the key assumption here is that these “superbabies” will naturally hold the “correct” moral values that they will then program into a superintelligent AI system, which will then elevate humanity into a golden period where we get to live in a techno-utopia amongst the stars.
which is pretty weird and has some uncomfortable implications
smart people are still capable of being pieces of shit. Eliezer’s whole “we need to focus everything on augmenting human intelligence” thing pretty much glosses over this. It only takes one group of superbabies/augmented intelligence humans getting into some fascist shit for this to blow up in his face.
Rationalism, among other things, is supposed to cure you of being a piece of shit, in fact it’s such a flawless epistemic hack that it’s a common belief among them that it’s impossible for two sufficiently rationalist individuals to disagree, as in come to different conclusions about the same assumptions. So if you think some rat influencer is full of shit, it could in fact be you that hasn’t yet attained the appropriate level of
thetansrationalism.So yeah they’ll just have the superbabies read the sequences and Harry Potter fan fiction.
Also, any talk of alignment should be seen in light of them being mostly ok with humanity going poof by this time next week, as long as the stage is set for whatever technological facsimile of consciousness they deem reasonably human-like to inherit the cosmos.
considering Yud’s previous comments on nuking data centres and bombing Wuhan, I wouldn’t be surprised if he’s cool with smart fascists programming their values into an AI and controlling it because “at least not all of humanity is dead, and there are humans living amongst the stars in a utopia!”
Superbabies is a backup plan; focus the energy of humanity’s collective genetic endowment into a single generation, and have THAT generation to solve problems like “figure out how to control digital superintelligence
The academic institutions in charge of exploring these ideas are deeply compromised by insane ideologies. And the big commercial entities are too timid to do anything truly novel; once they discovered they had a technology that could potentially make a few tens of billions treating single gene genetic disorders, no one wanted to take any risks; better to take the easy, guaranteed money and spend your life on a lucrative endeavor improving the lives of 0.5% of the population than go for a hail mary project that will result in journalists writing lots of articles calling you a eugenicist.
oh no, not a eugenicist!
I got caught on that quote too…
Superbabies is a backup plan; focus the energy of humanity’s collective genetic endowment into a single generation, and have THAT generation to solve problems like “figure out how to control digital superintelligence”.
Science-fiction solutions for science-fiction problems!
Let’s see what the comments say!
Considering current human distributions and a lack of 160+ IQ people having written off sub-100 IQ populations as morally useless […]
Dude are you aware where you are posting.
Just hope it never happens, like nuke wars?
Yeah that’s what ran the Cold War, hopes and dreams. JFC I keep forgetting these are kids born long after 1989.
Could you do all the research on a boat in the ocean? Excuse the naive question.
No, please keep asking the naive questions, it’s what provides fodder for comments like this .
(regarding humans having “[F]ixed skull size” and can therefore a priori not compete with AI):
Artificial wombs may remove this bottleneck.
This points to another implied SF solution. It’s already postulated by these people that humans are not having enough babies, or rather the right kind of humans aren’t (wink wink). If we assume that they don’t adhere to the Platonic ideal that women are simply wombs and all traits are inherited from males, then to breed superbabies you need buy-in from the moms. Considering how hard it is for these people to have a normal conversation with the fairer sex, them both managing to convince a partner to have a baby and let some quack from El Salvador mess with its genes seems insurmountable. Artificial wombs will resolve this nicely. Just do a quick test at around puberty to determine the God-given IQ level of a female, then harvest her eggs and implant them into artificial wombs. The less intelligent ones can provide eggs for the “Beta” and “Gamma” models…
But you don’t go from a 160 IQ person with a lot of disagreeability and ambition, who ends up being a big commercial player or whatnot, to 195 IQ and suddenly get someone who just sits in their room for a decade and then speaks gibberish into a youtube livestream and everyone dies, or whatever.
These people are insane.
195 IQ and suddenly get someone who just sits in their room for a decade and then speaks gibberish into a youtube livestream and everyone dies, or whatever.
I can’t even decipher what this is about. Like if you’re 195IQ you can invent Avada Kedavra in a decade?
esprit d’escalier
this whole “superbabies will save us from AI” presupposes that the superbabies are immune to the pull of LW ideas. Just as LW are discounting global warming, fascism etc to focus on runaway AI, who says superbabies won’t have a similar problem? It’s just one step up the metaphorical ladder:
LW: “ugh normies don’t understand the x-risk of AI!”
Superbabies: “ugh our LW parents don’t understand the x-risk of Evangelion being actually, like, real!”
this whole “superbabies will save us from AI” presupposes that the superbabies are immune to the pull of LW ideas.
Well, if they fall for less wrong, then we know they’re not all that smart
So AGI is 0.5-2 years away. After which the singularity happens and due to AI alignment we either are immortal forever, or everybody is diamondoid paperclips.
A normal human takes 18 years to grow to maturity. So for the sake of the argument (yes yes, don’t hand it to ISIS) a supergene baby can do that in 9 years. (poor kid). Those timelines seem at odds with each other (and that is assuming the research was possible now).
I know timelines and science fiction stories are a bit fluid but, come on, at least pretend you believe in it. I’m not saying he is full of shit but… no wait, I am saying that.
I took a look at the “AGI 0.5-2 years away” quote, and he said he’s using the term in the “pre-2020” way, where we would of “debated whether or not Claude Opus is AGI or not” so not exactly robot god, but he’s still expecting LLMs to be able to act as advanced secretaries for the government. I have no clue what he means by “cluster” and if he was being literal with the molecular stuff or that’s just figurative language
Still stupid tho
As we know, the critical age for a boy genius is somewhere from 11 (Harry Potter) to 15 (Paul Atreides), so the gene-enhanced baby ought to have a fair shot after a few months or so.
