Can’t recommend the book that coined this acronym enough: The WEIRDest People in the World
Book by Joseph Henrich.
It is such an eye-opening piece that explains so much of the world around us. He’s an anthropologist that goes into the psychology of it all. Touching on points like how religion plays a part in shaping the America of today and even how humans are worst at discerning faces today because we need to discern letters and words and dedicate brain power for that.
There are so many interest studies mentioned there, one that really stuck with me is how Protestant-raised Americans will work harder for the next day after having (reasearch-led) incestuous thoughts when compared to Catholics and Atheists.
He explains how monogamy is to blame for a lot of our western views today, and how Mormon towns in Utah were affected by not having monogamy as the basis of society (women there tend to prefer to be 2nd wives of a better man rather being the only wife of a lower-ranking man).
One of the wildest claims in there is the one that the north of Italy is more developed today because it was part of the Holy Roman Empire while the south wasn’t. About a thousand years separate these and he finds effects still. Mostly in connection to the spread of read/write to the public being a core tenant of Protestantism.
Anyway, this is not a summary of the book but instead a few points from it that really stuck with me after reading it. Fascinating stuff
Waterluvian 3 hours ago [-]
> humans are worst at discerning faces today because we need to discern letters and words and dedicate brain power for that.
I have absolutely nothing to back this up, but my gut tells me this risks being one of those bold claims that grows legs and runs for a while until we debunk it.
bitexploder 3 hours ago [-]
Yeah, has that Malcolm Gladwell knowledge porn vibe. A book that empowers its reader with secret knowledge of explanation that all fits together a little too neatly and loses nuance or is often just plain wrong.
Terr_ 2 hours ago [-]
I'm struggling to think of any way to test the hypothesis which is (A) practical and (B) accurate.
For example, suppose you sampled a group today and found an inverse-correlation between "good at recognizing many faces" and "good at recognizing written text"... That still wouldn't show that one facility grew causing the other to shrink, because maybe people are just born (or early-development-ed) with a certain bias.
raincole 1 hours ago [-]
After finding out even Think Fast, Slow (a book from a very creditable researcher and nobel laureate) is full of replication crisis, I approach pop-sci as entertainment instead of self-education.
blululu 1 hours ago [-]
*nobel memorial laureate.
This is exactly why people get annoyed with the branding of the bank of Sweden’s economics prize. We have yet to see the prize for chemistry awarded for research that does not reproduce.
cjauvin 3 hours ago [-]
One of my greatest pleasure of random walking the internet is building my list of possible next books to read.. thank you for this one!
DaveZale 3 hours ago [-]
yeah I just ordered a copy of "We Survived the Night" based on a post here. Never would have heard of it otherwise.
hiatus 3 hours ago [-]
Mind linking to the post for the curious?
minhaz23 30 seconds ago [-]
Would love to see that link and any other posts others might have run across here as well. I feel like pre covid it was common to see high Quality ask hn discussions with niche/prestigious book recommendations often. I dont see that as much now.
derefr 4 hours ago [-]
> One of the wildest claims in there is the one that the north of Italy is more developed today because it was part of the Holy Roman Empire while the south wasn’t. About a thousand years separate these and he finds effects still.
I would note that the north and south of Italy have very different geography and climate. Which can be upstream of all sorts of things, culturally. The geography of Italy's two halves support different types of economic activity; and the social realities of living within these different economies, naturally evolves into major differences in culture. (Compare/contrast: the differing cultures of coastal vs midwestern America. Now imagine that split with a few thousand more years for the divergence to take hold.)
History happens once; but geography is always affecting a nation, all throughout its evolution. So if you're looking for reasons that two sub-populations within a country might have noticeable differences today, differing geography is going to be the "horse", while history is more of the "zebra."
That being said: geography can also constrain history.
Southern Italy is almost entirely coastline, in a part of the world where, for much of the last ~2000 years, everyone was constantly invading everyone else by sea. Northern Italy was relatively-more immune to amphibious assault, as its capitals could be situated more inland. (Rome itself — the exception that proves the rule — was located in south Italy, but was defended from amphibious assault mostly by the Roman Empire's huge naval home-fleet being docked to the southern-Italian coast; not by anything inherent to its location. Once the Roman Empire itself went away, big rich cities in southern Italy suddenly became juicy targets for conquest and/or sacking.)
marcus_holmes 4 hours ago [-]
Also that Italy has only been a single country for a couple hundred years [0], so there will obviously still be regional differences.
Yes, your point and other points around the web I’ve seen make his argument about north and south Italy very controversial to say the least. He does have data to back it up, where he presents distance to nearest church as a predictor for how well a population will fare, and south Italy didn’t have the churches that north the Italy had
cortesoft 2 hours ago [-]
This would only work as evidence if church placement was random. Could be a correlation with a different cause.
notjoemama 3 hours ago [-]
I haven't read the book but it sounds really interesting. Regarding tone though,
> monogamy is to blame for a lot of our western views today
Does the author use the word "blame" to mean "the reason for" or do they present it as a critique of monogamy? Not a big deal, just made me curious when I saw that.
psidium 3 hours ago [-]
I meant it as “is responsible for” or “explains”. The author doesn’t seem to make any judgement in over the other, but he presents polygamy in a society as a causation for male violence. Sorry for that, English isn’t my first language
cortesoft 2 hours ago [-]
> women there tend to prefer to be 2nd wives of a better man rather being the only wife of a lower-ranking man
This seems like a bit of a tautology; how are they 'ranking' these men?
smelendez 1 hours ago [-]
I haven’t read it but I think that’s genuinely interesting and not obvious.
And probably could change based on the roles of first and second wives and, yes, how male status plays out and how it influences the life of the wife.
We don’t have legal polygamy but in many places there’s not much stopping people from living in an unmarried multi-woman household with a man (or vice versa). But it’s not a very common arrangement, and it’s interesting to think about why.
felipeerias 4 hours ago [-]
There is undoubtedly a real effect here, but IMHO one problem with the original article is that it treats the US as the only reference point.
On the one hand, this reflects the US dominant position in world affairs and the fact that probably most of the training materials come from there.
But on the other hand, there are some outlier results that are left unexplained. For example, ChatGPT is even more aligned with Japan than with the US.
coherentpony 3 hours ago [-]
> For example, ChatGPT is even more aligned with Japan than with the US
I have no expertise in this field.
Is it actually even more aligned? Or is it simply aligned with the elements of Japanese culture and/or media that are exported to the West?
cortesoft 2 hours ago [-]
I think the comment is basing it on the graphic from the article.
yellowapple 1 hours ago [-]
“Do non-American LLMs (e.g. DeepSeek, Mistral, Apertus) perform better or worse here? Do they have their own cultural biases in-built?”
I'm wondering the same thing, in addition to the related question of “Would an LLM perform better or worse if prompted with languages other than English?”.
tropdrop 58 minutes ago [-]
ChatGPT is worse in Russian. Example: after accurately noting that a name appeared in a particular Russian book, it asked if I wanted the direct quote in Russian. I said yes. At this point it switched to Russian output but could no longer find the name in that book, and then apologized for having used what seemed to have been "approximations" about the book before.
(I did then go and check the book myself; ChatGPT in English was right, the name is there)
ehnto 44 minutes ago [-]
I was using Qwen3 locally in thinking mode, and noted that even if it is talking to me in Japanese, it is doing it's "thinking" steps in English. Not having a full understanding of how the layers in an LLM handle language connections I can't say for sure, but for a human this would result in subpar outcomes.
For example (not actual output):
Input: "こにちは"(konichwa)
Qwen Thinking: "Ah, the user has said "こにちは", I should respond in a kind and friendly manner.
Qwen Output: こにちは！
It quiiiickly gets confused in this, much quicker than in English.
barnabyjones 2 hours ago [-]
I have saved instructions for Gemini to translate queries into the local language then retranslate the output back to English, when asking about non-English speaking countries/cultures. It seems to work fairly well, but I think it's just due to the different content trained in that language; obviously there would be more in depth discussion of Indonesian cuisine in Indonesian. Whether the country is rich or democratic shouldn't really affect the output.
decimalenough 2 hours ago [-]
The headline should retain the caps, since WEIRD here is not the same as regular weird.
simonw 5 hours ago [-]
WEIRD here stands for "Western, Educated, Industrialised, Rich, Democratic".
YurgenJurgensen 4 hours ago [-]
I’ve seen a massive uptick in the use of ‘weird’ as an insult (charitably because all the old insults get you shadowbanned on social media, less charitably because conformism is what the mainstream values more than anything), so the author isn’t even pretending to hide their agenda here.
cortesoft 2 hours ago [-]
I think they chose the WEIRD acronym to challenge the western centrism. For most of the readers of the book, the culture described by the WEIRD acronym is not only the normal culture, but is in many ways considered universal.
By calling it WEIRD, the author is trying to drive home the point that the vast majority of people in the world are NOT in that culture that many westerners feel is 'normal', which would make it 'weird' in the sense that it isn't actually the norm.
Now, I have a lot of problems with the book and his arguments, but I don't think there is anything sneaky or nefarious about the word choice, it is very up front and straightforward as to the reasoning behind it.
maxbond 2 hours ago [-]
WEIRD is not pejorative in TFA. There's no problem being WEIRD. I am WEIRD. What's alleged in TFA is that AI, as it's currently deployed, is implicitly chauvinistic towards perspectives other than WEIRD. This sort of thing has historically been a problem with AI/ML and automation in general. The classic example is cameras deploying autofocus features that fail on non-white faces (which has happened several times).
Poorly considered automation can create frictionless experiences for some and Kafkaesque experiences for the rest, where systems refuse to accept your atypical name, your atypical style of speaking is flagged as an indicator of fraud, etc. Automating processes involving people necessarily makes assumptions about those people, and such assumptions are often brittle.
For example, it's easy to imagine a resume filtering AI being implicitly prejudiced against people from Fictionalstan, because it was only trained on a few resumes from Fictionalstan and most of those happened to be classified as "unqualified". This is a danger anytime you have a small number of samples from any particular group, because it's easy for small sample sizes to be overwhelmed by bad luck.
In general I think these types of issues are best viewed as software bugs. It's a clearer and more actionable perspective than as ideological issues. If the software isn't serving some of our end users properly, let's just fix it and move on.
shermantanktop 3 hours ago [-]
The idea that “mainstream” values “conformism” seems like a relic of the 1980s. Have you looked around at public figures in the news? There’s less Debbie Boone and more Dennis Rodman going on. The freaks are flying their flags out there for everyone to see.
add-sub-mul-div 3 hours ago [-]
Could you explain the agenda to those of us like me, who missed it anyway?
DaveZale 3 hours ago [-]
I heard a congressman in a town hall meeting last night call a colleague "crazy liberal" - a psychotherapist called in and said don't use that word "crazy" - language is being perverted here
derektank 5 hours ago [-]
Interesting that the responses from ChatGPT on the World Values Survey correlated most closely with the responses from Australians and New Zealanders.
marcus_holmes 4 hours ago [-]
I expect (as TFA says) that they would most closely align with Californians, but that isn't in the data.
jdlshore 4 hours ago [-]
I imagine the culture of HRLF trainers affects things. Maybe there’s disproportionally more of them from Oz/NZ, as native English-speaking countries with possibly lower wages?
derektank 3 hours ago [-]
I had heard that a lot of human feedback was being provided by people in Nigeria, as it has a very large English speaking population (owing to its history as a British colony) while also having low wages. This was the explanation given as to why ChatGPT seemed to use the word "delve" so often, apparently it's used much more frequently in business contexts there.
Possible that they're using different sources of feedback for different training though
cortesoft 2 hours ago [-]
I wonder how ChatGPT and the like would do if you asked it to give a response as if they were a person from one of those other cultures.
In other words, is the issue in the defaults or is it impossible for AI to respond from other cultures?
qwertytyyuu 2 hours ago [-]
Oh it’s an arcronym. I was very confused for a good portion of the article
m0llusk 5 hours ago [-]
It seems like almost all contexts might get value from specialized training. People often vary radically depending on where they were raised and where they live, their occupation and social class, and a range of other factors. Even workers from essentially identical backgrounds but practicing different trades can have very different perceptions and framing for what might appear to be shared tasks.
5 hours ago [-]
DaveZale 5 hours ago [-]
Ummm... doesn't the AI have to scrape the data of those non- WEIRD cultures to work then? What am I missing here?
There are parts of the world where constant person-electronic connection isn't a thing. Is that your point?
psidium 5 hours ago [-]
I don’t have the data but I assume the corpus available to train an LLM is majorly in English, written by Americans and western counterparts. If we’re training the LLMs to sound similar to the training data, I imagine the responses have to match that world view.
My anecdote is that before LLMs I would default to search Google in English instead of my own native language simply because there was so much more content in English to be found that would help me.
And here I am producing novel sentences in English to respond to your message, further continuing the cycle where English is the main language to search and do things.
tropdrop 50 minutes ago [-]
In my experience, ChatGPT, at least, seems to have had multiple languages used to train its corpus. I am guessing this based on its interaction with me in a different language, where it changed English idioms like "short and sweet" to analogous versions in that language that were not direct translations.
But my guess is that the data sets used from the other languages are smaller (and actually, even if it had perfect access to every single piece of data on the internet, that would still be true, due to the astonishing quantity of English-language data out there compared to the rest. Your comment validates that). With less data, one would expect a poorer performance in all metrics for any non-Anglophone place, including the "cultural world view" metric.
klooney 4 hours ago [-]
And the RHLF was directed by Californians, and so the "values" are likely very California.
DaveZale 5 hours ago [-]
english is the lingua franca ;-)
YurgenJurgensen 4 hours ago [-]
“Fancy autocomplete better at completing documents similar to ones it has seen before” isn’t as headline-worthy.
nstj 59 minutes ago [-]
Well said. It's so odd that this isn't just the general societal talking point when it comes to LLM's
dwoldrich 4 hours ago [-]
This is some WEIRDO twist on the grievance olympics. AI is just software, is not alive, and most importantly, it is the current thing with which we can dissolve ourselves into.
I support the current thing, as should everyone.
Rendered at 05:04:37 GMT+0000 (Coordinated Universal Time) with Vercel.
It is such an eye-opening piece that explains so much of the world around us. He’s an anthropologist that goes into the psychology of it all. Touching on points like how religion plays a part in shaping the America of today and even how humans are worst at discerning faces today because we need to discern letters and words and dedicate brain power for that.
There are so many interest studies mentioned there, one that really stuck with me is how Protestant-raised Americans will work harder for the next day after having (reasearch-led) incestuous thoughts when compared to Catholics and Atheists.
He explains how monogamy is to blame for a lot of our western views today, and how Mormon towns in Utah were affected by not having monogamy as the basis of society (women there tend to prefer to be 2nd wives of a better man rather being the only wife of a lower-ranking man).
One of the wildest claims in there is the one that the north of Italy is more developed today because it was part of the Holy Roman Empire while the south wasn’t. About a thousand years separate these and he finds effects still. Mostly in connection to the spread of read/write to the public being a core tenant of Protestantism.
Anyway, this is not a summary of the book but instead a few points from it that really stuck with me after reading it. Fascinating stuff
I have absolutely nothing to back this up, but my gut tells me this risks being one of those bold claims that grows legs and runs for a while until we debunk it.
For example, suppose you sampled a group today and found an inverse-correlation between "good at recognizing many faces" and "good at recognizing written text"... That still wouldn't show that one facility grew causing the other to shrink, because maybe people are just born (or early-development-ed) with a certain bias.
I would note that the north and south of Italy have very different geography and climate. Which can be upstream of all sorts of things, culturally. The geography of Italy's two halves support different types of economic activity; and the social realities of living within these different economies, naturally evolves into major differences in culture. (Compare/contrast: the differing cultures of coastal vs midwestern America. Now imagine that split with a few thousand more years for the divergence to take hold.)
History happens once; but geography is always affecting a nation, all throughout its evolution. So if you're looking for reasons that two sub-populations within a country might have noticeable differences today, differing geography is going to be the "horse", while history is more of the "zebra."
That being said: geography can also constrain history.
Southern Italy is almost entirely coastline, in a part of the world where, for much of the last ~2000 years, everyone was constantly invading everyone else by sea. Northern Italy was relatively-more immune to amphibious assault, as its capitals could be situated more inland. (Rome itself — the exception that proves the rule — was located in south Italy, but was defended from amphibious assault mostly by the Roman Empire's huge naval home-fleet being docked to the southern-Italian coast; not by anything inherent to its location. Once the Roman Empire itself went away, big rich cities in southern Italy suddenly became juicy targets for conquest and/or sacking.)
[0] https://en.wikipedia.org/wiki/Unification_of_Italy
> monogamy is to blame for a lot of our western views today
Does the author use the word "blame" to mean "the reason for" or do they present it as a critique of monogamy? Not a big deal, just made me curious when I saw that.
This seems like a bit of a tautology; how are they 'ranking' these men?
And probably could change based on the roles of first and second wives and, yes, how male status plays out and how it influences the life of the wife.
We don’t have legal polygamy but in many places there’s not much stopping people from living in an unmarried multi-woman household with a man (or vice versa). But it’s not a very common arrangement, and it’s interesting to think about why.
On the one hand, this reflects the US dominant position in world affairs and the fact that probably most of the training materials come from there.
But on the other hand, there are some outlier results that are left unexplained. For example, ChatGPT is even more aligned with Japan than with the US.
I have no expertise in this field.
Is it actually even more aligned? Or is it simply aligned with the elements of Japanese culture and/or media that are exported to the West?
I'm wondering the same thing, in addition to the related question of “Would an LLM perform better or worse if prompted with languages other than English?”.
(I did then go and check the book myself; ChatGPT in English was right, the name is there)
For example (not actual output):
Input: "こにちは"(konichwa) Qwen Thinking: "Ah, the user has said "こにちは", I should respond in a kind and friendly manner.
Qwen Output: こにちは！
It quiiiickly gets confused in this, much quicker than in English.
By calling it WEIRD, the author is trying to drive home the point that the vast majority of people in the world are NOT in that culture that many westerners feel is 'normal', which would make it 'weird' in the sense that it isn't actually the norm.
Now, I have a lot of problems with the book and his arguments, but I don't think there is anything sneaky or nefarious about the word choice, it is very up front and straightforward as to the reasoning behind it.
Poorly considered automation can create frictionless experiences for some and Kafkaesque experiences for the rest, where systems refuse to accept your atypical name, your atypical style of speaking is flagged as an indicator of fraud, etc. Automating processes involving people necessarily makes assumptions about those people, and such assumptions are often brittle.
For example, it's easy to imagine a resume filtering AI being implicitly prejudiced against people from Fictionalstan, because it was only trained on a few resumes from Fictionalstan and most of those happened to be classified as "unqualified". This is a danger anytime you have a small number of samples from any particular group, because it's easy for small sample sizes to be overwhelmed by bad luck.
In general I think these types of issues are best viewed as software bugs. It's a clearer and more actionable perspective than as ideological issues. If the software isn't serving some of our end users properly, let's just fix it and move on.
Possible that they're using different sources of feedback for different training though
In other words, is the issue in the defaults or is it impossible for AI to respond from other cultures?
There are parts of the world where constant person-electronic connection isn't a thing. Is that your point?
My anecdote is that before LLMs I would default to search Google in English instead of my own native language simply because there was so much more content in English to be found that would help me.
And here I am producing novel sentences in English to respond to your message, further continuing the cycle where English is the main language to search and do things.
But my guess is that the data sets used from the other languages are smaller (and actually, even if it had perfect access to every single piece of data on the internet, that would still be true, due to the astonishing quantity of English-language data out there compared to the rest. Your comment validates that). With less data, one would expect a poorer performance in all metrics for any non-Anglophone place, including the "cultural world view" metric.
I support the current thing, as should everyone.