As slow as it may be to adopt, it's a welcome change to how people should do research. Why induce suffering on an animal when the potential for failure and or miscalculation is so high? It seems from the article a model/chip can be more insightful on how something will affect humans.
2) When testing in animals, first ensure that there is reasonable confidence the risk and severity of harm from the testing will be minimal. Only begin initial in vivo testing after this has first been achieved.
3) For test animals that are or aren't harmed, allow them to live happy, (relatively) normal, natural lives. Don't end their life early unnecessarily, with exceptions for humane euthanasia if the testing has unexpectedly caused significant, persistent suffering. Treat them as a pet owner would treat their beloved pet rat.
Some may consider it extreme, but, in my view, in an ideal world, people would treat testing in animals almost like testing in humans. The researcher should first ask "would I feel comfortable testing this in a voluntary human test subject?" Non-human animals would still be the first line of testing, since harm/death in a rat/mouse/chimpanzee is less morally bad than in a human, but I think this should be the mindset.
Most of what you said is already how it's done today, at least in the pharmaceutical industry.
> Some may consider it extreme, but, in my view, in an ideal world, people would treat testing in animals almost like testing in humans.
I don't consider it an extreme view, as lots of people share that view. But within the scope of pharmaceuticals, as opposed to cosmetics, I disagree with it because I think that's an impractical guideline to follow. Animal testing happens because there aren't good alternative ways to get safety data (on a per-treatment basis) without putting human lives at risk, and it's already a tiered system designed to balance scientific need against animal welfare. This is also why in standard medical trials proving safety comes before proving efficacy.
> The researcher should first ask "would I feel comfortable testing this in a voluntary human test subject?" Non-human animals would still be the first line of testing, since harm/death in a rat/mouse/chimpanzee is less morally bad than in a human, but I think this should be the mindset.
The problem is: how do you as a researcher become comfortable testing on a voluntary human test subject? You need safety data. How do you get the safety data? Right now, the main way is a tiered system that starts on a petri dish and eventually passes through mice, monkeys, etc. before humans.
It is a tiered system, at least if you look at big pharma. There are many stages in the pharmaceutical drug development process.
Also human testing (clinical trials) have different stages/tiers to maximize learning and minimize risk at every stage. A lot of companies trying to innovate here.
Meeting standards is nothing but a formality, a form to fill. The PhDs and lab staff doing the killing, if they have any qualms about the act itself or just the purpose of the killing, generally justify it away as being what they have to do to keep the job. The "humane" killing methods are learned on the job through trial and error, and such errors keep happening regularly to experienced staff.
Prior to lockdowns, many/most lab animals just got killed: interesting quality of life standard.
What's written in lab policies is not at all how it happens in reality.
I do feel bad for medical mice, especially things like SCID Mice https://en.wikipedia.org/wiki/Severe_combined_immunodeficien..., which are basically only born to be infected with something, and then die. Even more so in a recent case of academic fraud I've seen involving cancer testing, where the data was obviously faked, but mice still died for the data.
But yes, it's hard to test immune responses without hurting a living animal, since you'd have to somehow have a non-alive immune system to test on otherwise
FWIW, just watching a few pigs in an existing herd without interacting with them in any active way requires an extensive ethical approval process. Nobody would go through even that if they could gather the same data from a computer model.
I feel like "just a question of cost" is a heavy lifter here though.
Cost core to our resource allocation system, for better or worse, and if something costs a lot it usually means we can't do it as much. In this framing, cutting out animal testing would (often) mean medical research is slower. In this framing it's a question of where people are willing to put the dial on ethics vs results, and of course pumping the breaks in one place without universal buy-in just means the cruelty moves elsewhere.
I don't think the above 1-dimensional framing is the whole story though. You could imagine more standardized data sharing, so that the data that cost the lives of a few thousand rats could be used by others. And of course you could require (as drug trials do) that the study be registered before any experiment happened, and that the results be published regardless of the outcome. Some of these changes could actually make the science better, in addition to more ethical.
hopefully animal testing alternatives keep innovating enough that they, like some renewable energy sources today, become cheaper to make use of than their incumbents
I hope so too! Unfortunately animal studies are still one of the main pillars of physiology and drug development. They were even a requirement until very recently in the US, see the rule just approved by the FDA [1].
It’s not just a question of cost. Our biological mechanisms are so complex, there is no single system that can tell us whether a drug is 100% safe or not. That’s why drugs are eventually also “tested” in humans during clinical trials. How amazing would it be if we could throw any drug/compound into a computer, then also spit in it, and get a personalized efficacy score as a response…?
Potential for failure is high, of course, an experiment might not have the desired outcome and a death of a few hundred mice would not lead to a cure of some disease, sure.
But without trying for millions of times, scientists won’t find the one drug that works. If it saves at least person, isn’t the risk and cost of not trying to find the cure by all mean possible even higher?
Sacrificial animals are used to answer nearly all questions in medicine. I would be surprised if you could find a single therapeutic out of the last 3,000 that entered Phase I that didn't cite mouse experiments in its preclinical studies for efficacy.
The dirty little secret of human organoid models is that they are grown in a matrix (e.g. Matrigel) that is derived from a tumor that secretes this matrix, which is grown in a mouse host. The mouse of course is "sacrificed" to harvest this matrix.
IMO the most interesting research in this field is being done by DeepMind, which is to build an actual virtual cell [1]. Would be curious to know if the team / Demis is still finding adequate time to advance this field amidst the LLM boom.
[1] https://www.youtube.com/watch?v=CEOOMYxMvY4 (Note: I have heard Demis Hassabis speak about this topic in a few sources, this was the quickest one I could find where he talks in depth about it)
Very odd.... two months ago now I just bought a lazer cutter and some random acrylic off amazon to make little microfluidic devices because I thought they were neat and have some ideas to make them more generalized that i wanted to tinker with. Nice to see one pop up at me unexpectedly on an HN link.
There are ton of benefits to this too. Better for the animals but also more realistic conditions. For example XSphera https://www.xspherabio.com/ tests immunotherapy drugs against real tumors inside their actual immune system
Ah lest we forget about Jevons Paradox. With the ability to discover a nearly infinite number of potential breakthroughs, we are going to need more animal testing than ever!
What makes it more complicated is that regulatory authorities across different countries have varying views on what is considered 'acceptable'. As mentioned in the Harvard example from the article. This could mean that both animal and non animal testing will be required for the forceable future to satisfy both sets of views until there is sufficient alignment from the 'big' regulators imo.
Honestly I don’t get it. Why use a model of organs when you can use real organs in mice?
If society had to choose: cure one extra disease variant and sacrifice a million mice in the process, or don’t. Shouldn’t the answer be obviously to use mice to save people lives?
Honestly I don’t get it. How many lab rats is one person worth? Judging by some comments here, the answer would be they won’t trade mice lives for people lives?
Where do you draw the line and say that the arithmetic regarding suffering lab animals versus suffering humans does or doesn't add up?
The answer is far from obvious for many people and so obvious as to not need discussion for others. But the proportions of the population that fall into each group varies over time, from place to place, etc., as does the fraction of each group deciding for or against animal experimentation.
That whole argument is predicated on the false assumption that mice are accurate enough human analogs that you even can cure human disease with their help. There's a good reason why it's such a common comment to point out how "in mice" was omitted from a headline that otherwise sounds like a medical breakthrough.
If the simulated organs perform better for research and avoid needless animal suffering, then it's a win/win. Right now it's a win/lose at best, possibly a lose/lose if you factor in reproducibility issues using animal organs for research.
I think it’s important to realize, the mice are still dying as we speak for this purpose.
This is simply a step toward not needing to literally kill animals needlessly, if it succeeds. In the meantime, while stuff like this is really early on, we’re not going to stop animal testing.
I think for most people who are against animal testing it is not the death of the animals that is the principal problem, it is the suffering they undergo while alive.
It could and likely is the case that we can make better alternatives to animal testing but we're held back by the kinds of arguments you just made and the inertia in the system so we don't move past a flawed but functional paradigm based around animal testing.
In the next 18 months it will come out that Anthropic and/or OpenAI have killed ~1,000,000,000 mice to RL train models for drug discovery. May these model/chips be widely available and cheap enough soon!
Funny animal research story: University had an animal research committee to which anyone wanting to "use" an animal on university property had to submit proposals. Theatre department wants to "borrow" a lamb from the school's agricultural department research herd. Female character was to hold a live lamb during several stage scenes. Three shows a week for about a month. Everyone laughed. Then came the objection:
As the lamb would be exposed to uncontrolled people/pathogens it could not be used for further scientific studies nor returned to the herd. There was doubt it could even be sold for food. So, absent anyone taking it home as a pet, its appearance on stage was more of a death sentence than any scientific study!
the FDA wanted data from organs-on-a-chip or some comparable alternative.
This is the main takeaway from the article imho and great news! ~15 years ago those systems were were not very interesting for pharma companies (bayer used to have a campus a few s-bahn stops away) due to regulators requiring animal tests over those systems.
The massive increases of medical research spending to useful outcomes directly track the decrease in testing in humans and then animals. False positives from doing research in systems that are not the systems the medicine will be used in results in the observed waste and lack of progress compared to, say, the 1950s.
I am not saying we should not have stopped directly testing in humans or animals. But I am saying that the result has been worse outcomes. And any further shifts away from the biological systems that need the medicine (like using computational crap like LLMs) will further increase this disconnect.
I believe that the issue is not that we are not capable of, it's just that we don't want to do it or give priority to it.
All the while people with massive power and resources are just wasting them on gimmicks and plots to keep increasing their net worth, power, all of that.
The one who don't care have all the power, the ones who care have none.
As slow as it may be to adopt, it's a welcome change to how people should do research. Why induce suffering on an animal when the potential for failure and or miscalculation is so high? It seems from the article a model/chip can be more insightful on how something will affect humans.
It should be a tiered system, IMO.
1) Where possible, don't test in animals.
2) When testing in animals, first ensure that there is reasonable confidence the risk and severity of harm from the testing will be minimal. Only begin initial in vivo testing after this has first been achieved.
3) For test animals that are or aren't harmed, allow them to live happy, (relatively) normal, natural lives. Don't end their life early unnecessarily, with exceptions for humane euthanasia if the testing has unexpectedly caused significant, persistent suffering. Treat them as a pet owner would treat their beloved pet rat.
Some may consider it extreme, but, in my view, in an ideal world, people would treat testing in animals almost like testing in humans. The researcher should first ask "would I feel comfortable testing this in a voluntary human test subject?" Non-human animals would still be the first line of testing, since harm/death in a rat/mouse/chimpanzee is less morally bad than in a human, but I think this should be the mindset.
Most of what you said is already how it's done today, at least in the pharmaceutical industry.
> Some may consider it extreme, but, in my view, in an ideal world, people would treat testing in animals almost like testing in humans.
I don't consider it an extreme view, as lots of people share that view. But within the scope of pharmaceuticals, as opposed to cosmetics, I disagree with it because I think that's an impractical guideline to follow. Animal testing happens because there aren't good alternative ways to get safety data (on a per-treatment basis) without putting human lives at risk, and it's already a tiered system designed to balance scientific need against animal welfare. This is also why in standard medical trials proving safety comes before proving efficacy.
> The researcher should first ask "would I feel comfortable testing this in a voluntary human test subject?" Non-human animals would still be the first line of testing, since harm/death in a rat/mouse/chimpanzee is less morally bad than in a human, but I think this should be the mindset.
The problem is: how do you as a researcher become comfortable testing on a voluntary human test subject? You need safety data. How do you get the safety data? Right now, the main way is a tiered system that starts on a petri dish and eventually passes through mice, monkeys, etc. before humans.
> It should be a tiered system, IMO.
It is a tiered system, at least if you look at big pharma. There are many stages in the pharmaceutical drug development process.
Also human testing (clinical trials) have different stages/tiers to maximize learning and minimize risk at every stage. A lot of companies trying to innovate here.
I’m exited what more AI can bring to this space.
this is already what they do, you have to justify usage of animals and meet standards for quality of life to do any research at a real institution
also animal research is a huge pain in the ass, nobody is doing it unless they have to
That's a naive understanding of academia.
Meeting standards is nothing but a formality, a form to fill. The PhDs and lab staff doing the killing, if they have any qualms about the act itself or just the purpose of the killing, generally justify it away as being what they have to do to keep the job. The "humane" killing methods are learned on the job through trial and error, and such errors keep happening regularly to experienced staff.
Prior to lockdowns, many/most lab animals just got killed: interesting quality of life standard.
What's written in lab policies is not at all how it happens in reality.
I do feel bad for medical mice, especially things like SCID Mice https://en.wikipedia.org/wiki/Severe_combined_immunodeficien..., which are basically only born to be infected with something, and then die. Even more so in a recent case of academic fraud I've seen involving cancer testing, where the data was obviously faked, but mice still died for the data.
But yes, it's hard to test immune responses without hurting a living animal, since you'd have to somehow have a non-alive immune system to test on otherwise
That's already how it works.
FWIW, just watching a few pigs in an existing herd without interacting with them in any active way requires an extensive ethical approval process. Nobody would go through even that if they could gather the same data from a computer model.
I agree. If it’s a question of just cost - (animals being cheaper to just test on quickly and faster) animal testing shouldn’t be allowed.
I feel like "just a question of cost" is a heavy lifter here though.
Cost core to our resource allocation system, for better or worse, and if something costs a lot it usually means we can't do it as much. In this framing, cutting out animal testing would (often) mean medical research is slower. In this framing it's a question of where people are willing to put the dial on ethics vs results, and of course pumping the breaks in one place without universal buy-in just means the cruelty moves elsewhere.
I don't think the above 1-dimensional framing is the whole story though. You could imagine more standardized data sharing, so that the data that cost the lives of a few thousand rats could be used by others. And of course you could require (as drug trials do) that the study be registered before any experiment happened, and that the results be published regardless of the outcome. Some of these changes could actually make the science better, in addition to more ethical.
hopefully animal testing alternatives keep innovating enough that they, like some renewable energy sources today, become cheaper to make use of than their incumbents
I hope so too! Unfortunately animal studies are still one of the main pillars of physiology and drug development. They were even a requirement until very recently in the US, see the rule just approved by the FDA [1].
[1]: https://public-inspection.federalregister.gov/2026-19350.pdf
It’s not just a question of cost. Our biological mechanisms are so complex, there is no single system that can tell us whether a drug is 100% safe or not. That’s why drugs are eventually also “tested” in humans during clinical trials. How amazing would it be if we could throw any drug/compound into a computer, then also spit in it, and get a personalized efficacy score as a response…?
>Why induce suffering on an animal when the potential for failure and or miscalculation is so high?
Because that's drug testing in general, even with human subjects.
Potential for failure is high, of course, an experiment might not have the desired outcome and a death of a few hundred mice would not lead to a cure of some disease, sure. But without trying for millions of times, scientists won’t find the one drug that works. If it saves at least person, isn’t the risk and cost of not trying to find the cure by all mean possible even higher?
> But without trying for millions of times, scientists won’t find the one drug that works.
well isn't it mostly other way around? millions of times to LOOK FOR hidden SIDE-EFFECTS?
animal testing is mostly for answering 'is it safe?'
Sacrificial animals are used to answer nearly all questions in medicine. I would be surprised if you could find a single therapeutic out of the last 3,000 that entered Phase I that didn't cite mouse experiments in its preclinical studies for efficacy.
The dirty little secret of human organoid models is that they are grown in a matrix (e.g. Matrigel) that is derived from a tumor that secretes this matrix, which is grown in a mouse host. The mouse of course is "sacrificed" to harvest this matrix.
Is it a dirty secret? It's like saying the people running the ESG funds drive gas powered cars.
IMO the most interesting research in this field is being done by DeepMind, which is to build an actual virtual cell [1]. Would be curious to know if the team / Demis is still finding adequate time to advance this field amidst the LLM boom.
[1] https://www.youtube.com/watch?v=CEOOMYxMvY4 (Note: I have heard Demis Hassabis speak about this topic in a few sources, this was the quickest one I could find where he talks in depth about it)
Very odd.... two months ago now I just bought a lazer cutter and some random acrylic off amazon to make little microfluidic devices because I thought they were neat and have some ideas to make them more generalized that i wanted to tinker with. Nice to see one pop up at me unexpectedly on an HN link.
There are ton of benefits to this too. Better for the animals but also more realistic conditions. For example XSphera https://www.xspherabio.com/ tests immunotherapy drugs against real tumors inside their actual immune system
Ah lest we forget about Jevons Paradox. With the ability to discover a nearly infinite number of potential breakthroughs, we are going to need more animal testing than ever!
(Sarcasm but also some potential truth)
What makes it more complicated is that regulatory authorities across different countries have varying views on what is considered 'acceptable'. As mentioned in the Harvard example from the article. This could mean that both animal and non animal testing will be required for the forceable future to satisfy both sets of views until there is sufficient alignment from the 'big' regulators imo.
Honestly I don’t get it. Why use a model of organs when you can use real organs in mice? If society had to choose: cure one extra disease variant and sacrifice a million mice in the process, or don’t. Shouldn’t the answer be obviously to use mice to save people lives? Honestly I don’t get it. How many lab rats is one person worth? Judging by some comments here, the answer would be they won’t trade mice lives for people lives?
Where do you draw the line and say that the arithmetic regarding suffering lab animals versus suffering humans does or doesn't add up?
The answer is far from obvious for many people and so obvious as to not need discussion for others. But the proportions of the population that fall into each group varies over time, from place to place, etc., as does the fraction of each group deciding for or against animal experimentation.
That whole argument is predicated on the false assumption that mice are accurate enough human analogs that you even can cure human disease with their help. There's a good reason why it's such a common comment to point out how "in mice" was omitted from a headline that otherwise sounds like a medical breakthrough.
If the simulated organs perform better for research and avoid needless animal suffering, then it's a win/win. Right now it's a win/lose at best, possibly a lose/lose if you factor in reproducibility issues using animal organs for research.
I think it’s important to realize, the mice are still dying as we speak for this purpose.
This is simply a step toward not needing to literally kill animals needlessly, if it succeeds. In the meantime, while stuff like this is really early on, we’re not going to stop animal testing.
I think for most people who are against animal testing it is not the death of the animals that is the principal problem, it is the suffering they undergo while alive.
> How many lab rats is one person worth?
Zero, if there are alternatives.
If the alternatives are as fast and cheap as lab rat testing, sure.
If not, the alternatives will cost human lives and suffering.
But animal testing could be a local maxima.
It could and likely is the case that we can make better alternatives to animal testing but we're held back by the kinds of arguments you just made and the inertia in the system so we don't move past a flawed but functional paradigm based around animal testing.
In the next 18 months it will come out that Anthropic and/or OpenAI have killed ~1,000,000,000 mice to RL train models for drug discovery. May these model/chips be widely available and cheap enough soon!
Exactly 49 humans.
Funny animal research story: University had an animal research committee to which anyone wanting to "use" an animal on university property had to submit proposals. Theatre department wants to "borrow" a lamb from the school's agricultural department research herd. Female character was to hold a live lamb during several stage scenes. Three shows a week for about a month. Everyone laughed. Then came the objection:
As the lamb would be exposed to uncontrolled people/pathogens it could not be used for further scientific studies nor returned to the herd. There was doubt it could even be sold for food. So, absent anyone taking it home as a pet, its appearance on stage was more of a death sentence than any scientific study!
They opted for a doll.
Nice to see TissUse getting a shout out :)
This is the main takeaway from the article imho and great news! ~15 years ago those systems were were not very interesting for pharma companies (bayer used to have a campus a few s-bahn stops away) due to regulators requiring animal tests over those systems.It is perfectly acceptable for dogs to hunt and brutally kill rabbits and other animals (even pets). I wouls ban hunting first!
The massive increases of medical research spending to useful outcomes directly track the decrease in testing in humans and then animals. False positives from doing research in systems that are not the systems the medicine will be used in results in the observed waste and lack of progress compared to, say, the 1950s.
I am not saying we should not have stopped directly testing in humans or animals. But I am saying that the result has been worse outcomes. And any further shifts away from the biological systems that need the medicine (like using computational crap like LLMs) will further increase this disconnect.
“It’s people man! We’re the ones who are miserable, so why shouldn’t we get to try it all first?!”
- Sealab 2021
This was already figured out and built out years ago https://www.darpa.mil/research/programs/microphysiological-s...
better link https://www.darpa.mil/news/features/microphysiological-syste...
I believe that the issue is not that we are not capable of, it's just that we don't want to do it or give priority to it.
All the while people with massive power and resources are just wasting them on gimmicks and plots to keep increasing their net worth, power, all of that.
The one who don't care have all the power, the ones who care have none.