Although the diagnosis of multiple sclerosis relies on the demonstration of disease dissemination in space and time, the exclusion of other neurological disorders is also essential. The limited specificity of abnormalities disclosed by MRI may increase the likelihood of diagnosis of multiple sclerosis in patients affected by other disorders. The available criteria for diagnosis of multiple sclerosis have not taken advantage of the potential of MRI to detect features "not suggestive" of multiple sclerosis. Recognition of such features in the work-up of patients suspected of having multiple sclerosis may reduce the likelihood of a false positive diagnosis of the disorder in some, while suggesting the correct alternative diagnosis in other patients. On the basis of this, a workshop of the European MAGNIMS (Magnetic Resonance Network in Multiple Sclerosis) was held to define a series of MRI red flags in the setting of clinically suspected multiple sclerosis that is derived from evidence-based findings and educated guesses. The presence of such red flags should alert clinicians to reconsider the differential diagnosis more extensively. In this review we will report on the conclusions of this international consensus, which should represent a first step beyond the concept of "no better explanation", and inform future diagnostic criteria for multiple sclerosis.