Brain arteriovenous malformations (BAVMs) are an important cause of intracerebral hemorrhage (ICH) in young adults. Biological predictors of future ICH risk are lacking, and controversy exists over previous studies of natural history risk among predominantly ruptured BAVM cohorts. Recent studies have suggested that the majority of BAVMs are now diagnosed as unruptured lesions, and that the risk according to natural history among these lesions may be less than previously assumed. In the first part of this review, the authors discuss available data on the natural history of BAVMs and highlight the need for future studies that aim to develop surrogate biomarkers of disease progression that accurately predict future risk of ICH in BAVMs. The etiology of BAVM remains unknown. Recent studies have suggested a role for genetic factors in the pathogenesis of sporadic BAVM, which is further supported by reports of familial occurrence of BAVM and association with known systemic genetic disorders (such as Osler-Weber-Rendu disease, Sturge-Weber disease, and Wyburn-Mason syndrome). Molecular characterization of BAVM tissue demonstrates a highly angiogenic milieu with evidence of increased endothelial cell turnover. Taken together with a number of reports of de novo BAVM formation, radiographic growth after initial BAVM diagnosis, and regrowth after successful treatment of BAVM, these findings challenge the long-held assumption that BAVMs are static lesions of congenital origin. In the second part of this review, the authors discuss available data on the origins of BAVM and offer insights into future investigations into genetics and endothelial progenitor cell involvement in the pathogenesis of BAVM. Current treatment options for BAVM focus on removal or obliteration of the lesion in an attempt to protect against future ICH risk, including microsurgical resection, endovascular embolization, and stereotactic radiosurgery (SRS). In the third part of this review, the authors discuss available data on SRS in BAVMs and highlight the need for future studies on the radiobiology of BAVMs, especially in regard to biomarker detection for tracking SRS response during the latency period. Insights from future investigations in BAVM may not only prove important for the development of novel therapies and relevant biomarkers for BAVM, but could also potentially benefit a variety of other disorders involving new vessel formation in the CNS, including stroke, tumors, moyamoya disease, and other cerebrovascular malformations.