Genetic Engineering and unknown consequences

[Rainbow]

Rainbow, this I think refers to the paradigm of a broad philosophical movement called "extropianism", or a branch called "singularitarianism"

Thanks Stavia!

I realize that thinking in terms of singularities does not usually follow from the experiences of everyday existence: It can take a while to adjust.

Rainbow, there is no doubt whatsoever that the Singularity is approaching.

I'm sure there'll be growth and development, but I'm wondering how we can predict that the growth won't level off and plateau at some point. Max More mentions this idea early in the following video:

[MarcR]

J11, the IQ singularity can only unfold over generations and must overcome massive cultural inertia to do so. I think the AI singularity is much closer and will render our predictions about change on a biological time frame moot. This is an excellent, accessible introduction to the topic from WaitButWhy:

The AI Revolution: The Road to Superintelligence
The AI Revolution: Our Immortality or Extinction

[Brian4]

J11, I'm curious: do you think it's possible to say (without any amount of doubt) that there will be a cognitive singularity?

The empirical evidence – the Fermi Paradox – suggests that at a certain level of development, civilizations may just end. (May – or they ALL become silent, which seems unlikely.) Whether they end after a massive increase in intelligence (the term singularity seems wrong to me) or not, I have no idea. But I'm (barely) young enough to suspect that worrying about dementia is the least of my concerns. By the 2030s or -40s, I think, in all seriousness, the more serious threat will be hyperintelligent machines, most likely created in part by accident (self-learning voice recognition programs or something). So, myself, I think there will be a massive increase in intelligence (almost certainly machine intelligence), but I fear it will only go a bit beyond the level of the smartest humans before it becomes an insane nihilist and destroys itself and everything around it. I really hope I'm wrong, of course!

[Fiver]

Hold on. We skipped right over the Jetsons and my flying car - which never arrived, by the way - to the AI apocalypse. Then robot our overlords went from sane to insane, found meaning in life, then rejected it really, really quickly. I knew there was a reason I don't trust Alexa.

[mike]

China is already clamping down...we may not see this scientist for awhile..

[J11]

Good one Brian4!

I think that if I can find a highly plausible existential threat that is more scary than Alzheimer's to those at risk of dementing illness, well I guess you can say that I'll chalk up today as a pretty good day. Might just have earned by angel wings or something.

What should also be considered is that research has appeared reporting that risk of Alzheimer's can be reduced merely with increased intelligence (Mendelian randomization designs). This suggests that CRISPRing or even selecting for IQ uplift will help prevent AD. A large number of IQ variants have already been reported that makes this observation increasingly actionable.

[J11]

MarcR, yes I had thought that the Cognitive Singularity was decades away, though now I am not so sure. I thought that CRISPR would be delayed for perhaps decades as endless obstructions were advanced.

Yet, this story has me rethinking what is possible. So someone without institutional or financial support was able to CRISPR babies and this was financed out of his private resources? This is for me probably the most remarkable and unexpected aspect of the story. There are essentially no great barriers for this to roll out? A $100 DNA guide? The genome sequencing would not need to be done in-house. The actual back lab needed to do this might be minimal. Even for those who welcome the dawning of a new era of smarter people there should clearly be reluctance when contemplating a world of wild west CRISPRing. What could happen to humanity?

One thing that could happen is a much nearer term IQ uplift. The 19 SNPs below if CRISPRed homozygously could enhance human intelligence by 20 points. That would be a start there are other SNPs that could be edited; one could combine this with genetic selection and extreme mate selection. What is so remarkable here is that we are now talking about existing technologies! Research published within the last few days even suggests that at some loci off targets would not occur.

rs144336753 rs117895796 rs71413877 rs150421637 rs62155350 rs34305371 rs115693355 rs75177132 rs77999825 rs114468556 rs75756843 rs75308819 rs72624911 rs111530150 rs114952970 rs11596387 rs72673097 rs1368250 rs116386746

20 years from now is rapidly feeling more like a wish to avoid the future, then to actually accept that the future has already arrived.
For those astute readers the only logic fault that I can see above is that the above SNPs are not actually likely all causal. Gene editing them might have no effect on IQ. Of course it should not be unexpected that yet additional research will reveal the causal SNPs.

[Brian4]

Fiver, what I was really hoping for was the jetpacks. No luck. Damn it.

J11, thanks.

It doesn't surprise me that variants for higher intelligence would be associated with reduced risk of Alzheimer's (likely many other pathologies as well). Much of this could well be an indirect effect: smarter people are more likely to engage in healthy behaviors. And then there's the greater amount of "cognitive reserve" (a weird notion, I've always thought: I think what that often means is not that smarter people aren't getting AD, but that it's diagnosed later: if you're born with an IQ of 170, you could have moderate AD but still score in the high 20s on the MMSE).

But since many of the cells in the brain that matter are post-mitotic, I'm not sure intelligence-linked SNPs matter much when it comes to eliminating AD: (#1) looking into this won't help those already born (at least not much [1]), and (#2) the babies one might want to engineer or select for will be born, at the earliest, in a time (2020s, say) when an outright cure for AD will be available while they're teenagers (and at that point, Alexa starts reading Nietzsche and then blows everything up anyway, so...)

Brian

[1] To be sure, it might help understand aspects of the mechanisms of the disease, a bit at least.

[J11]

Oh, I don\t know just thought someone might find this amusing.

https://www.technologyreview.com/s/6124 ... ing-sperm/

[Brian4]

J11, Someone should tell these people that some studies show epsilon-4 carriers are smarter than non-carriers (even in developed countries).

"don\t ": I'm guessing you switch between different language keyboard layouts? I sometimes forget which one I'm using and write things like that.

Brian