Hacker News new | past | comments | ask | show | jobs | submit login
Humans are not automatically strategic (2010) (lesswrong.com)
60 points by walterbell 10 months ago | hide | past | favorite | 36 comments



So what’s the deal with lesswrong? People there seem to take a generally academic-ish or intellectual-ish tone, but there aren’t any sources in this article and:

> Why? Most basically, because humans are only just on the cusp of general intelligence. Perhaps 5% of the population has enough abstract reasoning skill to verbally understand that the above heuristics would be useful once these heuristics are pointed out.

Is a wild claim to make unsupported, right?


Less wrong can really only be understood in context, and the context is recovering children of ferverent evangelicals. If you grew up in a reasonable household, The thought “maybe I, a mildly gifted 12 year old, am right and every trusted adult I’ve ever seen and all of the books they have and all of the people they talk to are wrong” is going to be disproven every time. If you grow up in the sort of household that produces less wrongers, you need to take that sort of thought seriously or you will end up believing the Earth is 2000 years old and hurricanes are caused by gay people.


It is a wild claim. I think it's important to understand that LessWrong -- whatever it eventually became -- comes out of the SF Bay Area "Rationalist" community. Despite claims to generality, it's a narrow way of reasoning about the world with some odd outcomes. Personally, I think the notion of an "information hazard" is a fine story point in a Neal Stephenson story but taking the idea seriously enough to have nightmares about Roko's basilisk and to ban discussion of it[1] is pushing past the bounds of 'rationality' I think. That a chunk of the core LessWrong crowd went into Effective Altruism ends-justify-the-means think or into neo-reactionary circles -- to a less degree -- is also worth noting: despite claims with regard to general well-being both political ideologies are more than happy to cause mass suffering so long as the expected outcomes are good, eventually, maybe.

[1]: https://en.wikipedia.org/wiki/Roko%27s_basilisk#Reactions


This claim does not look so wild of you encounter people outside your usual educated and smart bubble. Just recently in the sauna a girl explained to me how salt is good to your soul. But not white chemical salt from the store. Only raw salt from under the ground, 10€ per small bottle. She does not sell it, just a victim herself.


Maybe she likes the way eating raw salt makes her feel.


No, the claim is still wild. Ignorance of chemistry in one person or woo-woo beliefs is not an indication for the article's general claims about humanity. You also do not know me or my social circle. There's a cynicism in your post that says maybe more about you than about the claim in the article, friend.


I was not that cynical before, but I live in a town where any encounter outside my friends makes me grind my teeth. Taxi driver tells me how covid is fake and all (sic) doctors are there to kill you, not help you. Doctor tells me how I need to believe in god and I will be cured. My mum believes in all kind of stupid stuff: astrology, fake science and totally fake history.

Even my close friend speaks about cosmic energies and healthy sound vibrations that channel meridians (the only exception I have among my friends).

I can't handle this shit anymore and I am loosing faith in humanity.


"Cos-playing academia" is not bad as a description of lesswrong...


My main issue with this article, as well.

Generally they’re essays about cognitive science so they feel a little more free to pontificate than, say, biologists are, but yeah I also would like to see a lot more sources. They are in their own discourse - note that this is a response to another (anonymous) post on the forum.

When you’re having a discussion that the scientific consensus (of the 2010s, at least…) finds alarmist and non-scientific, you can see why you’d be attracted to sources in your own little ecosystem. It is quite a diverse ecosystem, despite its shared interests


It's an extremely wild claim that demands some sort of supporting evidence.

But it's pretty standard sort of fare for lesswrong. Despite the air it tries to put on, lesswrong is essentially just an opinion site.


Lesswrong is a strange and fascinating place with some elements that I find disturbing and problematic and others that are quite admirable. It's a place where "semi-crank" people can go and have their ideas taken moderately serious. Like academia, there's a lot of reference to past discussion but if anything they take ideas more seriously than academics - academics will drop ideas that seem like dead-ends or counter-productive but lesswrong won't give up on a thing 'till it's proven wrong. A lot of ideas have been developed, some rigorous and worth knowing, some I find pseudo-intellectual hogwash.

The place indeed has a set of common assumptions that have developed over time and may seem strange to the average person - some of these are quite interesting and others I'd find rather twisted (recently upvoted articles: "AI Girlfriends Won't Matter Much"[1]).

There's too much to summarize but I occurs to me that programmers generally have a certain tendency to instrumentalize human beings and human qualities, treat these as things or as control problems and a lot of lesswrong concerns (rationality, AI and AI-safety, effective altruism, "reproductive strategies", etc) come of this. If anything, Lesswrong is resisted the movement we've seen of libertarians drifting to the "alt-right" and so it's one forums for "this kind of stuff".

Overall: worth looking at and even posting but also it's various ideologies have used by people doing real harm (Sam Bankman-Fried etc). Further, it can produce a situation where it's more reasonable concerns ("AI safety") get distorted by and tarred with it's more ideological concerns.

[1] https://www.lesswrong.com/posts/pGhpav45PY5CGD2Wp/ai-girlfri...


The first sentence seems fine (humans are somewhat obviously just on the cusp of general intelligence, in evolutionary terms).

The second sentence starts with a "perhaps", so isn't trying to present itself as fact, though agree is still overstated. My best guess is it's about right (I think 5% is somewhat of an underestimate, but not like massively so).

Seems like a reasonable criticism to me of the post. Standards for posts on LessWrong are mostly just like normal blog posts, though people try pretty hard to be precise in their guesses (hence the use of 5% in the above, which I think by people not used to LessWrong communication norms reads as making a reference to an established fact or scientific study or something).


I mean it’s not stated as a strong claim, but a rough guess given the “perhaps”. I do agree saying an exact number with 0 basis is not a great argument though.


Every time I read an article from there I think, "I wish I can slap a Wikipedia 'citation needed' on this"


They are 'less wrong' in the sense that you can't tell if they are right.


LessWrong is well, akshyually personified. It's where a bunch of guys with an IQ of 120 go to circle jerk each other to feel superior to people with an IQ of 110. Similar to reddit, whereas reddit is where the 110 IQ people go to feel superior to people with an IQ of 100.


> I have enough abstract reasoning ability to understand […] that ice cream is not healthy

Ice cream might be healthy though: https://www.theatlantic.com/magazine/archive/2023/05/ice-cre...


Can someone summarize the findings for those without a subscription? I can totally understand “due to some hidden dynamics, ice cream consumption is correlated with health in type 2 diabetics”, but I can absolutely not understand “ice cream is a healthy food to eat in general”.


From an earlier discussion on here:

>the "health benefits" of ice cream are not explored because the scientists "do not believe in them"

Hence the „might be“. So there have been some positive effects, but no deeper investigation yet so we can’t know for sure.


you can also prepare infinite ingredient based ice cream which might have different properties ;)


pass me the broccoli ice oat 'cream' please


we might be onto something here.


That sentence lacks context. It’s like saying strawberries are healthy. Makes no sense without any context. Diet and lifestyle can be healthy or unhealthy. Single ingredient, not really. There’s no measure of healthiness for a single ingredient and I’m pretty sure you could have ice cream every day and still be healthy long term.


> Diet and lifestyle can be healthy or unhealthy. Single ingredient, not really.

That’s false.

There are many ingredients and foods that we could name that are without dispute in and of themselves unhealthy, regardless of general diet and lifestyle.


Not really. That’s an orthorexic view of food.


> Why do even unusually numerate people fear illness, car accidents, and bogeymen, and take safety measures, but not bother to look up statistics on the relative risks?

Its all about the impact of the outcomes. No one is fearing stepping on a lego brick. Getting seriously ill or getting seriously injured in a car accident can be ruinous health wise as well as financially. They are serious implications. Some never recover from illness or car accident. You'll recover from stepping on a lego though ;)


> No one is fearing stepping on a lego brick.

No one? When my kids were little and lego bricks were hiding everywhere, I lived in overt fear of stepping on them in the middle of the night. It's why I developed a habit of wearing hard-soled slippers at home, even to this day.


You weren’t fearing your life, merely pain avoiding.


You're missing the point, all the things are very bad but have very different relative risks.


No, I don't think this is a great example...

Car accidents are something directly under your control if you're a driver and your choices dramatically affect your chances of getting in one. Fear of these events can change your behavior thereby changing the occurrence profile you fit under.


Related:

Humans are not Automatically Strategic (2010) - https://news.ycombinator.com/item?id=9948600 - July 2015 (11 comments)

Humans are not automatically strategic - https://news.ycombinator.com/item?id=1673144 - Sept 2010 (35 comments)


I would love to hear a comment from someone who found this useful! It’s well written, well reasoned, but IMO completely misses a key fact that makes this framing less useful: just because something isn’t intuitive doesn’t mean it’s not natural —- if anyone doesn’t find that obvious I can find some Chomsky clips giving the argument. I think the word “automatic”, which is not one I’ve seen much in CogSci lit, is responsible for this loss of resolution and ultimate misunderstanding.


So how can we get more strategic?


Require self-proclaimed economists to be rich and posters like these to be very disproportionately productive at something that is valued by at least one person that is not himself


This sounds like someone who has studied a lot of math or physics but has never even heard of the humanities, art, philosophy, etc.


Well, it was posted on lesswrong.com




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: