Hacker News Reader

Comment by derefr | Hacker News Reader

That mountain of evidence mostly says that the in vivo prenatal environment is kind of... awful, though. As in, mother and foetus competing for nutrients with better maternal genes resulting in better defenses against the placental "parasite", and thus worse developmental outcomes; and better foetal genes resulting in better parasitic strategies to penetrate the uterine wall, and thus worse perinatal health outcomes for the mother.

I'd honestly stake >$10k, at this point, on the hypothesis that "growing children from embroyos in vitro, hooked up to an unlimited nutrient drip" will result in children much healthier than "regular" children, along every axis we can think of. (Presuming, of course, that we first figure out every micro-nutrient such a drip would be required to provide.)

View on HN

What would happen though is eventually humans would lose the ability to naturally produce children.

I've seen this in a bunch of movies but it just occurred to me that this may eventually happen to humanity. Birthing children is dangerous for the mother and child, and if given the choice many women may decide to have one "grown" instead.

That's a situation we already face. There's a line of humans who can only reproduce with IVF.

That's not a situation "we" as a culture/civilization/species face, it's a situation they face. The rest of us are perfectly capable of doing what our bodies are designed to do without doctors and scientists and other forms of assistance.

IVF babies aren't guaranteed to be infertile, they just have a slightly higher chance of it (because of selection bias in the parents).

My wife and I might well have babies via IVF because she can't get pregnant while on chemotherapy - but that doesn't mean we carry any infertility genes. She just had the bad luck to have some cells mutate into cancer.

What is "naturally?" At some point in technological progress, we need to build machines to have children. But go a bit further along the line, and maybe we'd find the human body automatically building the machine that builds the human body—in effect, laying eggs.

> At some point in technological progress, we need to build machines to have children

Why?

Er, sorry, I just meant that as "at some [early] point along the timeline of the possible developments in biomedical technology, the world would look like one where—at least early adopters—would need a machine to make children." I didn't mean to imply that there's some sort of imperative to go toward that point.

To outbreed the bad guys.

There will be enough poor people that can't afford to have their children grown who will still be able to get pregnant.

But the continuation of the human "species" as a single entity would then assume/require class miscegenation to re-constrain the DNA from the in vitro-grown humans every few generations.

Otherwise, you'd now have two diverging human species, who might eventually not even be able to interbreed.

> Otherwise, you'd now have two diverging human species, who might eventually not even be able to interbreed.

It's a possible outcome in a few centuries. We might get one human species that would call themselves robot-overlords, those would be born in artificial wombs have artificially enhanced brains and have very prominent positions on Earth. The plebs would sleep on the stairs and drink their soylent green.

In any reasonably competent universe the costs would head towards zero.

I'll bet against you. In fact, you inadvertently bring up the very undoing of your hypothetical bet when you say...

>presuming, of course, that we first figure out every micro-nutrient such a drip would be required to provide.

It's important to understand that these micro-nutrients and other chemical signalling-agents vary through time, and in response to such things as social environment, bodily movement of the mother, sensory input (sight, hearing, olfaction, the haptic senses, etc), occasional punctual stress, pleasure etc.

You're ignoring a huge source of complexity which is the systematic variation of these signals in response to the external environment.

Temporal dynamics of signaling pathways matter.

That's simply not the kind of thing we're even remotely close to being able to simulate in a synthetic system. Hell it's not even something we're remotely close to understanding.

What kind of infrastructure will this require? You'll need researchers to learn how such an object can be feasible, engineers to design it, people to construct it, parts must be ordered (some of which you might not even have the capability to make yourself), raw materials to make the parts must be mined, logistics must be done to make sure all of these parts are in the right place at the right time, enterprises must be constructed around building, selling, and marketing these devices.

What kind of power dynamics does that create? It involves the work of hundreds of thousands of people across every sort of country spanning the entire globe, many of whom hold a vast amount of power over you and the rest of whom have no obligation except financial to cooperate with you.

What exactly do we gain from giving so many people such vast power over us?

Reply on news.ycombinator.com