One argument for the singularity is that artificial cognition will become so advanced that it will be able to alter itself, that eventually it will reach a point beyond which we won’t be able predict. This point of self-transformation functions iike an event horizon, beyond which we cannot see, or conceive. Hence it is metaphorically compared to the singularity at the middle of a black hole.
This is specious. First of all, there is no reason why in principle it would be impossible to speculate about the future — in other words no analogue to the physical limitations on the speed of light. Maybe as robots advance our understanding of them will advance with them, or maybe they’ll provide us some sort of gigantic tutorial so we don’t lag behind. As far as we know there is no past epoch which is separated from our present by an event horizon. If you asked Xerxes to speculate about the 21st century there is plenty he wouldn’t know but he would get some things right: there is sex, and wars, and banking, rivers are still important, people like dessert, and so on.
Secondly if the post-singularity state of the world is truly “inconceivable” then we have no grounds for saying it is singular, or not singular, or future, or past or anything. It’s just inconceivable. We can’t even say that it is not just extinction. We just don’t know. It’s a null claim.
Another argument that I just heard from a commenter here is that super-powers will tend towards the quick establishment of a super-empire, which will be unified. I don’t see why this should be the case — every global empire has fallen to internal division and revolt. Even if one of the super-advanced super-robots of the future were to discover a death ray at 10 am and conquer the world by 10:30, there’s no reason to believe that by 10:45 he would not be faced by revolts by whatever was the analogy of his generals and provinces. Perhaps even subsystems of his own super-self.
Would he not then try to incorporate every other being in the world until Borg-like, he reigned alone? This brings us I think to the nub of the mistake of the singularity fantasists. It assumes that given awesome power a super being would use it to be safe and alone rather than say use it to help the other robots do their thing. It assumes he would not want to have a romantic relationship, or friendship, or children, or just partners for different pursuits.
That is to say the singularity theorists think logic tells us that a barricaded lone ego is the logical result of progress. Actually they have a previous belief that a barricaded lone ego is the best thing or the most fundamental reality and therefore interpret motion towards that state as progress. As in so many things if you scratch a grand philosophy of history you find personal psychopathology.