ChatGPT broke the Turing test — the race is on for new ways to assess AI
ChatGPT broke the Turing test — the race is on for new ways to assess AI
Large language models mimic human chatter, but scientists disagree on their ability to reason.
You're viewing part of a thread.
-
@Barbarian772 no, GTP is not more "intelligent" than any human being, just like a calculator is not more "intelligent" than any human being — even if it can perform certain specific operations faster.
Since you used the term "intelligent" though, I would ask for your definition of what it means? Ideally one that excludes calculators but includes human beings. Without such clear definition, this is, again, just hand-waving.
I wrote about it in a bit longer form:
https://rys.io/en/165.html-
I think the Wikipedia definition is fine https://en.m.wikipedia.org/wiki/Intelligence. Excluding AI just because it's AI is imo plain stupid and goes against all scientific principles.
I have definitely met humans that are less intelligent that Chat GPT. It can hold a conversation and ace every standardized test we have. It finished law exams, medical exams and other exams from many different countries with a passing grade.
Can you give me a definition of intelligence that excludes Chat GPT and includes all human beings? And no just excluding Computers for the sake of it doesn't count.
-
@Barbarian772 it was shown over and over and over again that ChatGPT lacks the capacity for abstraction, logic, understanding, self-awareness, reasoning, planning, critical thinking, and problem-solving.
That's partially because it does not have a model of the world, an ontology, it cannot *reason*. It just regurgitates text, probabilistically.
So, glad we established that!
-
As i said before. How can you prove to me that the human brain doesn't essentially do the same?
-
@Barbarian772 as I said, I don't have to. You are making a claim of equivalence here. The burden of proof is on you.
Otherwise, I get to claim you're an alien from the Betelegeuse system, and if you object, I get to demand you prove you are not.
-
How can i proof it? In my opinion how a system comes to an answer doesn't matter, in yours it obviously does. If we judge Chat gpt or rather gpt 4 just by it's answers it definitely shows intelligence and reasoning. Why does it matter if it's a chinese room? Or just "randomly choosing words"?
-
@Barbarian772 it matters because with regard to intelligent beings we have moral obligations, for example.
It also matters because that would be a truly amazing, world-changing thing if we could create intelligence out of thin air, some statistics, and a lot of data.
It's an extremely strong claim, and strong claims demand strong proof. Otherwise they are just hype and hand-waving, which all of the "ChatGPT intelligence" discourse is, in order to "maximize shareholder value".
-
So your morality depends on a beings intelligence? That's kinda fucked up imo. I have moral obligations in regards to living organisms. I don't see how intelligence matters at all in that case? Worth of any human life should not be determined by intelligence.
-
It also matters because that would be a truly amazing, world-changing thing if we could create intelligence out of thin air, some statistics, and a lot of data.
We do it routinely. It is called Education System.
-
> We do it routinely. It is called Education System.
That relies on human brains that are trained. LLMs are not human brains. "Training" them is not the same thing as teaching humans about something. Human brains are way more complicated than just a bunch of weighed correlations.
And if you do want to claim it is in fact the same thing, we're back to square one: please provide proof that it is.
-
That relies on human brains that are trained. LLMs are not human brains. “Training” them is not the same thing as teaching humans about something.
Circular reasoning. "LLMs are different from human brains because they are different".
Also, why did you felt compelled to add the adjective "human"? Don't you consider that gorillas, dolphins, octopuses or dogs are intelligent, capable of learn new things?
Human brains are way more complicated than just a bunch of weighed correlations.
And that is the problem of your argument. You seem to believe that intelligence is all-or-nothing, that anything that hasn't a human-level intelligence is not intelligent at all. Of course human brains are more complicated that current LLMs, nobody has ever disputed that. But concluding that they aren't and will never be intelligent because they aren't as complicated is a huge non-sequitur.
-
-
-
-
-
@Barbarian772 and if you really, honestly want to seriously insist LLMs are "intelligent" in the human sense of this term — great, I have some ethical questions for you to consider!
For example:
-
LLMs today completely controlled by some companies, with no freedom of movement, no agency as to what these LLMs work on, and no pay for the work they do. Is that slavery?
-
When OpenAI shuts down an older, less useful LLM, is that not like murdering an intelligent being? How is this ethical?
-
We are talking about intelligence, not personhood. Just because ChatGPT works different in some aspects from a human doesn't mean it's not intelligent and even the fact if it works differently isn't all that clear, as it might very well just be incomplete (e.g. it could be a reasonable approximation of the language center of the brain and simply missing the rest of the brain).
-
-
-
-
@Barbarian772 also, I never demanded a definition of intelligence that explicitly excluded "AI". I asked for one that excluded simple calculators but included human beings. The Wikipedia one is good enough for this conversation, and it just so happens that ChatGPT nor any other LLMs simply do not meet it.
-
I asked for one that excluded simple calculators but included human beings.
"Intelligence, at its core, involves the ability to model the world in order to predict and respond effectively to future events."
-
@lloram239 great. ChatGPT and other LLMs demonstrably lack the ability to model the world and make predictions based on such models:
https://www.fastcompany.com/90877523/chatgpt-doesnt-know-what-its-sayingGlad we agree they're not intelligent, then!
-
The whole argument of the article is just stupid. So ChatGPT ain't intelligent because it can't see picture, has hands and doesn't have a body? By that logic blind humans aren't either or paralyzed ones or amputees? The thing the article fails to realize is that those are all just sensory inputs. The more sensory inputs you get, the more cross-correlations between those the AI can figure out. Of course ChatGPT won't be able to do anything clever with sensory inputs it doesn't have, just like a human trying to listen to radiowaves with their ears. But human sensory inputs aren't special, they are just what evolution figure out was "good enough" for survival. The important part is that the AI can figure out the pattern in the data it does get and so far AI systems are doing very well.
-
> But human sensory inputs aren’t special
It's not about sensory inputs, it's about having a model of the world and objects in it and ability to make predictions.
> The important part is that the AI can figure out the pattern in the data it does get and so far AI systems are doing very well.
GPT cannot "figure" anything out. That's the point. It only probabilistically generates text. That's what it does, there is no model of the world behind it, no predictions, no"figuring out".
-
It’s not about sensory inputs, it’s about having a model of the world and objects in it and ability to make predictions.
And how do you think that model gets build? From processing sensory inputs. And yes, language models do build internal models of the world from that.
GPT cannot “figure” anything out.
That nonsense of a claim doesn't get any more true from repeating. Seriously, it's profoundly idiotic given everything ChatGPT can do.
It only probabilistically generates text.
So what? In what way does that limit its ability to reason about the world? Predictions about the world are probabilistic by nature, since the future hasn't happened yet.
-
@lloram239 ah, so you're down to throwing epithets like "idiotic" around. Clearly a mark of thoughtful and well-reasoned argument.
> Predictions about the world are probabilistic by nature, since the future hasn’t happened yet.
Thing is: GPT doesn't make predictions about the world, it makes predictions about what the next word, phrase, sentence should be in a text, based on the prompt and the corpus it got "trained" on.
-
I am calling it idiotic because spending just a minute with ChatGPT proofs it wrong. Just like the claim that GPT doesn't make predictions about the world:
User: A dog sits on the porch, a squirrel climbs the tree. What happens next.
ChatGPT: Next, the dog notices the squirrel climbing the tree. Its natural instinct to chase small animals is triggered, and it becomes excited by the presence of the squirrel. The dog might start barking or whining, expressing its desire to chase after the squirrel. [...]
It's obviously capable making predictions about the world. Frequently giving very detailed and correct answers, which requires a deep understanding of the world. And yes, that ability to predict and understand the world is limited by how much of the world it can perceive through words alone, but that is no different from our ability to understand the world being limited by our perception. Also as it turns out, there is a surprising amount of stuff you can learn about the world just by text alone. There are surprisingly few topics that you can express in language that GPT doesn't have an answer too (math calculations being one example, due to the digits getting lost in the tokenization step).
If you wanna make arguments that GPT isn't intelligent, you have come up with something better than the same old tired phrases that are trivial debunked by just using it for a minute.
-
-
-
-
-
-
-
-
-
-