Twenty years after its recognition, HIV / AIDS has become the most important infectious disease globally and the leading cause of death in Africa. A preventive vaccine represents the best long-term hope for its control. The development of such a vaccine, however, has encountered a number of scientific challenges, including the lack of information on immune correlates of protection, the limitations in our understanding of the relevance of primate protection experiments in relation to vaccine-induced protection in humans, and the significance of genetic and immunologic variability of HIV strains for potential vaccine efficacy. Despite these uncertainties, the first phase I trial of an HIV vaccine was conducted in the United States in 1987. Since then more than 30 candidate vaccines have been tested in over 70 phase I / II clinical trials in both industrialized and developing countries. The first HIV vaccine trial in a developing country was conducted in 1993, six years after the first trial in the United States. Since then eighteen phase I / II trials and one phase III trial have been or are being conducted in developing countries, and additional phase II and III trials are planned to start in 2003. Most of these initial trials have been conducted in Thailand, but more recently trials have been initiated in Africa, Latin America and the Caribbean. Over the past years, the HIV vaccine development effort has followed three major overlapping paradigms. The first “wave” of candidate vaccines aimed at inducing neutralizing antibodies. The second wave focused on stimulation of CD8+ T-cell responses. The current “wave” of HIV vaccine research is aimed at optimizing both humoral and cell-mediated immune responses. The first generation of candidate vaccines (based on the HIV envelope protein) entered phase III efficacy evaluation in 1998, and definitive results from these trials will become available in 2003. Plans to ensure wide access to future HIV vaccines must be developed well in advance.