It may be wise to temper one’s enthusiasm for Musk’s hyperbole with a recognition that he has one talent only: the ability to extract huge sums of money from gullible investors. When it comes to delivery, it’s a whole different story. Tesla is still struggling to make a tiny fraction of the number of vehicles routinely turned out by real auto-makers. SpaceEx is successful despite, rather than because, of Musk and is about to commit the crime of polluting the skies with thousands of satellites that will wreck terrestrial astronomy (and he can’t be bothered to file flight plans with the relevant bodies). The Boring Company is pure hype and economically non-viable.
So when it comes to a fantasy about “connecting the human brain to AI” not only do we have to take Musk’s real-world record into account but also recognize that we’re at least 20 years away from having technology that does more than the occasional parlor trick. First of all, we still have no real understanding of how most of the brain works. Secondly, interfacing with the brain is a non-trivial problem (do we really want to have electrodes implanted into our brains, with the subsequent infection & tissue growth around them as the body’s defense mechanisms kick in?). And of course the brain works in a very different way from a computer chip so any meaningful inter-connection is at least another 20 years in the future.
While it’s charming to see the ease with which Musk pumps out hype and the enthusiasm with which a certain section of the population accepts his hype uncritically, it may be wiser to add a touch of caution and mix in some appreciation for the reality of the problem domain before becoming irrationally exuberant. Musk is a showman, and rarely bothers to learn enough about the things he’s excited about. That’s why there’s always a huge gulf between hype and execution in Muskworld.