Drug-induced toxicity is a major barrier to the development of new therapeutic agents. The majority of drug candidates that cause toxicity are screened out at the discovery or development stage. Some, however, are not detected until the drug is in late-stage clinical trials or has become available to the public. Often, this is caused by what is known as an 'idiosyncratic drug response', and several theories have been proposed regarding the underlying mechanisms for idiosyncratic toxicity. These include the formation of reactive metabolites in certain individuals due to the presence of polymorphisms in drug-metabolizing enzymes, immune-mediated responses to the drug (or one of its metabolites), the combination of drugs with low-level inflammatory reactions, and drug-induced mitochondrial toxicity. This review will focus on recent evidence supporting these theories, as well as some of the preclinical models that can be used to test and study idiosyncratic drug responses.