Hacker News Reader

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.

> The problem is: how do you as a researcher become comfortable testing on a voluntary human test subject?

I don't think you and GP are actually saying anything different. I interpreted their statement 'The researcher should first ask "would I feel comfortable testing this in a voluntary human test subject?"' not to mean don't test on animals if you wouldn't on a human, but to use your reaction and feelings to that to guide your judgement, and help judge what type of animal testing you're at, not to be a binary switch of whether any animal testing at all should be done.

> 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.

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

How do you feel for wild mice, who typically have short lives of borderline starvation before they are eaten by a snake, bird, or other predator?

Should hunter gatherers be thankful if we kidnapped them and tortured them to death in our labs?

What about people from Gaza?

Is it acceptable to do human testing as long as you pull the humans out of say east Africa during a famine?

These examples don't work. Hunter gatherers didn't suffer, and the other two are in outlier bad situations. Set them free in a stable situation and they'll live happy lives.

If that description of wild mice is accurate, there is no happy free life. The baseline is very different. I'm not going to try to say how that should affect the rest of the ethics, but it's an important point to consider, and not as easily dismissed as your examples would imply.

> Hunter gatherers didn't suffer

Of course they did and still do. They have far higher rates of infant and child mortality, death from curable disease, starvation, and violent death.

It’s inconsistent to call the natural condition of wild animals, surviving and dying in nature, "no happy free life," then exempt humans living under that same fundamental condition.

How do we draw the line, from which principle that isn’t ex post facto? Do no wild animals have a happy free life? Do all humans? Most? A minority? Across all of human history?

Even in the most cherry-picked minority of cases, someone born into a wealthy developed country, people are frequently born with or develop agonizing diseases. They die of cancer and dementia despite their wealth. They live in grinding poverty, or barely cling to existence through depression, anxiety and alienation. Many suffer so greatly in fact that they commit suicide.

Suffering isn't a state of existence reserved for animals that humans are exempt from except in rare cases.

So if it is not, then is the fact that people suffer tremendously just as a normal fact of life justification for nonconsensual fatal human testing?

You're trying too hard to find suffering. It should have been clear I was comparing to "short lives of borderline starvation before they are eaten" and "torture", not to a paradise without disease.

Most humans and most medium to large animals live pretty comfortable lives by default. It's easy and simple to draw a line where they're above the line.

The claim above is about mice. I don't know if nervous tiny prey have consistently worse lives than lab treatment in the wild, but it's not an immediately outrageous claim.

> You're trying too hard to find suffering

I don't know how you can say this, frankly, unless you do not see or experience any of this suffering yourself. If not, you should be extremely grateful for your luck.

You don't have to try hard, at all, to find it.

> Most humans and most medium to large animals live pretty comfortable lives by default. It's easy and simple to draw a line where they're above the line.

How are you defining this, and again by which principle do you draw this line which isn't some ex post facto version of "well humans are here, so I'm going to say everything to the left of here"?

> I don't know if nervous tiny prey have consistently worse lives than lab treatment in the wild, but it's not an immediately outrageous claim

We should be extremely skeptical of this claim because it's so convenient.

It absolves us of the guilt for the tremendous harms we're causing them by proposing something which we would see as obviously absurd if it were applied to us:

"A short, miserable life of having all sorts of horrifying experiments performed on you in a lab is better than being free to live your life and die in the world at large." - Person who benefits from performing experiments on you in a lab

Sure, I'm skeptical of the claim.

But your attempts to equate it to the human experience are a waste of time. Nobody actually thinks humans (or cats, or bison) face the experience that SoftTalker was trying to describe. When we go by the actual intent of the statement, either it's true and mice but not humans face that, or it's false and neither mice nor humans face that. Any experience that you can say is the human default in full honesty, is not the experience SoftTalker was trying to convey.

> short lives of borderline starvation before they are eaten

Ah yes, how silly of me to criticize this as a 2 dimensional statement utterly lacking in self-awareness.

Humans have always enjoyed objectively long lives, and mouse lives are objectively short. Relative to what? Let’s not think about that and while we're at it take the implication for granted that length of a life grants it weight.

This is why we have always honored the fact that tortoise lives are more valuable than those of humans after all.

Borderline starvation is definitely not something millions of humans face every day, or something most of humanity faced for most of history.

Being eaten is so horrible that you’re better off being tortured to death in a lab. Thankfully that has never happened to humans in the wild. Meanwhile, dying of war, famine, cancer or diarrhea is comparatively great, which is why using you in a lab would of course be unthinkable.

Mice struggling for food and survival in the wild are totally different from humans struggling for food and survival in the wild, because, see, uh...

That sentiment is a dismissive, incurious oversimplification of the entire existence of other species. It has so many blind spots, and is seemingly oblivious to the fact the same reasoning could be turned on us. If a hard, short, dangerous life justifies putting a creature in a lab, then a large share of humanity, past and present, would qualify for the same.

> Ah yes, how silly of me to criticize this as a 2 dimensional statement utterly lacking in self-awareness.

You can criticize it but you can't strawman it to oblivion and back.

> objectively

You know what they meant though.

Lines can be set.

And the higher you set the quality of life line, the nicer your lab needs to be to meet the "better than life outside" standard! It's not an excuse for unlimited torture.

> millions of humans

Again, this is about the default. "millions" is 0.1% of humans. And the real number is higher than that but not that much higher.

> Being eaten is so horrible that you’re better off being tortured to death in a lab.

No, that wasn't the argument. And lots of animals we eat do live perfectly good lives.

> Mice struggling for food and survival in the wild are totally different from humans struggling for food and survival in the wild, because, see, uh...

Much much less struggling and starvation and fear fear fear.

You're mocking the statement without trying to understand it. That's rude and extremely unproductive.

Criticize what they meant, not the dumbest interpretation you can think of. Prove their actual intended meaning wrong.

I mean, at least that pain isn't being forced upon them by some higher being for the purpose of scientific fraud or similar, or be deliberately bred to be immunocompromised

I bite the bullet on this one by saying we should in fact do something about that. All of it. And we will, when we can. Probably before 2400.

No matter how many checks, there's a human trial, and that may not go to plan either: http://news.bbc.co.uk/1/hi/england/london/4811626.stm

--

Two men who fell seriously ill following a clinical drugs trial remain in a critical condition but four others are showing signs of improvement.

All six are still in intensive care in Northwick Park Hospital, north-west London, after falling ill on Monday.

TeGenero, which manufactures the anti-inflammatory drug, says it has apologised to the men's families.

All patients recovered, the final patient left hospital after 3 months. Not ideal. The company went bankrupt and the outcome was better regulation in the UK and Germany; great wiki article here: https://en.wikipedia.org/wiki/Theralizumab including the relevant:

    Critics of animal testing have cited the case to argue that experiments on nonhuman animals, even in species closely related to humans, are not necessarily predictive of human responses, and cannot justify the harm inflicted on animals or the resultant risks to humans

This one says four months and the participant "had his toes and parts of his fingers amputated", indeed not ideal.

https://www.bbc.com/news/magazine-35766627

That was an antibody against human CD28, and I'd heard speculation at the time that the monkey trials failed to predict the human response due to differences between monkey and human CD28. That was apparently not correct, since "the sponsor had not only determined the relevant extracellular sequence (which is 100% identical to the human one) but also the binding affinities of the mAb, which were identical for human and monkey". That author goes on to propose a different explanation, though I don't know if any explanation is generally accepted.

https://pmc.ncbi.nlm.nih.gov/articles/PMC2990150/

It's funny to recall this in the context of Terence Tao's question of whether you'd take an AI-designed drug with no mechanistic insight. The sponsors of this trial had reason to believe they had strong mechanistic insight, but they were wrong. A pharmacologist from the nineteenth century would have probably run the trial more safely (lower starting dose, one participant at a time) without that false confidence.

Okay, so let's not do the human trials. Just be content with the modelling (which is known to be imprecise especially where it matters most), and just approve the drug. What would that even solve? Instead of 6 volunteers getting injured, all the customers/patients who bought and took the drug would get harmed.

There is always human trial — the whole history/timeline of people taking a particular medicine is effectively a decades-long human trial. That's actually the only way we learn about long-term side effects, see paracetamol for instance. All you're proposing is we remove a smaller, controlled pre-trial before releasing the medicine into the wilder population. Sorry, but that won't solve anything except for your feeling of moral indignation.

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.

that does not sound at all like my lab experience, nobody was doing ad hoc trial and error on basic animal management

> Treat them as a pet owner would treat their beloved pet rat.

They do that already? Then how come an estimated 192 million mice are killed each year?

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 think that oversimplifies it. One approach is use animals early because if there is no effect in animals, then you don't have an effect.

Once you have an effect, use in vitro models/cell culture to figure out potential mechanisms, then test again in animals to confirm.

If you are doing a PhD (or whatever), do you want to spend 5 years not knowing if your discovery only works in test tubes?

> reasonable confidence the risk and severity of harm from the testing will be minimal

this sounds good, but how do you actually define that? how harmful do you consider giving an animal a UTI (which has the chance to infect kidneys, etc if untreated)?

> For test animals that are or aren't harmed, allow them to live happy, (relatively) normal, natural lives

I agree on principle but for infectious disease you can't really do this without potentially releasing lab strains into the wild.

Something you didn't mention would be to try to use as many parts of an animal that you sacrifice as possible, but then again you're just cluttering expensive -80C freezer space with random tissues if you don't use them. But sometimes when doing ex vivo experiments you can turn one mouse intro multiple samples if you cut the tissues up and expose each one to different conditions, rather than treating the whole mouse. That can save mice..

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…?

Even within animal testing, the cost and convenience of animal testing could vary tremendously. Working in animal testing intrinsically means running a death-camp for mice. The mice aren't are barely even raw materials, they're basically stationery, more treated more like a mobile pH test-strip then a living creature.

There are ways that animal testing could be made far more humane (like birth control to prevent the kind of overbreeding that demands mass culling), but those, themselves, would be costly.

And in order to test cures, you must first inflict the animals with the disease. How many do you need? Are you sure? Because waiting to produce more later would also create costs. Cheaper just to breed a huge batch now and then cull the surplus.

... so how many extra do you want?

Animals are NOT cheaper. Pharma companies do everything to avoid animal testing.

For most drugs the only reasonable test animals are mice and rats. And they LOVE to naturally get cancer and other diseases, so you need to use large groups to get even a semblance of statistical significance.

Oh, and don't forget that it takes a lot of time to run the tests. And time IS money for pharma.

animals are very expensive, but usually you learn something about biology even if mouse results don't always translate to humans.

wasting time and money on a drug that only works in a test tube/organoid/co-culture model is also very expensive.

we should keep funding both, and if the non-animal are good enough, it won't be difficult to convince scientists to save money by using them.

Time to feed my beloved pet snake.