The concern expressed in The Center for AI Safety’s Statement of AI Risk seems justified to me, but it also seems to me that many of the signatories have still not grasped the real nature of that risk. It’s the second order effects that’ll do us in—not the singularity and its presumptively implacable AI overlords, but rather the symbiotic processes already inherent in pervasive computing, processes which we can all sense, but are still in denial about what it will take, in terms of an evolution in human consciousness, to successfully navigate those spaces which still exist between where the machine ends and we begin.
In his 1960 Critique de la Raison Dialectique, Jean-Paul Sartre indulged himself in a typically poetic digression about how we can’t tell—may never be able to tell—whether we’re dreaming the machine, or the machine is dreaming us. This is a commonplace now, but although it wasn’t entirely new in 1960, it was still controversial enough to meet with widespread ridicule among the opinion makers of the day. And of course Sartre was describing the strictly physical interactions of humans and industrial age assembly lines, when machines were dumb, and humans were still thought to be the masters no matter how deeply their own mental processes were conditioned by the mechanical repetitions of their jobs.
The machines today are no longer dumb, and we can no longer afford the illusion that we are the masters of either the physical or the mental aspects of the machine/human symbioses of the 21st century. I’m not sure why, but I’m not as bothered by this as the signatories of this letter are telling me I ought to be. It certainly isn’t because I’m an optimist in the narrow sense ot the term. I expect great darknesses in our future, but not the ones that are supposedly keeping the tech bros up at night. These latter day idiot savants aren’t the real heralds of our new distempered age, it’s the kids now glued to TikTok all day. What their stewardship of our future will look like remains beyond anyone’s current power to predict. To make a long story short, it’s not the end of humans that should concern us, but the end of humanism, which seems to be losing its grip on the tiller of this ship of fools we’re crewing well before a new helmsman is ready to take its place.