Since the early 1980s, Epstein-Barr virus (EBV) infection has been described as one of the main risk factors for developing multiple sclerosis (MS), and recently, new epidemiological evidence has reinforced this premise. EBV seroconversion precedes almost 99% of the new cases of MS and likely predates the first clinical symptoms. The molecular mechanisms of this association are complex and may involve different immunological routes, perhaps all running in parallel (i.e., molecular mimicry, the bystander damage theory, abnormal cytokine networks, and coinfection of EBV with retroviruses, among others). However, despite the large amount of evidence available on these topics, the ultimate role of EBV in the pathogenesis of MS is not fully understood. For instance, it is unclear why after EBV infection some individuals develop MS while others evolve to lymphoproliferative disorders or systemic autoimmune diseases. In this regard, recent studies suggest that the virus may exert epigenetic control over MS susceptibility genes by means of specific virulence factors. Such genetic manipulation has been described in virally-infected memory B cells from patients with MS and are thought to be the main source of autoreactive immune responses. Yet, the role of EBV infection in the natural history of MS and in the initiation of neurodegeneration is even less clear. In this narrative review, we will discuss the available evidence on these topics and the possibility of harnessing such immunological alterations to uncover predictive biomarkers for the onset of MS and perhaps facilitate prognostication of the clinical course.