It’s going to be tough to explore this through internet comments, but that just raises the question of “what do you mean by thought and intelligence?”, which then turns into “what do you mean by understanding?” and lots of other similar questions, down a deep rabbit hole. I don’t think it’s really possible to make strong statements either way until we’ve come up with a more coherent theory underlying basic terms like that. I’d love to see a rigid and objective definition that we can measure LLMs against.
I think you’re fishing for some philosophical discussion/redefinition of terms, but I’m not. If you are saying that any algorithm that makes a decision based on criteria is “thinking” or “Intelligent”, then sure, so are LLMs. That definition isn’t one that is accepted by almost anyone, but let’s say that it’s true. In that case, the autocomplete on your phone’s keyboard is thinking and intelligent, so is a random number generator, and so is even a basic if/else statement in programming.
When people discuss “thought” or “intelligence” typically the core definition comes from awareness. Awareness of the decisions that a being is making. LLMs do not have the capability of awareness. It doesn’t understand why it makes the decisions it does, it has no clue if it has created an answer out of thin air or not. It isn’t even aware that it gave an answer. When asked to explain its rationale, it will often contradict the conclusion it came to, because there is no awareness of a decision even being made, it has simply been trained on which words follow which words in an explanation, and strings them together.
I’m fishing for any useful definition at all. What does “awareness” mean? One definition is “conscious knowledge”, and you look up “conscious” to get “Characterized by or having an awareness of one’s environment and one’s own existence, sensations, and thoughts”. It’s circular! So I don’t care about what most people think, because most people can’t define the terms in any useful way. And this is why I dislike almost all discussion about AI, because people have strong opinions without ever considering the basic definitions of the words they’re using (not limited to any one group).
You bring up a really good example, and that’s actually exactly where I’ve ended up after thinking about it. Why isn’t an if/else statement the basic “atom” of intelligence, just like a bit is the basic atom of information? Sure, it’s not “intelligent” in the same way a human is, but a single bit isn’t “information” in the same way the complete works of Shakespeare is. It’s about degree rather than kind, though. There’s some more to think about down the path of compression representing intelligence. Maybe intelligence is using the minimal number of if/else statements to solve a problem, or something like that.
I’ll also go a bit further and propose that awareness/consciousness is defined as “having a world model”. That’s a useful definition, because it immediately makes clear that “consciousness” isn’t just a binary yes/no. The question isn’t “Is X conscious?”, it’s “What is X conscious of?”. Mice have a world model, but one that is mostly dwarfed in scope by a human’s world model. Humans are aware/conscious of many more things than a mouse, though a mouse does experience the world in some ways that a human doesn’t. LLMs provably have a world model, so according to this definition they’re conscious. I’m willing to accept that result, but would also have to note that it’s a world model built on experiencing the universe through an incredibly small peephole of human-written text content compared to the rich sensory experience of humans.
This also doesn’t address at all if they are intelligent or can reason, but I think it helps clarify the thought process to split it up like this. Something can have a large world model but low or no ability to reason (current LLMs). Something else can be very intelligent but have little world model (human brain in a vat, I guess). Normal humans have a large world model and ability to reason, so they are both highly conscious and intelligent.
All this being said, LLMs are clearly not the same as humans. They’ve been a great tech though, for forcing us to explain more exactly why that is. Much like how airplanes and birds both “fly”, but airplanes don’t flap their wings and birds don’t have jet engines.
LLMs generate tokens based on probabilities - they do not create thoughts that they can perform discrete logic with.
The chat bots are deceptive because you can ask questions with discrete logic requirements and they answer convincingly well, but that is because their training data set had many such questions in it, so its really token generation.
If you never played with a old school “chat with eliza” bot, its worth the effort. LLMs are just that super charged, there has to be some input to train on to make the response.
Of course people are trying to glue math and discrete algebraic systems on top of LLM output, but that still does not solve the problem of artificial general intelligence.
Why don’t they “create thoughts”? I mentioned this in another comment, but most discussions around AI are people talking past each other because they use the same words to mean different things.
It might seem absurd, but it’s a lot harder to define words like “thought” than you’d think, because often the definition just leads to more questions. Wikipedia for example says “In their most common sense, they are understood as conscious processes that can happen independently of sensory stimulation.”, but then what does “conscious” mean? Until we have a rigid definition for words like that all the way down to first principles, I wouldn’t agree with definitive statements.
ELIZA is fundamentally different from an LLM though, it’s much more an expert system.
I see what your doing, but your asking for too much formalism in a casual context. To satisfy the entire vocabulary from first principles would be a non-trivial task - its so daunting I don’t even want to attempt it here.
It’s going to be tough to explore this through internet comments, but that just raises the question of “what do you mean by thought and intelligence?”, which then turns into “what do you mean by understanding?” and lots of other similar questions, down a deep rabbit hole. I don’t think it’s really possible to make strong statements either way until we’ve come up with a more coherent theory underlying basic terms like that. I’d love to see a rigid and objective definition that we can measure LLMs against.
I think you’re fishing for some philosophical discussion/redefinition of terms, but I’m not. If you are saying that any algorithm that makes a decision based on criteria is “thinking” or “Intelligent”, then sure, so are LLMs. That definition isn’t one that is accepted by almost anyone, but let’s say that it’s true. In that case, the autocomplete on your phone’s keyboard is thinking and intelligent, so is a random number generator, and so is even a basic if/else statement in programming.
When people discuss “thought” or “intelligence” typically the core definition comes from awareness. Awareness of the decisions that a being is making. LLMs do not have the capability of awareness. It doesn’t understand why it makes the decisions it does, it has no clue if it has created an answer out of thin air or not. It isn’t even aware that it gave an answer. When asked to explain its rationale, it will often contradict the conclusion it came to, because there is no awareness of a decision even being made, it has simply been trained on which words follow which words in an explanation, and strings them together.
I’m fishing for any useful definition at all. What does “awareness” mean? One definition is “conscious knowledge”, and you look up “conscious” to get “Characterized by or having an awareness of one’s environment and one’s own existence, sensations, and thoughts”. It’s circular! So I don’t care about what most people think, because most people can’t define the terms in any useful way. And this is why I dislike almost all discussion about AI, because people have strong opinions without ever considering the basic definitions of the words they’re using (not limited to any one group).
You bring up a really good example, and that’s actually exactly where I’ve ended up after thinking about it. Why isn’t an if/else statement the basic “atom” of intelligence, just like a bit is the basic atom of information? Sure, it’s not “intelligent” in the same way a human is, but a single bit isn’t “information” in the same way the complete works of Shakespeare is. It’s about degree rather than kind, though. There’s some more to think about down the path of compression representing intelligence. Maybe intelligence is using the minimal number of if/else statements to solve a problem, or something like that.
I’ll also go a bit further and propose that awareness/consciousness is defined as “having a world model”. That’s a useful definition, because it immediately makes clear that “consciousness” isn’t just a binary yes/no. The question isn’t “Is X conscious?”, it’s “What is X conscious of?”. Mice have a world model, but one that is mostly dwarfed in scope by a human’s world model. Humans are aware/conscious of many more things than a mouse, though a mouse does experience the world in some ways that a human doesn’t. LLMs provably have a world model, so according to this definition they’re conscious. I’m willing to accept that result, but would also have to note that it’s a world model built on experiencing the universe through an incredibly small peephole of human-written text content compared to the rich sensory experience of humans.
This also doesn’t address at all if they are intelligent or can reason, but I think it helps clarify the thought process to split it up like this. Something can have a large world model but low or no ability to reason (current LLMs). Something else can be very intelligent but have little world model (human brain in a vat, I guess). Normal humans have a large world model and ability to reason, so they are both highly conscious and intelligent.
All this being said, LLMs are clearly not the same as humans. They’ve been a great tech though, for forcing us to explain more exactly why that is. Much like how airplanes and birds both “fly”, but airplanes don’t flap their wings and birds don’t have jet engines.
LLMs generate tokens based on probabilities - they do not create thoughts that they can perform discrete logic with.
The chat bots are deceptive because you can ask questions with discrete logic requirements and they answer convincingly well, but that is because their training data set had many such questions in it, so its really token generation.
If you never played with a old school “chat with eliza” bot, its worth the effort. LLMs are just that super charged, there has to be some input to train on to make the response.
Of course people are trying to glue math and discrete algebraic systems on top of LLM output, but that still does not solve the problem of artificial general intelligence.
Why don’t they “create thoughts”? I mentioned this in another comment, but most discussions around AI are people talking past each other because they use the same words to mean different things.
It might seem absurd, but it’s a lot harder to define words like “thought” than you’d think, because often the definition just leads to more questions. Wikipedia for example says “In their most common sense, they are understood as conscious processes that can happen independently of sensory stimulation.”, but then what does “conscious” mean? Until we have a rigid definition for words like that all the way down to first principles, I wouldn’t agree with definitive statements.
ELIZA is fundamentally different from an LLM though, it’s much more an expert system.
I see what your doing, but your asking for too much formalism in a casual context. To satisfy the entire vocabulary from first principles would be a non-trivial task - its so daunting I don’t even want to attempt it here.