SIVs are non-pathogenic in their natural hosts, but their transmission to a new host, such as humans for HIV-1 and HIV-2 as well as macaques for SIVmac, enable them to become highly pathogenic. HIV-1 and HIV-2 share 40–60% nucleotide sequence homology. The transmission rate of HIV-2 is lower because the viral load is generally lower in infected individuals. The natural history of HIV-2 infection is quite different from that of HIV-1. Although clinical symptoms of acquired immunodeficiency syndrome (AIDS) caused by HIV-1 and HIV-2 are indistinguishable, most people infected with HIV-2 do not progress to AIDS. One strong predictor of disease progression that distinguishes pathogenic HIV infection and non-pathogenic SIV infection is the activation of host antiviral defence including a prominent stimulation of T cells in the former but not the latter. Another possibility is that the natural hosts of SIVs might be the survivors of ancient SIV pandemics. One prediction is that HIVs and humans will eventually adapt to each other just like SIVs and their natural hosts. In this regard, AIDS might be considered an accident in which HIVs fail to adapt to humans or humans fail to adapt to HIVs. In support of this view, species-specific features in host restriction factors, such as TRIM5α and tetherin, can prevent SIV infection of humans. On the other hand, adaptive mutations and accessory genes such as Vpu, Nef, and Vif in HIVs have been found to counteract host restriction factors, which constitute the antiviral defence, in a host-specific manner. For example, a five-codon deletion in the cytoplasmic domain of human restriction factor tetherin results in the prevention of its interaction with SIVcpz Nef in humans [45]. On the side of the virus, some HIV-1 strains use their Vpu protein to degrade tetherin [45].