4 ms·

No, you need intelligent things to be intelligent. And it would probably be good to distinguish between "things that appear superficially intelligent" and "thi

No, you need intelligent things to be intelligent. And it would probably be good to distinguish between "things that appear superficially intelligent" and "things that have genuine capacity for internal experience" while you're at it, since the popular perception on this forum seems to be to just declare that physicalism/materialism is obviously correct and explanatory and we can just brush the hard problem of conciousness under the rug. People are so eager to endow the statistical model that has been designed to produce responses that are statistically feasable in a given context (a very impressive feat of technology to be clear) as having internal experience. Chomsky happens to be more right about LLMs than 90% of the popular discourse will ever be able to understand, which isn't that suprising considering the impact if this "crank's" earlier work.

astrange 2y ago
Chomsky's critique of LLMs was somehow not even correct about his own theories (https://x.com/profraha/status/1634311282135318529 https://x.com/profraha/status/1634311282135318529). And of course has the usual problem of 1. claiming LLMs are random statistical models (they aren't, the sampler is) 2. claiming instead of generating correct text they only "seemingly" generate correct text and not explaining why this is any less impossible. But people have built explicit grammar knowledge into LLMs (it's in llama.cpp) and nobody seems to need it to make English text. Of course the main reason Chomsky is a crank is that he's completely convinced he's right about politics despite his unbroken record of endorsing every mass murderer in the world as long as they're anti-American.
- monadINtop 2y ago
  [flagged]
- optimalsolver 2y ago
  I swear there's a personality type that simply can't resist injecting their politics into unrelated conversations. It's truly fascinating.
  - astrange 2y ago
    I don't think that the opinion "70s intellectuals' love of the genocidal Khmer Rouge is bad" is "my" politics. I think this is a commonly held opinion actually.
    - monadINtop 2y ago
      https://www.jstor.org/stable/25006809 https://www.jstor.org/stable/25006809 https://www.counterpunch.org/2021/09/10/noam-chomsky-and-the-khmer-rouge/ https://www.counterpunch.org/2021/09/10/noam-chomsky-and-the... stop spreading lies that have been debunked for decades, not everyone share your inability to parse information from outside your ideological bubble
int_19h 2y ago
Materialism is rather obviously correct when it comes to consciousness just based on what we know brain damage does to humans. But what does this have to do with problem of consciousness or internal experiences? There's no reason why the latter cannot be a material process. As for the linguists... people forget these days, but before the AI winter, there were many experiments trying to build an AI by modeling the language. And what they learned from it is that when you have enough compute and just throw it at a neural net, you get better results (there was even a joke about AI progress being inversely proportional to the number of linguists working on it). Which, to put it bluntly, means that our understanding of language is still very lacking, and quite likely to be fundamentally wrong some respects. I don't think those people are in a position to judge at that point; not until they figure out language enough to program something that can talk as well as GPT-4 does.
- monadINtop 2y ago
  There is zero understanding of the cause of internal experience. We can't even strictly define what it is. My contention isn't that materialism doesn't appear to be a reasonable canditate for cognitive process, but that we still have no idea HOW any of that happens. We may never know. People who claim that it is obvious that neural networks and LLMs replicate the functioning of a brain, any brain - let alone human, are just wrong. They are wrong to assume that it is obvious that just making bigger LLMs will somehow generate a being with the capacity of internal experience, whatever that may mean. They are wrong when they act like they have solved the problem of understanding cognition by just forgetting to mention the hard problem of conciousness. They are even wrong to simply assert that neurons and neural networks are "to do" with cognition. Hallucinogens and brain damage affect subjective experience, and both of those things are involved with the vascular system of the brain. Would I be right to say the vascular system produce cognition? Why doesn't anyone argue in favor of that view since it is not too far off from the same reasoning? Obviously, it's reasonable to assume that neurons are somehow involved with subjective experience, most serious neuroscientists and other researchers would hold that same assumption. But you need to EXPLAIN how, preferably with evidence. It is the burden of the one claiming they have finally figured it all out to present a convincing theory, or at least conjecture the path to get there. Loudly and smugly asserting that your assumption is correcter than the other people who have spent decades working on the problem doesn't make the problem dissolve, but it is especially irritating when the incentive of the loudest people doing so is in the pockets of the shareholders they are indebted to.

astrange 2y ago

Chomsky's critique of LLMs was somehow not even correct about his own theories (https://x.com/profraha/status/1634311282135318529 https://x.com/profraha/status/1634311282135318529). And of course has the usual problem of 1. claiming LLMs are random statistical models (they aren't, the sampler is) 2. claiming instead of generating correct text they only "seemingly" generate correct text and not explaining why this is any less impossible. But people have built explicit grammar knowledge into LLMs (it's in llama.cpp) and nobody seems to need it to make English text. Of course the main reason Chomsky is a crank is that he's completely convinced he's right about politics despite his unbroken record of endorsing every mass murderer in the world as long as they're anti-American.

monadINtop 2y ago
[flagged]
optimalsolver 2y ago
I swear there's a personality type that simply can't resist injecting their politics into unrelated conversations. It's truly fascinating.
- astrange 2y ago
  I don't think that the opinion "70s intellectuals' love of the genocidal Khmer Rouge is bad" is "my" politics. I think this is a commonly held opinion actually.
  - monadINtop 2y ago
    https://www.jstor.org/stable/25006809 https://www.jstor.org/stable/25006809 https://www.counterpunch.org/2021/09/10/noam-chomsky-and-the-khmer-rouge/ https://www.counterpunch.org/2021/09/10/noam-chomsky-and-the... stop spreading lies that have been debunked for decades, not everyone share your inability to parse information from outside your ideological bubble

int_19h 2y ago

Materialism is rather obviously correct when it comes to consciousness just based on what we know brain damage does to humans. But what does this have to do with problem of consciousness or internal experiences? There's no reason why the latter cannot be a material process. As for the linguists... people forget these days, but before the AI winter, there were many experiments trying to build an AI by modeling the language. And what they learned from it is that when you have enough compute and just throw it at a neural net, you get better results (there was even a joke about AI progress being inversely proportional to the number of linguists working on it). Which, to put it bluntly, means that our understanding of language is still very lacking, and quite likely to be fundamentally wrong some respects. I don't think those people are in a position to judge at that point; not until they figure out language enough to program something that can talk as well as GPT-4 does.

monadINtop 2y ago
There is zero understanding of the cause of internal experience. We can't even strictly define what it is. My contention isn't that materialism doesn't appear to be a reasonable canditate for cognitive process, but that we still have no idea HOW any of that happens. We may never know. People who claim that it is obvious that neural networks and LLMs replicate the functioning of a brain, any brain - let alone human, are just wrong. They are wrong to assume that it is obvious that just making bigger LLMs will somehow generate a being with the capacity of internal experience, whatever that may mean. They are wrong when they act like they have solved the problem of understanding cognition by just forgetting to mention the hard problem of conciousness. They are even wrong to simply assert that neurons and neural networks are "to do" with cognition. Hallucinogens and brain damage affect subjective experience, and both of those things are involved with the vascular system of the brain. Would I be right to say the vascular system produce cognition? Why doesn't anyone argue in favor of that view since it is not too far off from the same reasoning? Obviously, it's reasonable to assume that neurons are somehow involved with subjective experience, most serious neuroscientists and other researchers would hold that same assumption. But you need to EXPLAIN how, preferably with evidence. It is the burden of the one claiming they have finally figured it all out to present a convincing theory, or at least conjecture the path to get there. Loudly and smugly asserting that your assumption is correcter than the other people who have spent decades working on the problem doesn't make the problem dissolve, but it is especially irritating when the incentive of the loudest people doing so is in the pockets of the shareholders they are indebted to.