As a computational cognitive neuroscientist, for over 50 years, we certainly do not understand how the brain computes. There are lots of efforts out there giving us hints.
Neurolink is clearly giving us another tool and probably the right scientists and engineers to make the best uses of it.
What was not mentioned in this article, which I find a bit surprising, is the gross difference in scale of the measurements and actions. Those wires sound small, but they are not at neuronal scale, and they are vastly too few. Both by many orders of magnitude.
That said, Einstein proved the existence and size of atoms by observing and statistically modelling pollen dust shaking on the surface of water. So the data we get from neuorlink may tell us something we do not already know about how the brain computes.
I personally think there are better windows on how the brain computes. All the windows need to be tested against each other, since each reveal something fundamental about that computation.
My own study has led me to how the brain computes lies we tell and understand. Pigs lie, but unfortunately, cannot talk. I would be interested to see if he can get a glimmer on predication activity in the cortex. Most of the examples are in fact described as predications, but I do not believe analysed that way, yet. We do not have the tools that can watch the details of those computations.
Regarding the security issues, and being a computer security guy too, I think the parts of this article that discuss this are perfectly good and sound.
That said, I do not believe people understand how important understanding predications relate both to how the brain computes and the security implications of that because..
From that study we can do more important things than telling whether people are lying the way described, such as..
