The intensive care unit (ICU) was initially developed in the 1950s to treat patients who required invasive respiratory support and hemodynamic resuscitation. Since the beginning, ICU medicine has focused on maintaining sufficient arterial blood flow and oxygenation to provide adequate tissue oxygen delivery to forestall or reverse organ failure. Over time, ICU medicine became more intensive, with the administration of many diagnostic tests and monitors, invasive procedures, and treatments, often with scant evidence of benefit associated with them. An alternative perspective holds that ICU patients may represent a group of patients that is especially vulnerable to iatrogenic harm. We outline a case that presents common ICU dilemmas and discusses current data that propose that “less is more” when making key diagnostic or therapeutic choices in the ICU. Further, we assert that providers should skeptically consider common ICU interventions, trying to account for the potential unintended consequences of interventions. Finally, we suggest that the guiding principle of ICU medicine should be primum non nocere: in delicate situations, it may be better not to do something, or even to do nothing, rather than risk causing harm.