Interest in bedside ultrasound in the intensive care unit and emergency department has exploded in recent years. This interest is driven in part by the utility of ultrasound for procedural guidance. In most cases, enthusiasm outstrips current evidence. While ultrasound is often felt to be risk-free, the important risk of ultrasound is the chance of false diagnosis. The vividness of visual images may make practitioners especially prone to cognitive errors in interpretation. Possible applications of ultrasound include management of shock and respiratory failure, two complex syndromes with multiple aspects. Reasonable evidence supports use of ultrasound to guide volume expansion, although its value remains to be demonstrated in an explicit protocol. Other possibilities include ventilator titration and guidance of diuresis. While the literature is more complicated, there is some early evidence that lung ultrasound may improve the diagnosis of dyspnea, although these results have not been well validated. Centers should avoid premature loss of equipoise and participate in studies of explicit protocols that incorporate ultrasound.