People cannot distinguish GPT-4 from a human in a Turing test · HackerPapers discussions

People cannot distinguish GPT-4 from a human in a Turing test — mentioned by meinersbur

Mon, 04 May 2026 02:40:00 GMT

Since ELIZA, the more we try to apply the Turing Test, the more evidence we get that the Turing Test about human gullibility, rather than saying anything meaningful about the machine [1,2]. Surprisingly it turns out that even critical thinker Richard Dawkins is quite gullible. [1] https://medium.com/@innovariart/the-turing-test-no-longer-me... [2] https://arxiv.org/abs/2405.08007

Paper and discussion history. A mention is not an endorsement.

People cannot distinguish GPT-4 from a human in a Turing test — mentioned by stared

Sat, 07 Mar 2026 11:06:10 GMT

It is interesting that now to pretend you are a human, you need to play dumb. Vide superstimuli for AI (maths and philosophy) and humans (clickbaits and scams) https://x.com/repligate/status/1830331774875893925 or a prompt for "People cannot distinguish GPT-4 from a human in a Turing test" https://arxiv.org/abs/2405.08007.

Paper and discussion history. A mention is not an endorsement.

People cannot distinguish GPT-4 from a human in a Turing test — mentioned by coldtea

Tue, 10 Feb 2026 23:30:46 GMT

>Not even close. "Neural networks" in code are nothing like real neurons in real biology Hence the simplified. The weights encoding learning and inteconnectedness and nonlinear activation and distributed representation of knowledge is already an approximation, even if the human architecture is different and more elaborate. Whether the omitted parts are essential or not, is debatable. “Equations of motion are nothing like real planets" either, but they capture enough to predict and model their motion. >The "turing test" doesn't exist. Turing talked about a thought experiment in the very early…

Paper and discussion history. A mention is not an endorsement.

People cannot distinguish GPT-4 from a human in a Turing test — mentioned by acdha

Mon, 27 Oct 2025 01:44:38 GMT

This is the best one I’ve seen but it has the notable caveat that it’s a relatively short 5 minute chat session: https://arxiv.org/pdf/2405.08007 I do think we’re going to see this shift as AI systems become more commonplace and people become more practiced at recognizing the distinction between polished text and understanding.

Paper and discussion history. A mention is not an endorsement.

People cannot distinguish GPT-4 from a human in a Turing test — mentioned by falcor84

Thu, 31 Jul 2025 13:48:31 GMT

As far as I understand, Turing himself did not specify a duration, but here's an example paper that ran a randomized study on (the old) GPT 4 with a 5 minute duration, and the AI passed with flying colors - https://arxiv.org/abs/2405.08007 From my experience, AI has significantly improved since, and I expect that ChatGPT o3 or Claude 4 Opus would pass a 30 minute test.

Paper and discussion history. A mention is not an endorsement.

People cannot distinguish GPT-4 from a human in a Turing test — mentioned by tripletao

Wed, 02 Apr 2025 16:59:04 GMT

This appears to be the same two authors who reported that "People cannot distinguish GPT-4 from a human in a Turing test" back in May 2024: https://arxiv.org/pdf/2405.08007 That earlier result was because they botched the statistics, changing the test so it's no longer a binary comparison but still analyzing as if it was. They seem to have fixed that now, perhaps in response to reviewer feedback. This new preprint is the best LLM Turing test I've seen so far. That said, their humans sure don't seem to be trying very hard. The most effective interrogator strategies ("jailbreak" and "strange")…

Paper and discussion history. A mention is not an endorsement.

People cannot distinguish GPT-4 from a human in a Turing test — mentioned by FergusArgyll

Sun, 09 Feb 2025 11:45:01 GMT

The sys prompt given to run the turing test (from https://arxiv.org/pdf/2405.08007) actually works well. I'm honestly not sure I'd be able to tell (unless I test it adversarially e.g ignore the prompt, write a poem etc)

Paper and discussion history. A mention is not an endorsement.

People cannot distinguish GPT-4 from a human in a Turing test — mentioned by tripletao

Tue, 07 Jan 2025 02:55:03 GMT

> The setup Turing describes isn't the "both must sum to 100%" setup you're presenting. He has two different games being played [...] Turing introduces the game as a man imitating a woman, then modifies it to a machine imitating a human. In both of those games, the interrogator makes a binary choice between two witnesses, one real and one imitating. So P(imitator-judged-real) and P(real-judged-real) sum to 100% in both games. So in both games, a score of 50% means the imitator and the real witness are indistinguishable. I believe that's the reason why a score of 50% is treated as significant.…

Paper and discussion history. A mention is not an endorsement.

People cannot distinguish GPT-4 from a human in a Turing test — mentioned by robkop

Thu, 02 Jan 2025 22:33:45 GMT

Ahh, there's a bug with the z-index on the Turing one (I made it a "legendary card" for surviving 70 year), will fix shortly. Here's the link for the moment: https://arxiv.org/pdf/2405.08007 Also if you want to read the original Turing paper (it was interesting to look back upon, I think the future of benchmarks may look a lot more like the Turing test): https://courses.cs.umbc.edu/471/papers/turing.pdf

Paper and discussion history. A mention is not an endorsement.

People cannot distinguish GPT-4 from a human in a Turing test — mentioned by wat10000

Mon, 23 Dec 2024 15:13:42 GMT

Google must be really going downhill. DDG “gpt turing test” provides nothing but relevant links. Here’s a paper: https://arxiv.org/pdf/2405.08007

Paper and discussion history. A mention is not an endorsement.

People cannot distinguish GPT-4 from a human in a Turing test — mentioned by ben_w

Mon, 30 Sep 2024 19:51:56 GMT

Capabilities vs. "is this a good idea?" — seems like you're saying two things, one of which I strongly agree with and the other I strongly disagree with. On "should we": I agree completely. I've previously made a similar example with sociopaths, when people are incredulous as to the risks of an un-aligned AI. I've also previously averred that the first time we attempt a brain upload, we're likely to know so little about how our minds function that the upload will have every possible mental health disorder simultaneously, including many that we've got no words for because they're incompatible…

Paper and discussion history. A mention is not an endorsement.

People cannot distinguish GPT-4 from a human in a Turing test — mentioned by ben_w

Sat, 17 Aug 2024 17:37:48 GMT

> Noticed how everyone dropped the Turing Test like a hot potato as the gold standard for intelligence, the moment it became apparent that LLMs were about to pass it? Try to find a recent high-profile article invoking the Turing Test. Crickets. The intellectual dishonesty is nauseating. Not quite that bad, but I hear you. https://arxiv.org/html/2405.08007v1

Paper and discussion history. A mention is not an endorsement.

People cannot distinguish GPT-4 from a human in a Turing test — HN submission

Thu, 27 Jun 2024 16:15:48 GMT

People cannot distinguish GPT-4 from a human in a Turing test

1 point; 2 discussion replies on HN. Paper and source history. A submission is not an endorsement.

People cannot distinguish GPT-4 from a human in a Turing test — HN submission

Mon, 20 May 2024 15:09:13 GMT

People cannot distinguish GPT-4 from a human in a Turing test

3 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.