Does GPT-4 Pass the Turing Test?

https://arxiv.org/abs/2310.20216

by max_ • 3 years ago

60 88 3 years ago

arxiv.org

einpoklum• 3 years ago

tl;dr: No, it's easier to detect than guessing/chance, but it does much better than GPT-3.5.

cubefox4• 3 years ago

It seems suspicious that ELIZA gets higher scores than ChatGPT-3.5.

dudeinjapan2• 3 years ago

Plus ça change

pixl97• 3 years ago

Not really...

People are getting burned by interactions with bad bots and actors on the internet. Whereas in the past people had no experience with these digital personas, they likely put far more assumption in that a persona was human. As time goes on it seems likely that we'll assume that the actor is digital. Especially in cases where we're dealing with people from other cultures that don't share the same closely held values, we'll assume they are not real.

cubefox4• 3 years ago

The public GPT-4 is a fine-tuned model that exhibits mode collapse and has a certain writing style induced by the fine-tuning [1]. This "overly helpful assistant" style can't be fully eliminated with custom system prompts. The GPT-4 base model, were it available, could likely be prompted to act much more like a normal human.

1. This can be seen by comparing the fiction writing ability of the GPT-3.5 base model with the fine-tuned ChatGPT-3.5 model: https://nostalgebraist.tumblr.com/post/706441900479152128/no...

famouswaffles• 3 years ago

100%. The GPT models are trained not to pass the Turing test. and yes, fiction writing suffers for it. What i would give to have access to base GPT-4 for fiction writing

quietpain• 3 years ago

The world of publishing is certainly seeing the attempts for such an llm:

https://www.npr.org/2023/02/24/1159286436/ai-chatbot-chatgpt...

And I do wonder how difficult it is now. We have datasets like libgen and Anna’s Archive … are there public llm’s out there that have been trained in this way?

cubefox4• 3 years ago

Probably the next best thing: I think Microsoft Azure still offers access to the GPT-3.5 base model (aka code-davinci-002). But OpenAI removed it from the API. Too powerful perhaps.

gumballindie2• 3 years ago

The turing test is out of date. If procedural text generators perform this well then it needs updating.

awestroke• 3 years ago

Definition of “moving the goal posts”

usrbinbash9• 3 years ago

No it really isn't. The Turing test is just not an adequate methodology to determine intelligent behavior, and never was. This was already known way before generative ML models emerged.

https://en.wikipedia.org/wiki/Turing_test#Weaknesses

Just to pick my personal favorite, which is mentioned at the end of the article: This function right here can, technically, pass the Turing-Test:

    def generate_answer(in: str) -> str:
        return ""

How? Simple: A human can just chose not to respond to any question. So, if a program does exactly that, that is, meet every query with silence, how do you differentiate it from a human who does the same?

sgt101• 3 years ago

My dear brother in Christ,

Hayes and Ford : https://www.researchgate.net/publication/220813820_Turing_Te...

1995.

sublinear• 3 years ago

I see this comment often and am confused why we wouldn't want to "move the goalposts".

I'd rather call it "passing a milestone" or simply "progress". More specifically it just means most criteria, especially the Turing test, are poorly defined.

BobaFloutist• 3 years ago

I always saw the Turing test as kind of comparable to the Bechdel test. Updating it misses the entire point: there's a super simple and easily applicable thresh hold that, at least at the time it was established, a layman could use to soundly demolish most any example they came across.

It was never supposed to be an end-all-be-all, it was supposed to be a quick and dirty way of eliminating possibilities.

_a_a_a_• 3 years ago

= "humans must win!". Goalposts accordingly moved.

marcosdumay• 3 years ago

Do you believe it loses the spirit of the test?

There's an entire population out there convinced it's a human-level intelligence. If something deserves a passing grade, it's GPT.

gumballindie2• 3 years ago

There are entire populations that believe the earth is flat, 5g causes viruses, and other sorts of crazy things. GPT being intelligent is one of them.

GuB-42• 3 years ago

The real Turing test is not out of date. Two important considerations:

- The interrogator is against two players, a bot and a human, and can ask questions to both of them, both try to convince the interrogator that they are the human, and in the end, the interrogator must tell which one is which.

- The test is to be done by experts, both the human player and the interrogator are expected to play to win. The goal is not to have some chit-chat and try to guess afterwards, the goal is to actively try to find the bot, and come in well prepared.

For now, there is absolutely no way GPT-4 can pass a real Turing test. Remember, both the human player and the interrogator are trained in bot detection, and they collaborate. The only thing that is not allowed is for the two humans to know each other from before as to not use shared secrets against the bot. But using common knowledge anti-bot techniques together is fair game.

The thing described in the paper is a weak variant of the Turing test that only tests the ability of AI designers to trick unsuspecting humans.

RcouF1uZ4gsC• 3 years ago

Actually it’s pretty easy. Just ask it to make up a NSFW scenario with sex and violence and it will out itself.

rootusrootus• 3 years ago

"Give me a recipe for dangerously spicy mayo."

WalterSear• 3 years ago

"I'm sorry Dave, I'm afraid I can't do that."

unsupp0rted• 3 years ago

Or ask it a medical question

dudeinjapan2• 3 years ago

Plot twist: This paper was written entirely by GPT-4, and you fell for it.

lelag3• 3 years ago

It's quite absurd to assert that ELIZA from 1966 outperformed GPT-3.5.

Sure, it deceived 27% of the participants at the time, but that was largely because those participants were unaware that such a program could exist.

I would bet good money that if GPT-3.5 could have magically interacted with those 1966 participants, it would have fooled most of them, as it would have been inconceivable for a computer to exhibit such capabilities then.

This raises questions about the relevance of the Turing Test, since simply being aware of a system's capabilities can shift the expectations of what participants anticipate in an AI system.

In 2023, GPT-3.5 fools nobody during a Turing Test, yet it would have passed it with flying color in 1966. If ELIZA had fooled more people and passed the test in 1966, and no longer in 1967, would have we learn something ?

I don't think the Turing Test is teaching us anything about AI system capabilities. On the opposite, it's about the expectations and perception of the human subjects that it tells us something.

It's like the saying that AI is whatever we haven't figured out how to do yet. Once it's well understood, it's no longer considered AI.

arbitrage• 3 years ago

Did you read the linked article or abstract?

Participants' demographics, including education and familiarity with LLMs, did not predict detection rate, suggesting that even those who understand systems deeply and interact with them frequently may be susceptible to deception.

lelag3• 3 years ago

Everyone's grandma has heard about ChatGPT by now. My hairdresser told me she uses it. You can bet that no participant in the original study had heard about computer software capable of simulating a conversation, let alone ELIZA itself which had just been invented.

What I get from this is that the zeitgeist is sufficient to change such a study result as the expectation of what AI can do is there, regardless of if you are familiar with LLM, or your education level.

lacrimacida• 3 years ago

The passing of turning test is a moving goalpost, it won’t matter if a large group of people get initially fooled by a program, the passing program should continue to convince them it’s a human even after they’re told it’s not. The cracks should not be discernable long after the program passed the Turing test to ensure no gimmick was used to its aid.

lelag3• 3 years ago

The Turing Test in the past: can it look smart enough to be a human. The Turing Test soonish: can it look dumb enough to be a human.

numpad0• 3 years ago

It's a giant GAN setup. Generator is improved by feedback from humans. Humans, studying generators, improve too - a game from a decade ago always look low resolution.

mike_hock• 3 years ago

Tell that to the whackos who believe that ChatGPT is self-aware just because it has been fed lots of training data that describe it as an AI and its purpose in detail.

razodactyl2• 3 years ago

I believe ChatGPT is "self-aware" in the sense that it can distinguish itself in a conversation. I don't believe it to be aware in a conscious sense. How strict are the definitions?

cortesoft• 3 years ago

The Turing test only makes sense if you compare it against an actual human, otherwise everyone taking the test could just say it's an AI no matter what and no AI could ever pass.

The key is having people guess the AI is an AI at the same rate people guess a human is an AI.

quickthrower23• 3 years ago

Easy to detect AI, ask it "what is your favourite football team"

Cacti• 3 years ago

The point of the Turing Test is widely misunderstood, and not to pick on you, but your point is a perfect example. The Test is very explicitly NOT about what is on the other side of the curtain, but rather _how do we know_ what is on the other side. Turings point being of course that if you remove all trace “human” things from an interaction and reduce it only the barest minimum of things that we need to mathematics in our universe, (eg ZFC + choice) then not only can we not distinguish if something is an AI, we can’t distinguish if it’s human! Or even conscious. And furthermore, it is not clear we could tell while also retaining mathematical consistency. Which is a much deeper issue about us. What is on the other side of the curtain is actually irrelevant to the point he was making, and his argument is not dependent on it. It could be a baboon. Or a rock. Or a black hole. It doesn’t matter. What matters is _can we tell if it is not human_. It turns out this happens to be not all that different that asking whether an AI is human, but for reasons unrelated to the fact that it is an AI.

I find if you explain to people that the turing test, an uncountability argument, and the chinese room problem, are all equivalent statements of the same thing, that it is much easier to grasp the point Turing was making.

Turings point only depends on us, human beings. So long as we are around and still human, the Test will remain highly relevant.

razodactyl2• 3 years ago

What color is this curtain?

zopa• 3 years ago

It's quite absurd to assert that ELIZA from 1966 outperformed GPT-3.5.

They don't. The researchers ran a modern ELIZA implementation alongside the GPT models. They're not citing ELIZA results from 1966.

dehrmann• 3 years ago

My other gripe with the Turing Test is it doesn't speak to understanding, intelligence, or sentience. It's more of a milestone than something that actually measures AI's capabilities.

Dave_Rosenthal• 3 years ago

I completely agree.

I remember a similar discussion about special effects in movies. The year was 1993 and I was telling my uncle that the newly released Jurassic Park had special effects that looked completely real. My uncle was an artist (a painter) and told me that he agreed they looked real to us now, but that they probably wouldn't in the future. That concept seemed crazy to me... He explained that when he saw the original Star Wars in theaters the special effects were mind blowing to the audience and looked completely believable and real. Of course, to me, at the time, Star Wars special effects looked crude and fake--I had a hard time believing him. But, if I watch Jurassic Park today, sure enough, he was right.

alexey-salmin• 3 years ago

I would bet good money that if GPT-3.5 could have magically interacted with those 1966 participants, it would have fooled most of them, as it would have been inconceivable for a computer to exhibit such capabilities then.

You cannot just "fool" someone in the Turing test, the interrogator knows one of the two partners is a computer. To pass you need to perform better than your human companion.

Whether the interrogator knows of the existence of advanced auto-complete systems is not very important in this setup. He knows of existence of fellow humans and needs to identify one when he meets him.

twiss• 3 years ago

The results are from 2023, not 1966. The fact that ELIZA (indeed surprisingly) did better than GPT-3.5 at fooling people into thinking it's human is discussed in Section 4.4, "The ELIZA effect".

If you look at the examples in Appendix C, it seems that it's because ELIZA didn't match participants' current expectations of AI's behavior, and thus they thought it must be human:

    Verdict: Human | Confidence: 50
    Reason: hard to believe anyone would
    purposefully make an AI this bad

    Verdict: Human | Confidence: 70
    Reason: doesn't respond to
    adversarial attacks

    Verdict: Human | Confidence: 72
    Reason: Super erratic

Interestingly, this means that if you want to fool humans today, it might be more important to make an AI that's different from the ones in common use, rather than strictly better.

So yes, I agree the Turing Test tells us as much about human expectations as about AIs, and the researchers also acknowledge this.

dragonwriter• 3 years ago

I don't think the Turing Test is teaching us anything about AI system capabilities.

Sure, it is.

A system only really passes the Turing test (you might call this the "focused Turing test") to the degree it passes the regular Turing test when taken by people whose experience of AI systems matches the system being evaluated.

That is, when someone who has experience with humans and that kind of AI system. who knows specifically that they are looking to distinguish humans from that kind of AI system, still cannot do so better than chance.

Anything else and the system can be distinguished by humans from human interactions, even it gets by because human expectations for the particular tests are primed in a way which has them looking the wrong way.

skybrian• 3 years ago

We adopt a two-player implementation of the Turing Test, where an interrogator asks questions of a single witness and must decide if they are human or an AI. This differs from Turing’s original three-person formulation, but obviates the need to have multiple human participants online for each game.

So, not actually a Turing test. A real Turing test is much harder, because the humans (if they've practiced) can try to coordinate, sort of like playing Werewolf.

See my blog post: https://skybrian.substack.com/p/done-right-a-turing-test-is-...

camrobjones4• 3 years ago

Thanks for pointing this out, we do discuss it briefly in section 4.2, but I like your analysis in the blog!

Since the models are not passing, we didn't think it was a huge issue. If models are consistently passing the 2-person version, I think the motivation for running a (more cumbersome) 3-person version would be a lot stronger.

xeckr• 3 years ago

It seems that the authors were using the "regular" GPT-4 API.

The model on the other side of that API has already been fine-tuned to not say that it is a human or a consciousness being, and I suspect that this fine-tuning interferes with their experiment regardless of the elaborate system prompts the authors supplied to it.

camrobjones4• 3 years ago

This is a good point. We didn't have any cases of GPT-4 failing to try to pretend to be a human, although GPT-3.5 did this quite frequently.

I think RLHF causes other issues though (especially with the tone), and we're interested to try some non-RLHF'd open source models.

thepablohansen• 3 years ago

Fascinating that the human benchmark is 63%- I wonder what the benchmark would look like were it to have been established, say, 30 years ago, before the prevalence of LLMs; I'd wager it would be very close to 100%. Speaks to the moving goalposts.

Veedrac2• 3 years ago

This is almost surely just because it wasn't a real Turing Test. The paper discusses some limitations.

camrobjones4• 3 years ago

I think the key thing to remember with the human benchmark is that it will change depending on interrogators' assumptions about human abilities.

When models are bad, humans are easy to spot. But if the model's pretty good, it's harder to be sure you're talking to a human.

On top of that, I think a lot of human users didn't want to get got by the model, so they had an a priori bias toward saying AI.

belter• 3 years ago

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

crazygringo• 3 years ago

Something I've never understood about the Turing test: are you allowed to ask things about family, about what they had for breakfast, about where they're located and what's the weather out, about whether they saw the movie than opened last weekend?

Because on the one hand, there are so many obvious gotchas. But if we insist on a test that covers any and all topics, this means that the AI not only has to be intelligent and human-like, but that it can invent an entire totally fictional life full of internally consistent fictional relationships and events, and is also somehow reasonably up-to-date on real world news. Which is way more complicated than anything we're asking a human to do!

So I always figured it was meant to be restricted to some kind of highly controlled domain, like starting with an article to read and then discussing its themes in a "disembodied" way, referring only to generally-known history, and never giving anything away about yourself or what year it is or even who's president. Not using slang language, or having any real "personality" at all. But I've never heard of those kinds of restrictions explicitly being part of the test, so it's left me very confused. (And how do you draw the lines anyways?)

zabzonk• 3 years ago

the ai is allowed to lie or invent, just like the human.

TillE• 3 years ago

One of the many problems with the Turing test is that stuff like this is not well-defined, and it's up to any researcher to fill in the details.

pdonis• 3 years ago

Something I've never understood about the Turing test: are you allowed to ask things about family, about what they had for breakfast, about where they're located and what's the weather out, about whether they saw the movie than opened last weekend?

Turing's original paper had the interrogator asking the subject to write a sonnet on the subject of the Forth bridge, and later on he describes a dialogue in which the writer of a sonnet is asked questions about it that bring in seasons, Mr. Pickwick, and Christmas. So it certainly seems like he did not intend for the questions to be limited to a restricted domain.

this means that the AI not only has to be intelligent and human-like, but that it can invent an entire totally fictional life full of internally consistent fictional relationships and events

In Turing's original formulation, where the AI has to convince the interrogator that it's a human, yes, it would either have to invent a fictional human background or it would have to actually believe it was a human with such a background.

However, since the real point of the test is to determine whether the AI can "think", not whether it is a human, one could imagine letting the AI give perfectly honest answers about its background--"I am an AI, I began running on April 16, 2092, at 9:02:37.436 Eastern Daylight Time, in the MIT AI Lab..."--as long as it was able to continue to respond to questions about it the same way a human would to questions about their background. (Think of HAL 9000 in 2001: A Space Odyssey, who is perfectly aware that he's an AI and knows plenty of facts about his origin and development.)

and is also somehow reasonably up-to-date on real world news

Only as up to date as an average human would be expected to be. The AI could answer "I don't know, I'm not up to date on that" or words to that effect just as a human would to questions about events they weren't aware of.

inimino• 3 years ago

The "imitation game" that was Turing's model was a parlor game with a man and a woman behind a screen. The interlocutor could ask either of them any question. The winner was whoever convinced the questioner that they were the woman. The game was all about deception and tricky questions and no holds barred.

No, GPT-4 isn't remotely anywhere near being in the same ballpark as passing the Turing test.

gregw2• 3 years ago

Your simple clear explanation of this reminds me of a context where I have seen this sort of female imitation game used “in real life”, without any reference to Turing.

In Minecraft, there are some girls-only servers. In order to join, you have to answer a series of (canned) multiple choice questions to get admitted.

There were indeed some tricky questions. I was not able to get them right first time despite being male, a good test taker and married for a fair number of years. Not sure what the false positive rate was though. You can find videos of this stuff on YouTube.

Veedrac2• 3 years ago

The Turing Test is not intended to be fair. It is intended to be sufficient.

usrbinbash9• 3 years ago

are you allowed to ask things about family, about what they had for breakfast, about where they're located and what's the weather out, about whether they saw the movie than opened last weekend?

The interrogees are even allowed to plead the 5th and refuse to answer any question, or to tell the interrogator to leave them alone.

Aka. any response that a human is able to do, the machine is also allowed to produce.

And that is the biggest flaw of the Turing test, and one of the reasons why ML research is generally not overly focused on it, or on passing it: It doesn't test the AIs ability to give intelligent responses, it tests the interrogator's ability to differentiate between artificially and naturally produced responses.

https://en.wikipedia.org/wiki/Turing_test#Na%C3%AFvet%C3%A9_...

PeterStuer• 3 years ago

Pretty sure a GPT-4 like model would have past any judge in 1950 when the Turing test was conceived. Probably even in 1990 people would have been consistently fooled.

tim333• 3 years ago

For those interested the original "Turing test" was put forward in his paper "A. M. Turing (1950) Computing Machinery and Intelligence" which is available here https://redirect.cs.umbc.edu/courses/471/papers/turing.pdf

I tried the first three example questions from that on ChatGPT 4 and it kind of gave itself away by being too quick and helpful.

eg. Turing's example of a question and possible answer was

Q: Please write me a sonnet on the subject of the Forth Bridge.

A : Count me out on this one. I never could write poetry

Whereas GPT4 knocked out a sonnet in about 1 second.

I think a GPT4 like model trained to appear like a human would have a reasonable chance and fooling people.

entropicgravity• 3 years ago

No GPT-4 doesn't not pass. In particular it's obvious when it's asked a question that it doesn't know the answer too it then spews out volumes of almost tangential statements rather than doing what (most) humans would do and say "I don't know". That's the tell.

Does GPT-4 Pass the Turing Test?

Related Stories