Background: The effectiveness of psychological interventions is undisputed. But while in other fields of health care the safety of interventions is studied alongside effectiveness, adverse events (AEs) have only recently been assessed in clinical studies of psychological interventions. This critical review summarizes the definition, assessment and current research status of AEs of psychological interventions. Summary: AEs are defined as any untoward event or unfavorable change that occurs in the course of a psychological intervention. AEs that are caused by the intervention can be classified into side effects of correctly applied treatment, malpractice (i.e., incorrectly applied treatment) and unethical conduct (e.g., sexual abuse). Ideally, they are assessed by independent raters or alternatively by self-report questionnaires that should also cover serious adverse events (SAEs, e.g., suicide attempts or self-injurious behaviors). About 1 to 2 in 3 patients report at least 1 AE and results of meta-analyses suggest that treatments might differ in frequency and/or severity of AE and in treatment acceptability (measured as dropout rates). Key Messages: Measures of AEs and SAEs as well as more nuanced descriptions of dropout should be included in all clinical studies of psychological interventions. If this happens, we might learn that psychological interventions differ with respect to AEs, SAEs and acceptability. As many psychological interventions are about equally effective, they might one day be chosen based on differences in their safety profile rather than their differential effectiveness. Ideally, reducing AEs might also lead to more effective interventions.