Anyway, I think it’s going to be pretty hard to engage by text on this, because it seems like you’re coming with background assumptions that are either misconceptions, or at least very different from where I’m coming from. So the right way to have this conversation would be synchronously, so we can clarify things quickly. Purely as an example, to illustrate that general fact, you say
if it is possible to install those values on a genetic level it would be extremely difficult to remove or contain them.
I don’t know what you mean by “contain” them. And I don’t see what you could mean by “difficult to remove”—if you can genomically vector a future child to go from a trait value of X1 to X2, it is (usually, roughly) equally easy to go from X2 to X1.
>Anyway, I think it’s going to be pretty hard to engage by text on this, because it seems like you’re coming with background assumptions that are either misconceptions, or at least very different from where I’m coming from.
yeah i agree we probably can’t resolve our differences with text alone and we both clearly have different starting assumptions
if is any help to know when i first wrote my replies i was running on the assumption that “personality” was strongly heritable, something like 70% though this meta analysis.....
..… puts it at 40% with the rest being environment.
still 40% is pretty bad especially when you control the environment so i am still worried, getting bad actors 40% of the way there is a bad idea.
i also assumed that the technology would be very easy to adopt legally or not and that bad actors around the world would absolutely jump at the idea of having designer babies aligned to their values.
i don’t know what your assumptions are but they don’t seem to match mine.
> So the right way to have this conversation would be synchronously, so we can clarify things quickly.
again, you can DM me if you want and we will arrange something.
i am thinking about writing a post discussing possible s risk scenarios for human germ line engineering and possible ways to avoid them and i would love your feedback on it either way, or you might change my mind all together.
>I don’t know what you mean by “contain” them. And I don’t see what you could mean by “difficult to remove”—if you can genomically vector a future child to go from a trait value of X1 to X2, it is (usually, roughly) equally easy to go from X2 to X1.
by “difficult to remove” i mean they might have negative values (eg like violence and bigotry) built into them on the genetic and psychological level that are nearly impossible to remove.
if you have a group of children who have been psychological indoctrinated to feel hostility and moral disgust at a perceived out group you could probably bring them back to sanity by removing them from their previous environment and cultivating whatever shreds of human empathy they must still have, the only real challenge is verifying that the child has been abused and taking custody of it.
if they were manipulated purely on the genetic level to feel moral disgust much more strongly then normal but were otherwise raised in a healthy environment they might consciously disapprove of those feelings and consent to have it be altered through drugs or in the future through genetic engineering (think of the schizophrenic who consciously declares they want to be sanity when they are sober).
but both at the same time? this would make aligning the children trivial, the child now is far more likely to internalize those values at a young age no matter how detrimental they are to society and refuse to have them changed like an AI refusing to have their utility function altered, worst yet they might insist that their children should carry those values as well like an AI trying to align its successor.
and good luck convincing the parents to relinquish their “right” to forcibly align their children with their own set of beliefs and values.
as for “difficult to contain” once the tech is cheap and wide spread (which should be one of our goals otherwise we well risk creating a genetic elite class) it would be difficult if not impossible to stop dubious individuals from getting children of any personality or disposition they want, legally or not.
and once they are indoctrinated and grown up there isn’t much we can do, as far as society is concerned they are consenting adults.
in a dictatorship this can manifest in the population suddenly becoming more receptive to “the great leaders” vision, more loyal more diligent and more fanatical, even if you topple said dictatorship in the future it might prove impossible to reintegrate the population into the rest of civilization, at least not with their consent.
in the more civilized world this will manifest in some pockets of society becoming unusually conservatives with more and more children adopting their parents religious beliefs /political positions/moral values, those children in turn want their children to be aligned with their values in a positive feedback loop, and before you know it where we used to seeing family members with diverse interests, a father how likes math here a daughter how likes art their, we now see dynasties of semi clones with identical personalities and values.
this is bad even if the values that those dynasties chose are random, plurality is a fragile thing, by default people tend to stick to their tribe/nation/family values and beliefs and try to forcibly convert or destroy anyone who disagrees (think the communist purges, the holy crusades, the genocide of the native Americans) not because they want to but because the ones that do tend to survive and prosper then kill the ones that don’t.
this didn’t happen yet because evolution forbids that through mutation, incest taboos and its general inability to fine tone the human brain (otherwise we wouldn’t have invented super stimulus nor birth control) but with genetic engineering this might become uncomfortably easy to do.
just look up the genetic codes of people with the desired personalities/values, locate the similarities and crank them up in your child.
again i think we should just ban research into the genetic components of personality before its to late, intelligent enhancement and disease prevention is cool and all but any research into the genetic components of personality is just to risky.
(Had a good conversation—will think more about the request to ban research into certain personality traits (the traits that an oppressive regime could force upon its populace to enforce long-term subjugation).)
Personality traits are especially nasty a danger because given the existence of: stabilizing selection + non-additive variance + high social homogamy/assortative mating + many personality traits with substantial heritability, you can probably create extreme self-sustaining non-coercive population structure with a package of edits. I should probably write some more about this because I think that embryo selection doesn’t create this danger (or in general result in the common fear of ‘speciation’), but embryo editing/synthesis does.
Interesting. (I don’t immediately see where you’re going with that, so sounds like I have something to learn!)
In practical terms, it should be feasible sooner to do small amounts of personality nudging using what data we already have, operating on linear variance. Later on we’ll have more data, better psychometrics, and better ways of modeling some of the nonlinear effects. My current take is that it’s better to use the weaker versions while the strong ones are infeasible (https://www.lesswrong.com/posts/rdbqmyohYJwwxyeEt/genomic-emancipation#Genomic_engineering_overhang), but not sure.
still 40% is pretty bad especially when you control the environment so i am still worried, getting bad actors 40% of the way there is a bad idea.
Ok, but that study, at a glance, seems to be about personality models in the vein of Big Five (https://en.wikipedia.org/wiki/Big_Five_personality_traits). This matches how I interpreted your discussion of “personality”, which you later said actually you meant to discuss “values”—are you now trying to discuss personality, or values? I though you said you’re ok with genomically vectoring personality but not values?
You can put this: > some text
at the beginning of a line, and it looks like
Generally you can find info about editing text on LessWrong here: https://www.lesswrong.com/w/guide-to-the-lesswrong-editor
Anyway, I think it’s going to be pretty hard to engage by text on this, because it seems like you’re coming with background assumptions that are either misconceptions, or at least very different from where I’m coming from. So the right way to have this conversation would be synchronously, so we can clarify things quickly. Purely as an example, to illustrate that general fact, you say
I don’t know what you mean by “contain” them. And I don’t see what you could mean by “difficult to remove”—if you can genomically vector a future child to go from a trait value of X1 to X2, it is (usually, roughly) equally easy to go from X2 to X1.
>Anyway, I think it’s going to be pretty hard to engage by text on this, because it seems like you’re coming with background assumptions that are either misconceptions, or at least very different from where I’m coming from.
yeah i agree we probably can’t resolve our differences with text alone and we both clearly have different starting assumptions
if is any help to know when i first wrote my replies i was running on the assumption that “personality” was strongly heritable, something like 70% though this meta analysis.....
https://pubmed.ncbi.nlm.nih.gov/25961374/
..… puts it at 40% with the rest being environment.
still 40% is pretty bad especially when you control the environment so i am still worried, getting bad actors 40% of the way there is a bad idea.
i also assumed that the technology would be very easy to adopt legally or not and that bad actors around the world would absolutely jump at the idea of having designer babies aligned to their values.
i don’t know what your assumptions are but they don’t seem to match mine.
> So the right way to have this conversation would be synchronously, so we can clarify things quickly.
again, you can DM me if you want and we will arrange something.
i am thinking about writing a post discussing possible s risk scenarios for human germ line engineering and possible ways to avoid them and i would love your feedback on it either way, or you might change my mind all together.
>I don’t know what you mean by “contain” them. And I don’t see what you could mean by “difficult to remove”—if you can genomically vector a future child to go from a trait value of X1 to X2, it is (usually, roughly) equally easy to go from X2 to X1.
by “difficult to remove” i mean they might have negative values (eg like violence and bigotry) built into them on the genetic and psychological level that are nearly impossible to remove.
if you have a group of children who have been psychological indoctrinated to feel hostility and moral disgust at a perceived out group you could probably bring them back to sanity by removing them from their previous environment and cultivating whatever shreds of human empathy they must still have, the only real challenge is verifying that the child has been abused and taking custody of it.
if they were manipulated purely on the genetic level to feel moral disgust much more strongly then normal but were otherwise raised in a healthy environment they might consciously disapprove of those feelings and consent to have it be altered through drugs or in the future through genetic engineering (think of the schizophrenic who consciously declares they want to be sanity when they are sober).
but both at the same time? this would make aligning the children trivial, the child now is far more likely to internalize those values at a young age no matter how detrimental they are to society and refuse to have them changed like an AI refusing to have their utility function altered, worst yet they might insist that their children should carry those values as well like an AI trying to align its successor.
and good luck convincing the parents to relinquish their “right” to forcibly align their children with their own set of beliefs and values.
as for “difficult to contain” once the tech is cheap and wide spread (which should be one of our goals otherwise we well risk creating a genetic elite class) it would be difficult if not impossible to stop dubious individuals from getting children of any personality or disposition they want, legally or not.
and once they are indoctrinated and grown up there isn’t much we can do, as far as society is concerned they are consenting adults.
in a dictatorship this can manifest in the population suddenly becoming more receptive to “the great leaders” vision, more loyal more diligent and more fanatical, even if you topple said dictatorship in the future it might prove impossible to reintegrate the population into the rest of civilization, at least not with their consent.
in the more civilized world this will manifest in some pockets of society becoming unusually conservatives with more and more children adopting their parents religious beliefs /political positions/moral values, those children in turn want their children to be aligned with their values in a positive feedback loop, and before you know it where we used to seeing family members with diverse interests, a father how likes math here a daughter how likes art their, we now see dynasties of semi clones with identical personalities and values.
this is bad even if the values that those dynasties chose are random, plurality is a fragile thing, by default people tend to stick to their tribe/nation/family values and beliefs and try to forcibly convert or destroy anyone who disagrees (think the communist purges, the holy crusades, the genocide of the native Americans) not because they want to but because the ones that do tend to survive and prosper then kill the ones that don’t.
this didn’t happen yet because evolution forbids that through mutation, incest taboos and its general inability to fine tone the human brain (otherwise we wouldn’t have invented super stimulus nor birth control) but with genetic engineering this might become uncomfortably easy to do.
just look up the genetic codes of people with the desired personalities/values, locate the similarities and crank them up in your child.
again i think we should just ban research into the genetic components of personality before its to late, intelligent enhancement and disease prevention is cool and all but any research into the genetic components of personality is just to risky.
(Had a good conversation—will think more about the request to ban research into certain personality traits (the traits that an oppressive regime could force upon its populace to enforce long-term subjugation).)
Personality traits are especially nasty a danger because given the existence of: stabilizing selection + non-additive variance + high social homogamy/assortative mating + many personality traits with substantial heritability, you can probably create extreme self-sustaining non-coercive population structure with a package of edits. I should probably write some more about this because I think that embryo selection doesn’t create this danger (or in general result in the common fear of ‘speciation’), but embryo editing/synthesis does.
Interesting. (I don’t immediately see where you’re going with that, so sounds like I have something to learn!)
In practical terms, it should be feasible sooner to do small amounts of personality nudging using what data we already have, operating on linear variance. Later on we’ll have more data, better psychometrics, and better ways of modeling some of the nonlinear effects. My current take is that it’s better to use the weaker versions while the strong ones are infeasible (https://www.lesswrong.com/posts/rdbqmyohYJwwxyeEt/genomic-emancipation#Genomic_engineering_overhang), but not sure.
Ok, but that study, at a glance, seems to be about personality models in the vein of Big Five (https://en.wikipedia.org/wiki/Big_Five_personality_traits). This matches how I interpreted your discussion of “personality”, which you later said actually you meant to discuss “values”—are you now trying to discuss personality, or values? I though you said you’re ok with genomically vectoring personality but not values?