Despite calls by some commentators for disclosing incidental findings in genetics research, several factors weigh in favor of caution. The technology of genetics has the power to uncover a vast array of information. The most potent argument for restraint in disclosure is that much research is pursued without consent so that the individual participant may not know that research is being conducted at all. Often the work is done by investigators and at institutions with which the person has no prior contact. Past practice is also relevant; genetics researchers historically have chosen not to disclose incidental findings, of which misattributed paternity and pleiotropic alleles such as ApoE have been the most common. Many people choose not to have genetic tests when given a choice. It may be desirable to discuss the topic of incidental findings when consent for research is obtained, but given the risk of unwanted surprise when there has been no prior discussion, the potential utility of incidental findings should be very high before they are even offered to individuals.