Identification of protein mutations that drive cancer is a major challenge. A primary reason is that driver mutations are principally identified by their high frequency even though they can also be rare. Driver mutations can locate at functional (binding or active) sites. We dub these orthosteric drivers. However, often they are allosteric drivers. Identification is particularly formidable for rare allosteric drivers. Autoinhibition, where a segment of the protein covers its functional site, is a common allosteric regulation mechanism. A modest shift in the equilibrium can switch the system from the autoinhibited to the active state. This can suggest why (i) mutations are likely to evolve to target it; (ii) inhibitors can straightforwardly relieve the autoinhibition but not vice versa; and why (iii) mutations that relieve the autoinhibition are likely to be drivers—even if they are rare. We explain in simple terms the linkage between allosteric driver mutations, release of autoinhibition, free energy landscapes, and targeted pharmacology in precision medicine. We review the literature and propose new concepts in identification of rare drivers in this framework.