The world optimized itself.
I think you have a good point buried in this post, but this doesn't seem like it. The world didn't optimize itself. Humans optimized for chicken volume, taste, and other parameters, and generally ignored the suffering and well-being of the chickens. It wasn't 'the world' and it didn't just optimize. It optimized certain variables at the expense of others.
Also, I'd push back on the idea (popularized probably most effectively by Pinker) that we're doing 'so much better' now than we were 200 years ago. Again, for certain parameters, yes. For many others, no. The biggest factor in these discussions that you're sort of pointing to is risk. We have increased per capita GDP, health outcomes, etc. But we have also massively increased X risk. We now have multiple non-negligible vectors for causing widespread catastrophe and possible extinction, not just among humans, but against all life on the planet. That in itself complicates the picture of how well we're doing.
Which sort of dovetails with your point, which is that many times things are going great, until they aren't. But you don't specifically frame it as an increase in risk. You seem to be attributing it to a sort of semi-random unpredictable switch from doing well to doing poorly, as a sort of inevitable cost of optimization. I don't think that's right.
We now have multiple non-negligible vectors for causing widespread catastrophe and possible extinction, not just among humans, but against all life on the planet.
Is there any vector other than misaligned ASI that could kill all bacteria on the planet and end biological evolution? I couldn't name one but perhaps there are indeed some I'm unaware of yet?
"a sort of semi-random unpredictable switch from doing well to doing poorly, as a sort of inevitable cost of optimization" I mean, yeah - that's Goodhart's Law. I made a follow-up post explaining it more.
Meditations On Moloch This is the dream time Forages vs farmers. Why the age of elm will happen were the posts which led me to similar conclusions as the author, needless to say AI causing painless extinction is much more bearable of a future than Moloch trap killing humans from inside. Maybe I am not hansonian enough to find these prospects comforting and I don't think the eternal subsistence of "poor people smiling" equilibrium will last. I have lost weeks of sleep over this.
I find myself in the rather peculiar position of fully agreeing with your argument and your conclusions - but disagreeing with the critical central point upon which your argument hinges!
I do not think chickens were ever doing well. I think that "doing well" requires more than just an existence that's pain-free relative to your ancestors' existences, and moreover I think it's essentially orthogonal to population size.
(The argument that population size is orthogonal to wellbeing is pretty self-explanatory so I'll skip that and focus on the potential other requirements of "doing well")
Imagine offering people the same deal the chickens got: "Stop scratching a living in the dirt, in constant fear of predators, and instead be owned, as a form of property, by some superior species. You'll be fed, kept warm, protected from predators and from disaease. Some of you will live for weeks before we kill and eat you, others for years (but of course we'll take and eat your eggs every time you ovulate), and others we'll compel to fight to the death for our entertainment." How far below the poverty line would you have to go - in other words, how bad would the human equivalents of "scratching a living" and "fear of predators" have to get - before most people would start to want to take that deal? I'm not convinced even the inmates of Devil's Island would have taken that deal.
Now suppose, instead of it being a deal offered to humans at all, the superior species simply decided for us. I think most people would agree that, even if overall suffering were reduced, average lifespan were improved, deaths to predation and disease were reduced, etc. etc., that the humans who had been compelled into this arrangement, even if their lives had been entirely brutal beforehand, would not be "doing better".
...but why? If the immediate suffering is reduced, averaged over the whole species, why can't we say that the species is doing better? I think the answer might be some weird hybrid of game theory and deontology like "If your wellbeing is dependent on the caprices of some entity vastly more powerful than you and that doesn't really care about you (in other words "if you're property"..) such that you have a high quality-of-life iff the powerful entity happens to want things for itself that happen to be good for you, then you're not doing well; what you have is just an illusion of doing well".
In short: "doing well" = "immediate quality-of-life" + "future security", and chickens never had the latter term.
Note that this equation doesn't include any terms for things like "freedom" or "self-determination"; my formulation assumes those things are only valuable instrumentally, insofar as they generally provide higher security than does being some other entity's owned property. I'm sure many would argue that "freedom" or whatever is valuable in and of itself, entirely apart from any security benefits it offers, which if it were true would strengthen the argument even further, turning it into: "doing well" = "immediate quality-of-life" + "security" + "freedom", with chickens now missing two of the necessary terms. (But as I say, I don't rely on "freedom" being a final value for the purposes of this argument)
I think the chicken part of your argument is entirely sound, however, since you're essentially still making the same point that chickens never had the freedom-of-action and self-determination they'd have needed to establish sustainable future security.
I think your analogy to humans entirely survives the added "security" constraint, too: because, like chickens, we also don't have the freedom-of-action and self-determination to establish future security for ourselves. I think our future security depends upon the caprices of vast alien entities that humans don't control and don't fully understand, which we happen to call "nation states" and "corporations", that inexorably pursue their own incentive gradients (such as "profit", "power", etc.) regardless of what the people that comprise them might want to do, and that dictate the terms and conditions (literally as well as figuratively) of almost every aspect of our lives.
So long as what the corporations and nation states want is broadly aligned with what we want, we have a high value for the equation's "immediate quality-of-life" term and we might think that things are going really well for us (the aforementioned "illusion of doing well"). But we're not doing well: we need future security too, and we don't have it.
Rapid comment:
I agree a bit and think animal welfare (and AI) are a key area to criticise capitalism on.
But I also think that chicken welfare is likely to improve.
Thanks Nathan.
Yeah agreed chicken welfare is likely to improve.
I seem to remember you relying on the "human progress outside view" in your reasoning about AI - maybe I'm misremembering. My point is more about how much evidence that outside view should be, on an abstract level, rather than any concrete criticism of capitalism. Like, obviously humans would do better for a while, so it cannot be that surprising, so it cannot be that much evidence.
I think my outside view is something more like: humans tend to solve problems in the interests of humans. That doesn't feel that damaged by this piece.
I see - yeah I agree that is different from the main thrust of the piece. I think I address this in footnote 9 a bit, where I call it the "humans will keep becoming more powerful" outside view.
My main objection to that would be that there's not a very strong correlation between humans getting what they want and the total welfare of the world (Agricultural Revolution, factory farming), so the fact that that outside view is strong doesn't seem very reassuring.
(How do you feel about the question of whether the total welfare of the world today is negative because of factory farming?)
Actually, no I changed my mind. It's still captured by the core conceptual point of the piece, there's just no good vivid examples of it like the chicken example.
It's still completely expected a priori that the dominant class of agent at a given stage of runaway Darwinian competition would solve most of their problems and be very powerful. This would be the case even if they were to then proceed to get outcompeted. So if someone hasn't integrated this conceptual point, it should be an update downwards on how much evidence that outside view is for them.
There's also the classic meta-epistemic debate here, on how to weigh empirical outside views against gears-level models (even if the "gears-level model" here is something almost tautological).
Interesting read, thank you for sharing. I think I would distinguish between two "optimization processes" that have molded the fates of living things on this planet (I don't love the term "optimization process" but it will do).
1. Evolution by natural selection which selects for "fitness"
2. Human intentional selection, which selects for usefulness to humans.
As a general principle, I do think you are on to something with the idea of a "honeymoon phase". The general pattern is, new niche opens up, one specie moves into this niche and dominates it (honeymoon phase), other species move into the niche making it much less comfortable and much more competitive than it used to be.
I do think though, that with the emergence of humans on the planet, the second optimization process is dominant. Like you mentioned with the example of the chickens, I do think it does make sense to translate the idea of a honeymoon phase to this optimization process as well, but we must account for the fact that unlike natural selection, what is driving the train now are human preferences. This makes me more optimistic than you are that barring major cataclysms or the emergence of superhuman AIs, humanity will have a bright future.
One last note, if AI development continues and we do enter a post human AI age, then the world might be molded by what the AIs want rather than what we want. In that world, humans might indeed enter a honeymoon phase where the AIs treat us well at first and then really poorly after (like the chicken).
Would be interested in hearing your thoughts on this, and thanks again for sharing!
Hanson seems to argue that we're in an overhang, where we've improved our conditions more rapidly than our biology can catch up. Though, eventually, the general tendency for strategies that reproduce more widely to dominate will inevitably lead to Malthusian conditions.
Where your post doesn't map neatly to this is in how you've selected a case where artificial selection is at play. If humans stopped existing or stopped needing to farm chickens their population would collapse. Maybe to zero. Or maybe some of them could survive and adapt back toward their status quo ante.
That's interesting, because you've suggested that their "state of nature" before humans was preferable to where they've ended up today. If humans went away, their conditions might improve, even if their population fell.
Given that, I think the scenarios you're describing might be a special case of "entity that has become instrumental to another entity's optimization function".
A blunt analogy might be, humans use chickens as a source of energy and protein. Our economic system uses humans for energy production and supply chain maintenance.
Is an analogy like that what you had in mind?
If so, I think that analogy would be worth probing more deeply. Because if you take it at face value, you could almost argue that a "return to nature" à la Kaczynski could be rephrased as breaking free of a higher order optimizer. Which is a fun thought.
This is such a good phrasing, thank you. I was looking for the name of this concept in relation. Have a look at https://thezvi.wordpress.com/2025/12/22/the-revolution-of-rising-expectations/
I think that technology increasingly pushed the surplus avaliable to us above the required spend to keep up with local status competition. This was the case for the boomers and their parents also to some degree.
This is less the case for the subsequent generations.
Expectation has again risen to cover avaliable resources. Let us pray avaliable resources do not shrink further.
The darwinian holiday is over. (Again, excellent phrasing, thank you.)
I hope my post can add something novel, that I haven't come across in the pre-existing discussion (see my footnote 1) of the specter of long-term evolutionary competition: the observation that this at first predictably benefits you (an instance of Goodhart's Law), a vivid example, and a memetic handle for the concept: The Darwinian Honeymoon.
I think there's a few distinctions worth making here. First of all, we don't know what "optimization" looks like for superintelligent AI. It might as well decide happiness is the thing it should be optimizing and create a highly contagious virus that makes humans functionally useless, but gives them the best phenomenological experience.
There's also a timeline where humans could plausibly close the gap with direct biological engineering. I am not sure whether training on human written training text could plausibly reach a point where it saturates the highest level of human intelligence. Maybe there will be no realistic path to get much further in an accelerated sense, I am not sure if we can assume acceleration picks up forever.
I think that people underestimate the potential of direct gene editing. Using CRISPR-based direct gene editing, that already works in humans. (in 4-5 generations, humanity would probably end with people who have IQ scores that are 8-15 SD's above the current mean, considering the additive effect of superintelligent researchers further optimizing this path), and would have a level of intelligence that also makes even the smartest humans in their own fields seem like apes compared to them.
My conclusion is that humans still have enough control over the development of AI that these alternatives feel like real options and real timelines. I am not sure for how long this window will hold. I predict that the moment AI gains cognitive agency over humanity, it will be the one making the decisions, not humans. And all these alternative timelines will collapse into just "what the AI decides".
I think it's worth trying to be much more specific about the mechanism and timeline of this "default outcome". Does it mean "imminently, standards of living will go down, as we run out of resources and overpopulate and compete"? That seems unlikely to me. Acquiring new resources, or using those resources more efficiently to make our lives better, happens on much shorter time scales than making new humans.
Does it mean "in a few hundred thousand years, space mormons will overpopulate the galaxy and run out of resources, because they had a more effective meme to encourage reproduction"? That is significantly less concerning of a problem to me, and it presupposes that the space mormons can't intelligently change course in order to protect their standard of living. Which brings us to another question:
Does the "nihilistic optimization" process under scarce resources bottleneck the well-being of general intelligences? Don't we think that conscious cooperation lets us forecast, understand, and communicate to solve coordination problems? Hasn't it done so over and over in the past, even when resources have been scarce?
AI x-risk can be expressed as a darwinian competition / gradual disempowerment problem, as can the risks of a permanent authoritarian state (like what chickens live under), but I don't think it's the best framing for those issues. e.g. chickens' standard of living could be significantly improved with the right political coalition, and that coalition won't be prevented from forming by fundamental darwinian aspects of reality.
I'm purposefully being very vague about the specifics, because I'm fundamentally only trying to make a conceptual point - about how strong the evidence from the outside view should be.
I am more optimistic than this. I think the main reason to expect things to get better for humans (if there is't an extinction-level disaster) is that this is an aspect of the future which humans can steer, and, while there is a conflict between individual preference and collective outcome involved, the conflict is not very strong.
People want to have descendants (at least, a significant fraction of people, and a fraction that by default will increase over time). People don't want those descendants to live in a malthusian world. These are sort of in conflict, in that if everyone tries to have the maximum possible number of descendents and you repeat this for enough generations, by the power of exponential growth, eventually you exceed the available resources and some people have to starve. But, while people have a preference for descendents, they mostly don't have a meta-preference for their descendents to share that preference. This means that if people coordinate to pick an ideal population-growth stopping point, there can be a many-generations gap in between the last generation that deeply valued having lots of descendants, and the generation that didn't get to have that. And there are a lot of generations left, so many chances to coordinate.
1
Crossposted from Substack and the EA Forum.
A common argument for optimism about the future is that living conditions have improved a lot in the past few hundred years, billions of people have been lifted out of poverty, and so on. It’s a very strong, grounding piece of evidence - probably the best we have in figuring out what our foundational beliefs about the world should be.
However, I now think it’s a lot less powerful than I once did.
Let’s take a Darwinian perspective - entities that are better at reproducing, spreading and power-seeking will become more common and eventually dominate the world.[1] This is an almost tautological story that plausibly applies to everything ever, agnostic to the specifics. It first happened with biological life in the last few billion years and humans specifically in the last hundred thousand years. Eventually, it led to accelerating economic growth in the last few thousand years, and in the future it will presumably lead to the colonization of the universe.
My core point is this: It makes complete sense that this nihilistic optimization process at first actually benefits some class of agent - because initially, the easiest way to keep growing is to use some class of agent in the world and incentivize it by satisfying its preferences. But then, as the optimization becomes more and more advanced, it stops being beneficial - because there are almost certainly some more evolutionarily fit configurations out there than the class of agent that the process just happened to start out with. It’s a “Darwinian honeymoon”.[2]
That’s very abstract - I’ll give an example. The population of the red junglefowl, the wild ancestor of the chicken, was in maybe the tens of millions, concentrated in the Southeast Asian jungle.[3] Then, around seven thousand years ago, humans started to domesticate them into Gallus Gallus domesticus, the modern chicken, and chicken farming spread through Persia and the Middle East into Europe. By 1700, there were maybe a billion chickens in existence.[4]
At that point - before the Industrial Revolution began - poultry farming was still a small-scale, household affair. Chickens were kept in small wooden coops or henhouses at night to protect them from predators, and let out to free-range during the day. Around half of chickens were surplus roosters, young males not needed for breeding and killed at 4–6 months of age. The other half were mostly laying hens, killed at 2–4 years of age. The most common methods of killing were neck wringing and beheading - both of which result in instant death if done correctly.
This was an enormous increase in the total welfare of poultry: a 100x increase in population, being given extra food by humans, and far more painless deaths than in the wild (where they died from predation, disease, and only rarely old age).[5] From the chickens’ perspective, it would have been an unimaginable, mindboggling blessing. Heaven on Earth. Really take in how extraordinary this is - it’s probably larger than the increase in human welfare from 1800 to today that we are all so excited about! (which involved only an 8x increase in population)
Early domestication was intended for cockfighting to a surprising extent (with use for meat and eggs only starting later), an activity presumably very distressing for the roosters. So there was even moral progress over time!
Chicken intellectuals would probably have been saying stuff like this:
“Maybe us and our human friends will colonize the universe together, and transform the trillions of planets of the Milky Way into endless spacious coops to cluck around in. What a bright future we have!”
Then… things changed.
Today, male chicks born at egg-laying facilities are useless to the industry and are killed on their first day of life - typically by being fed alive into high-speed grinders (maceration) or suffocated with CO₂, with billions killed this way every year globally. Female chicks have a portion of their highly sensitive, nerve-dense beaks cut or burned off without anesthetic within days of hatching, to prevent them from stress-pecking each other. Broiler chickens have been selectively bred to grow so fast - reaching slaughter weight in around 6 weeks - that their hearts, lungs, and legs frequently cannot support their own bodies, leaving many unable to walk, suffering heart attacks, or dying from organ failure before they even reach slaughter. Those in conventional housing live on floors covered in accumulated feces, causing chronic ammonia burns to their feet, legs, and eyes. Laying hens spend their entire lives in battery cages with roughly 61 square inches of space each - less than a single sheet of A4 paper - unable to spread their wings, turn around, or take a single step in any direction. When egg production slows, they are sometimes intentionally starved for up to two weeks in a practice called “forced molting”, to shock their bodies into another laying cycle. At slaughter, chickens are shackled upside down while fully conscious, dragged through an electrified water bath, and passed through an automated throat-cutting blade - a process with a well-documented failure rate. The ones that are improperly stunned or miss the blade entirely proceed directly into scalding tanks of near-boiling water, where they are drowned or boiled alive while conscious.[6]
Chickens had normal, happy lives for most of history because humans needed to “cooperate” with them, in a way - there was no alternative to free-range, and confining them in tight spaces would have made disease too much of a problem. Chickens had some negotiating power, so to speak, just by virtue of existing in their natural form, with their natural preferences.
But then, humans invented antibiotics and vaccines to make extreme density viable. They synthesized vitamin D and added it to feed to make sunlight unnecessary. They discovered mechanized processing made farms with millions of birds possible. They figured out breeding science to grow chickens to unhealthy sizes. They found the right culture, and enough psychological distance in the industrial process, to disregard the chickens’ suffering.
The world optimized itself.
People sometimes use chimpanzees as an analogy for AI - we will be to AIs like chimpanzees are to us, and look at the difference in power there. But this can feel dissonant with the fact that humans are obviously doing amazing right now, in a way that apes never were.
The chicken analogy may help bridge the gap. Chickens were doing so well - until they weren’t.
This makes it feel a lot less surprising that humans are doing much better than 200 years ago, and returns us some extent to a state of ignorance about the future.[7] Of course we’d be doing well for a while - that’s how this stuff works.[8][9][10]
You might just be in the honeymoon phase.
People that have talked about this being potentially The Most Important Thing for thinking about the long-term future and influenced me: Nick Bostrom, Robin Hanson, Allan Dafoe here and here (I’d recommend the latter), Dan Hendrycks, Carl Shulman, Andrés Gómez-Emilsson.
It’s also just Goodhart’s Law, I think.
The source for all facts in this post is Claude Sonnet 4.6, and it’s factchecked by another instance of Claude Sonnet 4.6
A lot of uncertainty on these numbers, obviously.
Quick note: I haven’t thought/read about animal welfare, and especially wild animal welfare, much. I take suffering-focused views like Brian Tomasik’s seriously. (and chickens are r-selected and prey animals, which may be bad?). So I am very uncertain in general.
But here, I’m mostly saying that from a “common sense” point of view, these lives seem alright - just like human lives with a distressing death at the end seem alright. So if you believe the human progress prior I’m arguing against, you should also believe this.
The other major counterargument I can think of is that the individual chickens die earlier, especially with the many unneeded roosters. The red junglefowl, in contrast, lives 5 to 8 years in the wild. But from a total utilitarian perspective, this is not that relevant.
Claude wrote this paragraph with some editing and guidance from me.
In a more mundane context, all this also made me more sympathetic to left-wing skepticism of unrestrained capitalism. We just can’t control it or know where it’s going, long-term.
Also, experts tend to think that the Agricultural Revolution (and possibly even the Industrial Revolution for a bit) reduced aggregate human welfare at first. So the outside view of “humans will keep doing better” has failed before - it’s strictly weaker than the outside view of “things will become more optimized”.
My guess is that the total welfare of the world today is probably negative anyway due to factory farming, so that’s another way the optimistic outside view, globally speaking, doesn’t get off the ground.
The big counterargument is that humans are much more powerful and competent than chickens were - we controlled their lives, after all, and it made it easy for us to “disempower” them. Of course, I agree that the situation is different - the “humans will keep becoming more powerful” outside view fares a lot better than the “humans will keep doing better” outside view - even the Agricultural and Industrial Revolutions involved increases in power for the people at the top. That’s the big question - if humans have enough power and competence to stave things off.
I guess I’m trying to shift the frame - getting outcompeted is the default outcome, unless something weird and unprecedented happens (permanent alignment, value lock-in, stable global singleton, something).
(And at the end of the day, power is not the end goal - welfare is. So it should also worry you that that outside view doesn’t seem very strong at all. That’s more in the direction of something like authoritarianism risks.)
Very quick, very speculative, lightly held thoughts on how the honeymoon may already be ending: Humans may be drifting away from our natural instincts due to the differing incentives of the modern capitalist world: Lower marriage rate, lower fertility, etc. Worsening teenage mental health may also be a sign of “underlying unhealthiness”/goodharting in some way. And increasing political polarization could be interpreted as human social instincts getting hijacked by larger entities (“egregores”) into larger-scale, more optimally power-seeking conflict. (Of course, there are other possible hypotheses for all these, and they’re completely independent from the main gist of the post).
And AI replacing us is obviously another option.
Also, I’d be remiss not to mention Ben Garfinkel’s blogpost on another concrete mechanism for how our current Golden Age could come to an end: Is Democracy A Fad?