Flavivirus infection usually results in fever accompanied by headache, arthralgia, and, in some cases, rash. Although the symptoms are mild, full recovery can take several months. Flaviviruses encode seven nonstructural proteins that represent potential drug targets for this viral family. Focusing on the Zika virus NS2B-NS3 protease, we uncovered a unique inhibitor, MH1, composed of aminothiazolopyridine and benzofuran moieties. MH1 inhibits ZVP with a biochemical IC50 of 440 nM and effectively blocks cleavage of ZVP substrates in cells. Surprisingly, MH1 inhibits the other flaviviral proteases at least 18-fold more weakly. This same phenomenon was observed in assays of the viral cytopathic effect, where only Zika virus showed sensitivity to MH1. This selectivity was unexpected since flaviviral proteases have high similarity in sequence and protein structure. MH1 binds at an allosteric site, as demonstrated by its ability to stabilize ZVP synergistically with an active site inhibitor. To understand its selectivity, we constructed a series of hybrid proteases composed of select segments of ZVP, which is sensitive to MH1, and dengue virus protease, which is essentially insensitive to MH1. Our results suggest that MH1 binds to the NS3 protease domain, disrupting its interaction with NS2B. These interactions are essential for substrate binding and cleavage. In particular, the unique dynamic properties of NS2B from Zika seem to be required for the function of MH1. Insights into the mechanism of MH1 function will aid us in developing non-active-site-directed, pan-flaviviral inhibitors, by highlighting the importance of evaluating and considering the dynamics of the NS2B regions.