Is ‘self-conscious intelligence about consequences and impact’ even possible, given the unknowable ways innovations may be used and develop?