5 ms·

GPT4o still can't reason. It is super fancy autocomplete. https://i.imgur.com/z83umbk.jpeg https://i.imgur.com/z83umbk.jpeg Here I change a widely known riddl

GPT4o still can't reason. It is super fancy autocomplete. https://i.imgur.com/z83umbk.jpeg https://i.imgur.com/z83umbk.jpeg Here I change a widely known riddle to the opposite answer, and I manage to make it state them both as the answer.

jordanpg 2y ago
Also, even if this oft-repeated trump card is true, why are you so sure this is different from how our own brains functionally work?
- codexon 2y ago
  Because if this is how our brains really worked, then chatgpt wouldn't be beating most humans at standardized tests and then failing this absurdly easy question that even an elementary school kid could pass. It is simply regurgitating this phrase without even considering that it is stating the exact opposite of the answer it just gave, simply because most answers to this riddle on the internet say this at the end. > This riddle plays on the assumption that a surgeon is typically male, but in this case, the surgeon is the boy's mother. So from this 1 failing, you can see that it is a copy and paste machine, and it doesn't even understand that it is contradicting itself.
  - jordanpg 2y ago
    Your counterexample doesn’t prove that this isn’t how our minds functionally work, except better. No amount of counterexamples could.
Eisenstein 2y ago
'Fancy autocomplete' is a thought-terminating cliche. You can stump a person with a riddle or a logic puzzle or an optical illusion.
- gfourfour 2y ago
  It’s no more thought-terminating than “intelligence,” which is extremely loaded and causes people to make assumptions about these models that work backwards from the “intelligent” label rather than forwards from the tech itself
  - Eisenstein 2y ago
    So? Stop using thought-terminating cliches.
- andy99 2y ago
  "How do you know we're not just next token predictors" is the thought terminating cliche . We know that's what LLMs are. It was certainly eye opening to see how far that gets you. But any deeper claims about intelligence or reasoning need real evidence or at least a proposed line of reasoning. "We don't know how intelligence works so it might be that" doesn't count.
  - Eisenstein 2y ago
    When did I say "How do you know we're not just next token predictors"? The fact is that people who say 'it is just fancy autocomplete' are using a thought-terminating cliche, and a 'I stumped an LLM' proves nothing.
j45 2y ago
It's a bit more than fancy autocomplete. It's very much feels to be gearing towards figuring out the gist of a search engine you may be trying to complete and put together by reading a few links.
macspoofing 2y ago
"Super fancy autocomplete" may not be that different to us, or at least some substantial part of us. Do you think when you speak colloquially with a friend or a colleague, you are engaging in a deep reasoning exercise? When you speak, the next set of words you utter feels like a 'fancy autocomplete' because you don't think through every word, or even the underlying idea or question that was presented - you just know how to respond and with what set of words and sentences.
- codexon 2y ago
  > Do you think when you speak colloquially with a friend or a colleague, you are engaging in a deep reasoning exercise? When I speak colloquially, I have an underlying idea rooted in a world model to be expressed. I don't spit out 1 word at a time based on the previous words I already said.

jordanpg 2y ago

Also, even if this oft-repeated trump card is true, why are you so sure this is different from how our own brains functionally work?

codexon 2y ago
Because if this is how our brains really worked, then chatgpt wouldn't be beating most humans at standardized tests and then failing this absurdly easy question that even an elementary school kid could pass. It is simply regurgitating this phrase without even considering that it is stating the exact opposite of the answer it just gave, simply because most answers to this riddle on the internet say this at the end. > This riddle plays on the assumption that a surgeon is typically male, but in this case, the surgeon is the boy's mother. So from this 1 failing, you can see that it is a copy and paste machine, and it doesn't even understand that it is contradicting itself.
- jordanpg 2y ago
  Your counterexample doesn’t prove that this isn’t how our minds functionally work, except better. No amount of counterexamples could.

Eisenstein 2y ago

'Fancy autocomplete' is a thought-terminating cliche. You can stump a person with a riddle or a logic puzzle or an optical illusion.

gfourfour 2y ago
It’s no more thought-terminating than “intelligence,” which is extremely loaded and causes people to make assumptions about these models that work backwards from the “intelligent” label rather than forwards from the tech itself
- Eisenstein 2y ago
  So? Stop using thought-terminating cliches.
andy99 2y ago
"How do you know we're not just next token predictors" is the thought terminating cliche . We know that's what LLMs are. It was certainly eye opening to see how far that gets you. But any deeper claims about intelligence or reasoning need real evidence or at least a proposed line of reasoning. "We don't know how intelligence works so it might be that" doesn't count.
- Eisenstein 2y ago
  When did I say "How do you know we're not just next token predictors"? The fact is that people who say 'it is just fancy autocomplete' are using a thought-terminating cliche, and a 'I stumped an LLM' proves nothing.

j45 2y ago

It's a bit more than fancy autocomplete. It's very much feels to be gearing towards figuring out the gist of a search engine you may be trying to complete and put together by reading a few links.

macspoofing 2y ago

"Super fancy autocomplete" may not be that different to us, or at least some substantial part of us. Do you think when you speak colloquially with a friend or a colleague, you are engaging in a deep reasoning exercise? When you speak, the next set of words you utter feels like a 'fancy autocomplete' because you don't think through every word, or even the underlying idea or question that was presented - you just know how to respond and with what set of words and sentences.

codexon 2y ago
> Do you think when you speak colloquially with a friend or a colleague, you are engaging in a deep reasoning exercise? When I speak colloquially, I have an underlying idea rooted in a world model to be expressed. I don't spit out 1 word at a time based on the previous words I already said.