This paper argues against considering incidental fndings (IFs) as potential benefts of research when assessing the social value of proposed research, determining the appropriateness of a study's risk/beneft ratio, and identifying and disclosing the risks and benefts of participation during informed consent. The possibility of generating IFs should be disclosed during informed consent as neither a risk nor beneft, but as a possible outcome collateral to participation. Whether specifc IFs will be disclosed when identifed is a separate question whose answer is material to determining whether IFs constitute a risk or a potential indirect beneft of participation. Finally, three types of IF should be distinguished and treated diferently during informed consent: those that will be routinely generated (e.g., results of testing to determine study eligibility), those that can reasonably be characterized in terms of their nature and frequency of generation (e.g., misattributed parentage), and those of unpredictable nature and frequency that can be characterized only in general terms. Research protocols should provide a rationale for sharing or not sharing IFs of these three types with participants. Regulatory review of such plans should not, however, be confused with regarding IFs as potential benefts when assessing the study's risk/beneft ratio or merit.