For years, researchers have searched for ways to replace the insulin-producing beta cells of the pancreas to treat diabetes. Much attention has been paid to the identification of stem/progenitor cells that can be expanded and differentiated towards beta cells in view of obtaining a renewable source for cell transplantation therapy. Several cell types, including islet cells themselves and the surrounding exocrine cells, revealed a marked plasticity in their differentiation and/or proliferation capacity in vitro or in vivo. Despite a lot of research, there is no decisive evidence for the existence of adult stem cells in the pancreas. Many observations attributed to stem cells may be explained by the plasticity of mature somatic cells under specific experimental conditions. Embryonic stem cells show the greatest promise for generating functional beta cells and represent a renewable and therapeutically useful source. Implementation of knowledge gathered from developmental biology resulted in an efficient generation of definitive endoderm progenies including pancreas progenitors from human embryonic stem cells. Differentiation of functional beta cells from such progenies could be demonstrated following implantation in mice. These breakthroughs of the last decade indicate that clinical application of human embryonic stem cell-based diabetes therapy is getting closer.