This is changing at an astonishing rate, though. The development of GWAS combined with huge genetic datasets and associated IQ test information has opened the floodgates to finding more and more genes that predict the variation in IQ. Of course, it's still an open question whether these have other effects, and whether their effects are additive (linear), but the field does seem to be making extremely rapid progress...
'gwern has put together some links on genetic associations with IQ:
ASOIF is fiction. We live in reality. The problem is far more often that we're held by obsolete and suffocating rules than that we're moving too fast. I'd rather try new things than hunker down in fear of the unknown.
I think GRRM very carefully made by both the positive and negative parts key to the framing of Qyburn; he's quite clearly both a brilliant scientist who rebelled against obsolete and conventional ideas which did not work, and, a monster who is utterly indifferent (or so nearly so as makes no difference in practice) to the suffering inflicted on the pursuit of his research, either directly or in his efforts to maintain the support he needs to continue his work.
GRRM tends not to make unidimensionally good or evil characters in ASOIAF (there's a few that seem that way on the evil side, but they are mostly also ones we don't know as much about.)
We are all utterly innocent and unwilling participants in being born. You can either take the random cards that nature deals you, or we can try to act to have better outcomes. Just because you refuse to take action doesn't absolve you from taking a suboptimal action.
It's actually very saddening to watch a completely avoidable & preventable problem, whose N-amount of obvious solutions end up being shoved under the rug because we refuse to draw the line.
Very few people are honest with themselves and admit the end goal is not making better, its making more controllable/more in aim with how the rulers want their population to be.
"I detest the precautionary principle."
Sometimes, a little bit of caution is what keeps someone alive long enough to reproduce. Plenty of Darwin Award winners could have used a little more caution.
Even if it isn't down-right engineering/splicing/etc, and using only the parents' egg/sperm combinations, they can still offer to "move our society" forward at a snail pace.
I personally see no reason to be against it, as it's the logical extension of natural evolution. It's just more refined, scientific, guided, and devoid of some of the randomness or trial-by-error properties that are inherent with natural evolution.
This kind of proof is of course impossible, and in practice it means "never do anything".
The fun thing about is that people only advocate applying it to things they don't want to do, never on their own favorite projects
So it happens that genetics doesn't have to catch up - we can have this knowledge without knowing it.
And also. People will question genetists' judgement. Some of the changes they're proposing have never passed statistically significant testing. But nobody ever questions neural network.
> I'm pretty sure that if we threw a lot of genomes, healthy and defective, to a neural network, and then showed it a genome and told "make it healthy", it would.
It's not quite that simple.
In order to use ML on actual organisms, you'd need a way to try a lot of different modifications and then evaluate them after actual growth.
Doing that kind of experimentation with human embryos, fetuses, and beyond has huge ethical minefields as well as many biological engineering challenges.
You're optimizing in a space of millions of discrete dimensions (one for each base pair), with little knowledge about independencies. This is in contrast to tasks like image recognition, where we can make use of the spacial structure of the pixels to build effective models like convnets.
Additionally, medical datasets, especially ones with genetic data, rarely contain more than a few hundred datapoints, which is not where you want to be for deep learning.
Machine learning is still useful in this area, especially to find genes or gene combinations with high correlation to certain diseases. But we are very far away from having a model that maps genome -> healthiness.