I am a neuroscientist. I wish this company luck. I hope they have come upon a solution that helps people gain some control over their lives.
That being said, having decades of experience with implanted neurophysiological devices, my initial though is 'Putting probes in small diameter blood vessels, partially occluding them and creating turbulence and eddies in the blood flow and possibly dislodging sclerotic debris from the vessel walls - what could possibly go wrong?'. Perhaps the technology from stents mitigates these concerns, but I hope they take a cautious approach when testing this.
Honestly I think the best brain-computer interface will likely be optical, using small arrays of LEDs (think mobile phone screens) and light sensors to measure blood oxygenation from the surface of the cortex. Optical devices are becoming ever more flexible, miniaturised, and energy-efficient and measuring blood oxygenation with light doesn't block blood vessels or make the brain a pin-cushion (as best I can tell the approach Musk's company is taking). Moreover, most implanted probes, like electrodes, will be encased by thin films of scar tissue eventually that will diminish the biological signals, but certain wavelengths of light can penetrate tissue. Anyway, researchers have been working on this problem for least least half a century - an easy solution that has been overlooked is improbable.
