The relentless expansion of the HIV pandemic has demonstrated that the need for a vaccine is desperate. However, the development of an effective vaccine against HIV is a formidable challenge. It is likely that a successful vaccine will have to induce an immune response consisting of not only neutralizing antibodies targeted to conserved epitopes of the viral envelope and cytotoxic T-lymphocytes targeted to a variety of viral antigens, but also local mucosal immunity. Furthermore, a vaccine should induce broad-spectrum immunity covering all HIV subtypes. It is unlikely that a single vaccine will achieve all this, and a combination of vaccines will probably be necessary. Although no efficacious HIV vaccine is available yet, definite progress has been made. It was demonstrated that chimpanzees could be protected from both cell-free and cell-associated HIV challenge. Protection from mucosal challenge was also demonstrated in several studies and limited cross-protection between HIV subtypes was observed in several animal models. In spite of these successes, much remains to be done. Prototype vaccines studied to date have only induced short-lived immune responses and elicited no antibodies able to neutralize clinical isolates of HIV-1. Novel ways of producing HIV-1 envelope antigens may lead to improved antibody responses and raise the chances of a vaccine inducing long-term protective immunity.