Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

IQ tests results for AI

trackingai.org

Post reply on HN
Print view View original

231–240 of 359 posts

Re: IQ tests results for AI

#231

Post by hemabe »

jacquesm wrote: post #181
hemabe wrote: post #73

Earlier quoted context omitted.

IQ is largely genetic, even if some people claim otherwise. The evidence for this is now overwhelming: even when different ethnic groups grow up in very similar conditions in the same country, the PISA (which correlates r=0.9 with IQ) scores measured vary greatly. For example, among second-generation children in Germany, there are significant differences in PISA scores. Polish children achieve similar or even better…

Remarkable how you jumped from 'IQ is largely genetic' to 'Turkish children remain at the same poor level that their parents achieved in the tests'. Don't you see your the mistake in your reasoning there? Probably that too can be explained by genetics or maybe by a failing education system but the point is: there are very dumb Germans and very smart Turkish people who would still score different on an IQ test in Germ…

Germans have an average IQ of around 100 (it used to be 105 before immigration). Turks have an IQ of 85-90. If you are not biased, you can clearly see these differences in everyday life. With a few exceptions, the intellectual achievements of fellow citizens of Turkish origin tend to be rather low. The situation is different when you look at migrants from high-IQ countries, such as China (who have an IQ of 104). These children often achieve amazing results, for example in the International Science Olympiads. The US is actually only able to achieve top places in the Math Olympiad thanks to Chinese migrants. Almost the entire US team consists of people with a Chinese background. Look at this picture of the US-Team for 2023 if you want to know, what I mean: https://maa.org/wp-content/uploads/2024/05/Screen-Shot-2024-...

Re: IQ tests results for AI

#232

Post by jacquesm »

hemabe wrote: post #231
jacquesm wrote: post #181

Earlier quoted context omitted.

Remarkable how you jumped from 'IQ is largely genetic' to 'Turkish children remain at the same poor level that their parents achieved in the tests'. Don't you see your the mistake in your reasoning there? Probably that too can be explained by genetics or maybe by a failing education system but the point is: there are very dumb Germans and very smart Turkish people who would still score different on an IQ test in Germ…

Germans have an average IQ of around 100 (it used to be 105 before immigration). Turks have an IQ of 85-90. If you are not biased, you can clearly see these differences in everyday life. With a few exceptions, the intellectual achievements of fellow citizens of Turkish origin tend to be rather low. The situation is different when you look at migrants from high-IQ countries, such as China (who have an IQ of 104). Thes…

Could you maybe fuck off with the overt racism? It would make HN instantly a lot better. You are well into 'not even wrong' territory.

Thanks.

Re: IQ tests results for AI

#234

Post by kcplate »

notahacker wrote: post #169
kcplate wrote: post #116

Earlier quoted context omitted.

There are a lot of people with high IQs that appear to be incapable of performing basic tasks too

Well yeah, I'd be wary of an "all our candidates are MENSA members" staffing agency as well :). But human intelligence isn't quite so easily overfitted to a training set consisting of multiple choice matrices and logic puzzles as neural networks

Sure, I’d probably go the extra step of sorting out any applicant who advertised their MENSA membership…but that’s just me finding those folks pretty insufferable.

But back to the point it’s really more about if the tool for the job can’t do the job…they are the wrong tool for the job. Tool of course being device, AI, or…human.

Re: IQ tests results for AI

#235

Post by jacquesm »

holbrad wrote: post #193
froh wrote: post #91

Earlier quoted context omitted.

especially Germany is a very poor example for this claim as school performance (PISA) in this country correlates with the parents academic background more than anything. if your parents aren't educated you can be intelligent AF and still will fail the German school system while you can be dumb as a brick but still make Abitur if your parents drill you through. in Germany. and Welt is a media source with a right conse…

Surely this is completely logicalal if IQ is largely heritable. The smarter richer parents are more likely (But not guaranteed) to have smarter children. So it would be completely expected for the smarted person you know to come from a rich family even if their environment had no effect. (Though it likely does)

> Surely this is completely logicalal if IQ is largely heritable.

> So it would be completely expected for the smarted person you know

Maybe we should make the occasionalal exception.

Re: IQ tests results for AI

#236

Post by fortran77 »

cedilla wrote: post #41
nsoonhui wrote: post #31

Earlier quoted context omitted.

Another component of this theory concerning g is that it's largely genetic, and immune to "intervention" AKA stability as you mentioned. See the classic "The Bell Curve" for a full exposition. Which makes me wonder what's the point of all the intervention in the form of teaching/parenting styles and whatnot, if g factor is nature and immutable by large? What's the logic of the educators here?

"The Bell Curve" is, let's say, highly controversial and not a good introduction into the topic. Its claim that genetics are the main predictor of IQ, which was very weakly supported at the time, has been completely and undeniably refuted by science in the thirty years since it's publication.

I've read the book. I says that aspects of IQ can be heritable, but doesn't ever say "genetics are the main predictor of IQ".

Quoting direcly from the book: "It seems highly likely to us that both genes and the environment have something to do with IQ differences” and the book states that "the exact contribution of genes versus environment is unknown."

Re: IQ tests results for AI

#237

Post by mrDmrTmrJ »

ok_computer wrote: post #168
jacquesm wrote: post #139

Earlier quoted context omitted.

It used to be that society worked just fine with people of all grades of smarts. But we're rapidly getting to the point that to be able to earn a living wage you need to be above average, especially if you are sole income provider for a while. AI is further steepening that S curve's mid-section.

I think that has more to do with our willingness or ability to value labor in a highly abstracted overseas and automated economy. In addition, there has been a complete disconnect between $1USD purchasing power and generation ability based on capital scale. I don't know what financial crisis or tax policy or free trade agreement or visa program that stems from. I think that in the knowledge worker class, people tend…

"It does feel like a squeeze just functioning in the current - housing, and grocery market though"

If housing were far cheaper and traded just at the cost of new construction. ($250/sq-ft for new build 6 story, $400/sq-ft for 30 story mass timber, $600/sq-ft steel and concrete). We'd see that people can easily live in the current job market!

The fundamental problem in our economy is the artificial scarcity of housing (through local regulation) in the cities and towns where the economy is booming.

Re: IQ tests results for AI

#238

Post by abullinan »

sigmoid10 wrote: post #22

Big caveat here: This website's method doesn't work at all for humans the way it works for LLMs. For humans, there is a strict time limit on these IQ tests (at least in officially recognised settings like Mensa). This kind of sequence completion is mostly a question of how fast your brain can iterate on problems. Being able to solve more questions within the time limit means you get a higher score because your brain…

Mensa really needs to be left out of these discussions. It’s not scientific, it is just a money grab for people who need intellectual validation. You can be admitted with a top 10% SAT score and no in-person testing at all. The in-person testing is in three parts, one part is a memory test, the second part is a Mensa test, the third part is the Weschler test. Source: I joined in 1995 because I needed intellectual validation. :)

Re: IQ tests results for AI

#239

Post by RcouF1uZ4gsC »

gpt5 wrote: post #6

The way human IQ testing developed is that researchers noticed people who excel in one cognitive task tend to do well in others - the “positive manifold.” They then hypothesized a general factor, “g,” to explain this pattern. Early tests (e.g., Binet–Simon; later Stanford–Binet and Wechsler) sampled a wide range of tasks, and researchers used correlations and factor analysis to extract the common component, then norm…

Good point.

There is probably a correlation between how fast a human can do math problems and how intelligent they are in general.

But a very trivial python program running on a normal computer will beat the fastest human at math problems in terms of speed. Even though it does nothing else useful

Re: IQ tests results for AI

#240

Post by Elle_Benjamin »

I realize this question is extremely petty, but it has professional implications for me. Chat GPT-4 gave my dissertation a B+ and GPT-5 gave it an A. The dissertation is in analytic metaphysics. Which model is more reliable for this kind of cognitive task?
Post reply on HN