logoalt Hacker News

jgalt212yesterday at 1:34 PM11 repliesview on HN

when was the turing test beaten?


Replies

simonhyesterday at 1:42 PM

1966

https://en.wikipedia.org/wiki/ELIZA_effect

It turns out the limiting factor isn't how sophisticated algorithms are, it's how gullible humans are.

show 4 replies
Someone1234yesterday at 1:39 PM

The EU just had to pass a law to force companies to disclose if a customer service agent is AI or Human. It is beaten.

https://commission.europa.eu/news-and-media/news/safer-and-m...

show 3 replies
bnchrchyesterday at 1:39 PM

According to Psychology today, April this year by GPT 4.5

https://www.psychologytoday.com/ca/blog/the-digital-self/202...

show 1 reply
mdp2021yesterday at 1:55 PM

The Turing test is more complex than what gets suggested.

And the "popularized" version is faulty also since it uses an ideal, abstract human judge (like the "spheroidal economic agent").

But if you want to add declinations to the said popularized image of the Turing test, you may add Maxim Lott's IQ tests at trackingai.org . Between the end of 2024 and the beginning of 2025 LLMs reached an equivalent IQ of 100, for example.

goda90yesterday at 1:36 PM

I think it was determined that the Turing test is too easy because humans are too easily fooled.

show 1 reply
3836293648yesterday at 1:41 PM

~1960

ELIZA beat the Turing test and then everyone forgot about it. Humans are just really terrible at recognising robots.

mdp2021yesterday at 2:11 PM

Or can we reframe it: when did humans start losing the (so-called) "Turing test".

I think there are elements showing lowering of performance and expectation.

shricyesterday at 1:43 PM

It depends who takes the test. I am not yet, to my knowledge, fooled by AI.

I've tried [1] and I almost 100% detect which is the AI. I really want to convince myself I have failed, does anyone know of a better site/resource for this?

I know it might be moving goalposts but I would consider AI to have passed in a well and truly undisputed manner when [2] is resolved.

But in a more practical sense, if AI can impersonate humans so well today then why are state of the art frontier models so obviously AI when they create PRs, commit messages, documentation, etc. Are the companies deliberately making them unnatural?

[1] https://turingtest.live/

[2] https://www.metaculus.com/questions/11861/date-when-ai-passe...

show 2 replies
intrasightyesterday at 1:36 PM

2050 I'm guessing

Rover222yesterday at 3:54 PM

are you living under a rock?