manuscript. Unreliable predictions can occur when an artificial intelligence (AI) system is presented with data it has not been exposed to during training. In this study, we develop conformal predictors for AI-assisted prostate pathology. We show how these predictors can be used to detect unreliable predictions due to changes in tissue preparation techniques in different laboratories, digitization utilizing different digital pathology scanners, and the presence of atypical prostatic tissue, such as variants of prostatic adenocarcinoma and benign mimics of cancer. We believe conformal prediction can have widespread utility in ensuring patient safety of clinically implemented AI systems. See the repository (https://github.com/heolss/Conformal_analyses) for up to date information.