I'm not convinced that's true. There are countless innovations that we were able to get to work despite having little understanding of the underlying theory. Gi

I'm not convinced that's true. There are countless innovations that we were able to get to work despite having little understanding of the underlying theory. Given how adaptable both biological and artificial neural networks are, I'd say that the hardware is far and away the primary limiting factor, not our theoretical understanding. Once we get a safe and reliable connection between brain and computer, we can figure out the rest through trial and error.