We’re not going to get intracortical Brain-computer interface anytime soon. But I find it quite generative to ask—how close can we get to the sorts of interfaces we might want, through modalities we might achieve today?