frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Artificial Womb Technology: Survey of 750 Researchers

3•biostark•2h ago
Hey HN, thought some of you might want to AMA, here's the context:

I wanted to know if artificial wombs would happen in my lifetime. I couldn't find a straight answer, so I started emailing researchers.

I’ve seen photos of lambs floating in plastic bags crop up constantly in these discussions. They're from Children's Hospital of Philadelphia, it’s not an artificial womb. CHOP's device keeps the lamb submerged in fluid so its fragile lungs aren't exposed to air, and gives oxygen to its blood using the same machines we use for adults with lung failure. This works for a 22-week fetus, which mostly just needs blood and nutrients, but it doesn't give an embryo somewhere to implant and develop from scratch.

I looked into what it would take to support an embryo earlier than that, and it turns out the mother does a lot we don't know how to replicate in a lab. I emailed someone at CHOP and they confirmed as much, a few of their colleagues called it a 'pipe dream' in a 2023 paper. I figured I'd just keep emailing researchers.

I went looking for real forecasts on getting this to work end-to-end. Metaculus predicts ~2044, but the comments cite similar articles about embryo research. The most credible-looking was a 2021 NYT piece about mouse embryos developing for ten days, so I emailed Dr. Hanna's lab. What they're doing is growing embryos in a nutrient bath: the nutrients soak in from the outside, but only so deep, so the cells inside eventually starve. I asked whether we’d need an artificial placenta to solve this problem, but they said they're sticking with the nutrient bath method since it's good for making embryos to study.

Matt Krisiloff mentioned on Twitter that at least four startups were working on this, so I kept emailing around. Haven't heard of anything interesting yet, but a researcher at CHOP told me some of them are literally repackaging incubator tech and presenting it as the real thing.

I’m running a fact-finding mission now, since information on this topic is weirdly hard to find. I mapped out some of the problems we’d need to solve for artificial wombs to work and recovered contact details from people working on research that’s relevant. I ended up with around 750 contacts including embryologists, cultivated meat people who work on keeping tissue alive outside of the body, bioprinting researchers, that sort of thing.

I’ve had a few of my own questions answered in this way, so I figured I'd open it up to you. If there's something you've always wondered about, even if it feels basic, I'm happy to look into it.

What would you want to ask?

Comments

haebom•1h ago
I'm not sure, but I hope they make it quickly. Personally, I think this is more revolutionary than going to space.
medinot•23m ago
What approaches could you use to engineer a synthetic placenta?

From what I understand of current research, the bottleneck right now is finding a way to supply an embryo with oxygen and nutrients, and remove waste products, that doesn't rely on the the use of a 'nutrient bath'. This method relies on diffusion. Diffusion is not sufficient to supply and remove the relevant substances as it grows. The embryo literally starts dying from the inside out as it's too far from the nutrient bath.

You need to involve the circulatory system - this is the literal problem it solves. How you do that however I do not know.

4B If Statements

https://andreasjhkarlsson.github.io//jekyll/update/2023/12/27/4-billion-if-statements.html
1•damethos•1m ago•0 comments

Muen – An x86/64 Separation Kernel for High Assurance

https://muen.codelabs.ch/
1•LiamPowell•2m ago•1 comments

$100k Robot Dogs That Look Like Billionaires? The Artist Explains

https://www.wsj.com/arts-culture/fine-art/beeple-robot-dogs-regular-animals-art-basel-nft-84043fb9
1•cs702•3m ago•0 comments

The Missing Device That Changes AI Forever – The Story of the Memristor [video]

https://www.youtube.com/watch?v=AHZxhdjCBSo
1•lucavice•5m ago•0 comments

The Silent Leak: How One Line of Go Drained Memory Across Goroutines

https://medium.com/harness-engineering/the-silent-leak-how-one-line-of-go-drained-memory-across-t...
2•gearnode•8m ago•0 comments

Treating Documentation as a Product

https://compositecode.blog/2025/12/03/treating-documentation-as-a-product/
1•Kerrick•9m ago•0 comments

Dolt Is as Fast as MySQL on Sysbench

https://www.dolthub.com/blog/2025-12-04-dolt-is-as-fast-as-mysql/
1•thunderbong•9m ago•0 comments

What Is Modal?

https://modal.cx/blog/what-is-modal/
1•todsacerdoti•9m ago•0 comments

OpenAI has trained its LLM to confess to bad behavior

https://www.technologyreview.com/2025/12/03/1128740/openai-has-trained-its-llm-to-confess-to-bad-...
1•cainxinth•13m ago•1 comments

The Cult of Therapy

https://www.profgalloway.com/the-cult-of-therapy/
1•Brajeshwar•14m ago•0 comments

A New Governing Ecosystem Is Evolving

https://www.noemamag.com/a-new-governing-ecosystem-is-evolving/
1•Brajeshwar•14m ago•0 comments

Growing Number of Satellites Will Leave Streaks on Photos from Space Telescopes

https://e360.yale.edu/digest/satellites-orbiting-telescopes-photos
2•Brajeshwar•14m ago•1 comments

What imitating an iconic robot reveals on vocal imitation in parrots&starlings

https://www.nature.com/articles/s41598-025-23444-7
1•bookofjoe•16m ago•0 comments

Carv: Whoop for Skiing

https://getcarv.com
1•danielfalbo•16m ago•0 comments

AI Slop Is Ruining Reddit for Everyone

https://www.wired.com/story/ai-slop-is-ruining-reddit-for-everyone/
4•thm•17m ago•2 comments

Robot Dog Billionaires Take Photos and Poop Them Out

https://petapixel.com/2025/12/05/beeple-robot-dogs-photos-elon-musk-mark-zuckerberg-miami-art-basel/
3•b0ner_t0ner•19m ago•1 comments

Every class I went to during the first week of fall

https://mitadmissions.org/blogs/entry/every-class-i-went-to-during-the-first-week-of-fall/
1•samuel246•26m ago•0 comments

Fast Median Filter over arbitrary datatypes

https://martianlantern.github.io/2025/09/median-filter-over-arbitrary-datatypes/
1•martianlantern•28m ago•0 comments

Artificial intelligence research has a slop problem

https://www.theguardian.com/technology/2025/dec/06/ai-research-papers
1•n1b0m•28m ago•1 comments

Another AI slop story: ChatGPT vs. Human

https://joshua.hu/ai-slop-story-nginx-leaking-dns-chatgpt
2•detaro•29m ago•1 comments

Show HN: Built an all-in-one feedback board for SaaS apps – Ship features

2•rahulbstomar•30m ago•2 comments

What if our ancestors didn't feel pain the way we do

https://www.theatlantic.com/magazine/2026/01/human-ancestors-emotion-history/684959/
2•HR01•31m ago•0 comments

Dependent Names with a Little Encouragement

https://consteval.ca/2025/09/27/dependent-names/
1•HeliumHydride•32m ago•0 comments

HTML as an Accessible Format for Papers

https://info.arxiv.org/about/accessible_HTML.html
4•el3ctron•36m ago•1 comments

AoCO 2025: Division

https://xania.org/202512/06-dividing-to-conquer
3•HeliumHydride•40m ago•0 comments

Carlo is no longer maintained

https://github.com/GoogleChromeLabs/carlo
3•keepamovin•41m ago•0 comments

Nemawashi: A Powerful, Forgotten Japanese Innovation and Change Management Tool

https://www.changebase.app/blog/nemawashi-japanese-change-management-tool
3•mooreds•42m ago•0 comments

A 2kb library that makes WebWorkers enjoyable

https://github.com/GoogleChromeLabs/comlink
3•keepamovin•42m ago•0 comments

Make images smaller using best-in-class codecs, right in the browser

https://github.com/GoogleChromeLabs/squoosh
2•keepamovin•43m ago•0 comments

Measuring AI impact like it's 1995

https://www.swarmia.com/blog/measuring-ai-impact-like-1995/
2•mooreds•44m ago•0 comments