Ethical principals demand that science foster human goodness and flourishing—serving as a tool to help, not harm, humanity. This chapter considers as a case study the Bucharest Early Intervention Project (BEIP), a randomized controlled trial comparing the effects of foster care as an alternative to institutional care for young children in Bucharest, Romania. The study results suggested that early institutionalization leads to profound deficits in many domains. The ethical issues triggered by these experiments span the domains of randomization to harm, clinical equipoise, standard of care, researcher ambivalence, duty to rescue, political cooptation and conflicts of interest, and logicality. The alleged methodological rigor of the randomized controlled BEIP has a seductive internal consistency, but careful analysis reveals that gloss distracted the scientific community from our fundamental ethical obligation to provide authentic witness to, and compassionate care for, the most vulnerable among us.