Resolving Artificial Intelligence's Trust Problem