News for hackers. Stuff that matters.
Re:Super-Intelligent Humans Are Coming by jerf on Thursday October 16, 2014 @03:53PM (#8467218)
From the article:

"This means that there must be at least thousands of IQ alleles to account for the actual variation seen in the general population. A more sophisticated analysis (with large error bars) yields an estimate of perhaps 10,000 in total."

The odds that you can flip a few thousand genes without harm seem to me to approach 0 rapidly. None of these genes are, after all, coding for "intelligence"... they're doing some sort of chemical thing, after all. Heck even if it produces a viable embryo it seems more likely to produce insanity than superintelligence (assuming we could even distinguish the two).

That said, yeah, I'm just hypothesizing, but it seems far more likely to me that rather than some sort of monotonic increase in the "intelligence" factor that we can just pile on freely that there will be any number of incompatible additions, things that where adding either one or the other of some modification takes you to the optimal point but having both takes you beyond, and, well, all the other exciting and whacky things we see in the genetic world. In the arbitrarily-complicated n-dimensional space of "intelligence" just ramping all the available controls to "maximum" doesn't strike me as likely to produce an optimal result.

(I did, however, resist giving a8da6b0c91d's answer; a priori we can't tell whether the Ashekenazi Jews are made susceptible by their intelligence, or if they merely got something nasty co-selecting with their intelligence. Perhaps somebody somewhere has made that particular study, I don't know, but my point is merely that study would be required on a case-by-case basis. That can't be used as a general argument when we're discussing intelligently modifying genetics.)

Re:Super-Intelligent Humans Are Coming by gwern on Thursday October 16, 2014 @06:13PM (#8468132)
> The odds that you can flip a few thousand genes without harm seem to me to approach 0 rapidly.

I have bad news for you, then: in the general populations, there's not merely thousands of SNPs, there are hundreds of thousands! I guess we're all going to die tomorrow now that we've realized this.

> In the arbitrarily-complicated n-dimensional space of "intelligence" just ramping all the available controls to "maximum" doesn't strike me as likely to produce an optimal result.

Yet, breeding works. Chickens and cows, for example, are scores of standard deviations away from their ancestors on some traits like weight or milk production, and while they may not be exactly as healthy or robust as their ancestors, they have hardly gone extinct.

Re:Super-Intelligent Humans Are Coming by jerf on Friday October 17, 2014 @12:15PM (#8471671)
"I have bad news for you, then: in the general populations, there's not merely thousands of SNPs, there are hundreds of thousands!"

Evolution isn't direct manipulation... part of what has evolved in the genome is the ability to be evolved in the first place. The genetic operations being proposed by this article bears no resemblance to the operations used by evolution.

While I've previously indicated my disagreement with the other sibling chain of discussion going on here, the car analogy isn't the worst. You can't take a sedan and just Push Everything To Eleven (individually, one criterion at a time, with no attention paid to integration) and expect a high-performance vehicle to result.

In math terms, we have a function f(x, y) -> IQ, and where we are now, we know going in the positive x direction will raise IQ and we know going in the positive y direction will increase IQ. We have no guarantee whatsoever that x + y will result in an increase in IQ. We certainly have even less guarantee when we start piling on the dimensions, and my comment about expecting a non-viable embryo is because in fact the xs and ys and zs and so on aren't just affecting IQ (since there aren't actually genes for that) but are having other effects as well. The odds of crossing into a nonviable regime somewhere, even from a starting position of known functionality, strike me as almost 1.

I sense I may be goring a bit of a sacred ox here, so let me remind you of my supreme confidence that genetic engineering can indeed raise IQ. It's just going to be harder than what this article proposes to produce a supermegagenius.

Re:Super-Intelligent Humans Are Coming by gwern on Friday October 17, 2014 @12:57PM (#8471941)
> Evolution isn't direct manipulation... part of what has evolved in the genome is the ability to be evolved in the first place.

What on earth do you think breeding does? It selects for increasing proportions of genetic variants in the phenotypically above-average members of the flock. What difference is there between creating a predictive score for an embryo and choosing to implant the highest-scoring one, with making the changes predicted to increase scores with something like CRISPR?

> You can't take a sedan and just Push Everything To Eleven (individually, one criterion at a time, with no attention paid to integration) and expect a high-performance vehicle to result.

Organisms are not cars. This is remarked upon by everyone how biology designs things in a very different fashion from humans, and yet this knowledge gets thrown out as soon as inconvenient...

> In math terms, we have a function f(x, y) -> IQ, and where we are now, we know going in the positive x direction will raise IQ and we know going in the positive y direction will increase IQ. We have no guarantee whatsoever that x + y will result in an increase in IQ.

Please, look up the twin and GCTA studies and what is meant by 'additive'. If there were those complex wiggly interactions, they would not show up as hits from the GWAS studies since they're not additive, and they would not contribute to the large additive fraction of heritability but the other parts.

Re:Super-Intelligent Humans Are Coming by jerf on Friday October 17, 2014 @02:25PM (#8472538)
"What on earth do you think breeding does? It selects for increasing proportions of genetic variants in the phenotypically above-average members of the flock."...

... using certain operators that are not arbitrarily recombining DNA. "Crossover" and such may be somewhat oversimplified versions of what real life does, but neither are the real life genetic operations anywhere near "completely free" recombinations.

Your arguments based on conventional evolutionary operations do not apply to cases where we are engineering freely, without regard to what appears where on chromosomes or any of the other myriad ways we've evolved the ability to safely evolve.

I do also feel like perhaps you are sneaking a step in where the intelligent manipulator double-checks whether the gene combinations make sense, which, if so, would be subtly begging the question as my point is precisely that we would have to check.

Besides, if I may flip the burden of proof around for a moment and appeal to something that may only be a heuristic rather than a solid logical argument, do you really think superintelligence is going to be this easy? "Just" look up all the thousands of genes that contribute to intelligence, assume they must all be doing it linearly (in the "linearly-combinable" sense of the term), and flip them to "smart" rather than "dumb"? Really? It's never that easy, even for things multiple orders of magnitude simpler than genetically engineering intelligence. It would boggle my mind if the path, or even a path, to human superintelligence could be so thoroughly expressed in so few bits.