The generation of induced pluripotent stem cells (iPSCs) from somatic cells demonstrated that adult mammalian cells can be reprogrammed into a pluripotent state by introducing defined transcription factors. iPSCs show almost identical properties in self-renewal and pluripotency, and can circumvent ethical concerns because they do not use embryonic materials. Therefore, iPSCs from a patient's somatic cells have great potential in studying drug development and regenerative medicine. Several human disease models have already been established using patient-specific iPSCs from Parkinson's disease and familial dysautonomia. Moreover, the correction of genetic defects by homologous recombination has already been accomplished with Fanconi anemia patient-specific iPSCs. However, the generation of patient-specific iPSCs for clinical application requires alternative strategies, because genome-integrating viral vectors may raise tumorigenic risk after transplantation. Moreover, the use of iPSCs for eventual clinical application is limited by the low efficiency of current methods for reprogramming. Studies on the mechanism underlying the reprogramming and on establishment of non-integration methods contribute evidence toward resolving the safety concerns associated with iPSCs. Small molecules involved in the epigenetic modification and signaling pathway not only improve reprogramming efficiencies, but also bypass the addition of certain reprogramming factors. However, reprogramming somatic cells purely by small molecule treatment still remains a challenge. Here, we review recent progress made by the use of transcription factors and small molecules that can either replace reprogramming factors or enhance reprogramming efficiency. We also discuss the progress that has been made in the rapidly moving iPSC field, with an emphasis on understanding the mechanisms of cellular reprogramming and its potential application to cell therapy.