Fair. But if AGI takes as long as it took DNA, we won't be around to see it. If we want AGI, we need something better than fumbling around at random in the dark.
That doesn't make any sense either, once we came up in the picture what took evolution millions of years humans do in just a few decades: look at how we bred animals and plants and completely transformed them without having any idea about DNA or evolution itself.
But here we're talking about creating a new thing with a desired property that we can't even define, from scratch- not by speeding up its evolution in a thing that already has that property, like we did with animals and plants.
Not to mention: far as I can tell nobody has yet been able to breed, say, sharks, and completely transform them to be useful to humans, like cows or potatoes.
Sorry, I don't understand what you mean by 'this "human exceptionalism"'. What is going to happen in the near future?
[I wrote a longer comment where I made many assumptions about what you might have meant but I hate doing that, so sorry if you happened to read it. It wasn't offensive or anything, I just genuinely have no idea what you mean and so I replaced it with this comment. I am sincerely asking you to explain what you mean because I didn't get it. Apologies! Communication is hard]
i think googling it would be helpful, but in an overly general way, human exceptionalism is the idwa that there is sometjing unique or special associated with the human experience or with human experience.
The broader point i was making in the original comment is agaisnt this sentiment. Humans abiluty to unserstand why things happen bears little relevance to them happening. We now know little if anything about why the human consious experience happens or why it works, yet here we are.
Humans can make gun powder or a water wheel with no understanding of electron orbital theory or an concept of gravity that outlines orbital mechanics.
The shocking effectiveness of piling up large amounts of complexity seems to be sufficient many of the phenomena that humans seem to desure a much more mechanistic explanatiom for. In this sense, it could be that complexity is all you need.
and to bring it full circle, what about the complexity we see in non human, non animal systems? if complexity is all you need, what might be happening beyond the veil of human assumptioms around the consious experience?
Well, I don't think that complexity "is all you need". "All you need" for what, anyway?
I think I understand what you mean by human exceptionalism, but I think it's besides the point. We don't even understand the intelligence of unicellular organisms, or insects. We have no "AI" that's half as smart as a cockroach.
The point is, there was never anything consequential that we invented, or discovered, all by chance. People always had some understanding of how things work, even if it wasn't understanding in the context of modern science, so even when it was only empirical knowledge, not formalised in the language of mathematics; even if it wasn't full knowledge, and yes, there are still many things we don't quite understand today. Off the top of my head, if I got that right, we don't understand exactly why or how anesthesia works. And yet, we have trained professionals who study and practice anesthesia, and they generally manage to avoid killing their patients, most of the time.
Unfortunately, we have nothing like that about intelligence. There's people studying the subject, coming at it from many different points (insect intelligence, IQ metrics, psychology, neuroscience, what have you) but there is not a body of knowledge that we can apply practically to show how intelligence works, and what happens when we do this, or what happens when we do that, to it.
Which means, if we do achieve "AGI" or whatchamacallit, at the present time it's not going to be because someone knew how to do it, but because someone got really lucky (or unlucky, perhaps).
