As regulators make a move to supervise emerging AI technologies, the question inevitably emerges of whether we’ve embarked on a Sisyphean task. Should we be regulating the technology itself? Or might it be better instead to design governance frameworks within which AI will operate?