Although basic research on human embryonic stem cells (hESCs) at the laboratory bench has progressed with enviable speed there has been little head way in terms of its clinical application. A look at the Internet however shows several stem cell clinics worldwide offering direct transplantation of undifferentiated hESCs to patients for the cure of a variety of diseases before bona fide evidence-based results can be demonstrated from large controlled studies. This raises concern because reliable protocols have to be first developed to resolve the three major hurdles delaying clinical trials such as inadequate cell numbers, immunorejection and tumorigenesis. Cell expansion methods using bioreactors, rotary culture and mitotic agents have now been developed to generate stem cell derivatives in large numbers. The problem of immunorejection can now be overcome with the development of indirect and direct reprogramming protocols to personalize tissues to patients (human induced pluripotent stem cells, hiPSCs; nuclear transfer stem cells, NTSCs; induced neuronal cells, iN). However, hESC, hiPSC, and NTSCs being pluripotent have the disadvantage of teratoma formation in vivo which has to be carefully addressed so as to provide safe stem cell based therapies to the patient. This review addresses the issue of tumorigenesis and discusses approaches by which this concern may be overcome and at the same time emphasizes the need to concurrently explore alternative stem cell sources that do not confer the disadvantages of pluripotency but are widely multipotent so as to yield safe desirable tissues for clinical application as soon as possible. J. Cell. Biochem. 111: 769–781, 2010. © 2010 Wiley-Liss, Inc.