Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Measurement of IQ seems to be lacking due to it missing the goal of estimating capacity to learn independent of external factors (I.e. either very *shallow* types of diversity [sex, race, etc], or more meaningful types of diversity [language, sensory abscences, etc]).

It's problematic because sometimes IQ tests will require a mix/match of: - Literacy - English language understanding - Extensive Vocabulary - Sight / Hearing

Of course literacy worked around with filling in pictures, but that just shifts us to the requirement of certain sensory input. Requiring English can be worked around by translating to any language, but doesn't address literacy requirements. One big issue I have is seeing vocabulary involved in these tests, which is outright not testing intelligence, but rather experience with a language.

For example, try testing ChatGPT for it's IQ. You'll see very clearly that an IQ test is not made for a system which doesn't have visual capabilities. Perhaps it would be able to from the domain of given the chance - but we don't know because the test is flawed. That is precisely the thing an IQ test should be able to measure - capacity to learn, regardless of the sensory input fed to the system.



Consider applying for YC's Fall 2026 batch! Applications are open till July 27.

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

Search: