Informed consent has been a mainstay of all ethical research guidelines since the 1970s, but the proliferation of field experiments in the social sciences – which include audit experiments, correspondence experiments, canvasing experiments, social media experiments, and information experiments – has brought with it an increasing resistance to procuring informed consent. This essay grapples with the now common practice of denying research subjects an opportunity to voluntarily consent to participate in research. It provides a framework for thinking about virtual consent, a situation in which the researcher consents for participants. Drawing on a Rawlsian thought experiment, I argue that ethical research is that to which a reasonable person, not knowing whether she would be the subject or the scientist, would consent. This type of reasoning provides a way for thinking about potential downstream consequences not just for the individual subject, but also for society writ large. Yet, because virtual consent does not entail voluntary participation, in constitutes a bronze standard, rather than a best practice, for ethics in experiments.