The information on the toxicity of pharmaceuticals on human health is of paramount importance. While animal models are a conventional approach to toxicological assessment, computational methods have been on the rise in the last couple of decades. These methods, spanning from simple structure–activity relationships to more complex models, provide cost-effective and ethical alternatives to traditional methods. This chapter provides an overview of the application of computational methods for the prediction of the human toxicological endpoints of pharmaceuticals. The representative examples of the prediction of high-priority toxicity endpoints for pharmaceuticals in the drug development process and regulatory toxicology were introduced.