HIV-1 can be detected in the brain/CNS, and more conveniently in cerebral spinal fluid (CSF), at all times after infection. Its presence reflects a number of processes ranging from the trafficking of infected T cells through the establishment of an independently replicating population (compartmentalization) where the virus has evolved to infect a host cell with a low density of CD4 (CD4low phenotype). A deeper understanding of these multiple processes has come from a clearer definition of viral entry phenotypes. A common misconception in the HIV-1 field is that all viruses that use CCR5 (R5 viruses) are macrophage-tropic. In fact, macrophage-tropic/ viruses, with their ability to enter cells with a low density of CD4 (as seen on macrophages), are rarely found in the blood and have not been detected among transmitted/founder viruses. The main form of HIV-1 found in the blood, and the form that enters the CNS by trafficking in infected T cells, uses CCR5 but requires a high density of CD4 (as is seen on CD4+ T cells) for efficient entry; this predominant form of HIV-1 is more appropriately called R5 T cell-tropic. This clearer understanding of HIV-1 entry phenotypes has allowed a reassessment of when and where macrophage-tropic/CD4low viruses evolve and their role in pathogenesis. Earlier studies highlighted the detection of CD4low viruses in the CNS and their link to severe CNS disease at late stages prior to death, such as HIV-associated dementia (HAD). The evolution of macrophage-tropic viruses can be viewed as an evolutionary path the virus follows when its target CD4+ T cells become limiting, a situation that is especially common behind the blood-brain barrier in the CNS. Persistent viral infection in the brain is likely to be very different from infection of CD4+ T cells in tissues such as lymph nodes, spleen, and GALT. The question of viral escape in the CNS during suppressive therapy, either transient or persistent, can now be addressed in the context of viruses that are adapted to replication in this environment. Similarly, rebound virus in the CSF after treatment interruption may provide insight into the presence of compartmentalized reservoirs.