Due to very high exposure to low-severity crashes, the vast majority of crashes that cause injury are of relatively low severity. Restraint systems that can adapt to the crash severity have a potential to reduce injuries in these high-exposure crashes. Even a small reduction in injury risk is likely to have a substantial practical effect due to the very high exposure to low-severity crashes. In this investigation two different adaptive restraint systems were investigated using contemporary technological constraints and new, proposed injury criteria target values balanced for equal risk of injury for all body regions:​ 15% injury risk for mid-severity crashes and 2% injury risk for low-severity crashes. It was found that both restraint systems performed better than a state-of-the-art restraint system in meeting the proposed injury criteria targets. Though this should not be considered a proof-ofconcept, these results do suggest that it may be feasible to reduce injury occurrence in crashes that are already of very low risk but result in high injury frequency due to very high exposure. Future work should include developing methods to assess the robustness of such systems, across dimensions of crash configuration, occupant, and sensing variability and its uncertainty that are present in the field.