Although there is growing concern as to the questionable value of animal-based methods for determining the safety and efficacy of new medicines, which has in turn led to many groups developing innovative human-based methods, there are many barriers to their adoption for regulatory submissions. The reasons for this are various, and include a lack of confidence that the available human-based methods, be they in vivo, in silico or in vitro, can be sufficiently predictive of clinical outcomes. However, this is not the only problem: the issue of validation presents a serious impediment to progress, a particularly frustrating situation, in view of the fact that the existing animal-based methods have never themselves been formally validated. Superimposed upon this is the issue of regulatory requirements, where, although regulators may be willing to accept non-animal approaches in place of particular animal tests, nowhere is this explicitly stated in their guidelines. Such problems are far from trivial, and represent major hurdles to be overcome. In addition, there are a range of other barriers, real or self-imposed, that are hindering a more-predictive approach to establishing a new drug's clinical safety and efficacy profiles. Some of these barriers are identified, and ways forward are suggested.