[divan]

> LLM are mathematical models that, given your question, return a sequence of words based on a very complex probability model.

How is this different from how humans respond? Our model could be just some orders of magnitude more complex. Or do you think there some _fundamentally_ different things are going on in the human brain?

[akiselev]

I think that's moot: based on the universal approximation theorem [1], a big enough model is indistinguishable from the human brain, regardless of whether the mechanism of action is fundamentally the same or not. I believe this applies to anything that can somehow be modeled with a continuous function - whether that's possible for the human brain is an open question, though we only need a certain fidelity to be useful.

The more useful question is: can the token prediction model scale to the level of a human intelligence within a reasonable power budget compared to a brain? It's comparing apples to oranges right now but the human brain consumes under 20 watts, a tiny fraction of the TDP of a single A100 GP, and the state of the art isn't even close in performance. We've got a long way to go before we can conclusively answer these questions.

[1] https://en.wikipedia.org/wiki/Universal_approximation_theore...

[checkyoursudo]

This is not the unchallenged, consensus position, though. A competing position within cognitive science is that intelligence requires embodiment, perception, metacognition, curiosity, etc., and that these factors that allow for the emergence of intelligence are indispensable, more or less.

See, e.g., https://plato.stanford.edu/entries/embodied-cognition/

I won't claim to know which is correct, or even if some other alternative is correct; however, this is not settled at all.

I do think it will some day be possible to simulate all of the embodied cognition above, which may truly render this discussion moot, but that LLMs are not doing that at all.

[jasmer]

What is "perception, metacognition, curiosity"

Seriously. How does a bag of random particles have those things?

It cannot, by our own definition.

Hence the metaphysical problem of 'consciousness' as it relates to our variation of scientific materialism.

I suggest the pragmatic approach is along the lines of what the OP said aka 'sufficiently large neural net will be indistinguishable from human' and that's it. We will see things that we can de facto contemplate as 'curiosity' 'perception' 'meta-cognition' if we want to, especially if we start to develop a more meta understanding of these systems, or not, and that's it.

We'll probably be arguing about 'cognition' long, long after we have variation of AI that kind of seem to be AGI. By many measures we are already kind of there Chat GPT will fool humans probably most of the time and that's that.

[oxfordmale]

You know what a book is and can reason about it. As a human you can provide answers that go beyond the knowledge you have ingested.

For example, if you ask ChatGPT if books can be fluorescent it says no. However, as an adult you know someone somewhere has made a book with fluorescent images, as it is a cool thing. You are combining knowledge from two different fields (books + fluorescence) and establishing the likelihood of someone being able to combine them.

[ai-punter]

lol, really?

Me: Can books be fluorescent?

GPT-4: While the term "fluorescent" is typically used to describe substances that can absorb light and then re-emit it, often at a different wavelength, there's no inherent reason why a book couldn't be made with fluorescent properties.

This could be achieved by using fluorescent ink, dyes, or paints on the cover or the pages, or by incorporating fluorescent fibers into the paper itself. When exposed to ultraviolet light (often called "black light"), these materials would glow. This might be used for aesthetic reasons, for practical reasons like aiding reading in low-light conditions, or for interactive elements in children's books or art books.

It's important to note that this is not a common feature for most books, as it would increase production costs and might not be appealing or necessary for all readers or all types of books. Also, long-term exposure to ultraviolet light can be damaging to many types of paper and ink, which could reduce the lifespan of the book.

As of my knowledge cutoff in September 2021, I don't have any specific examples of fluorescent books, but that doesn't mean they don't exist or couldn't be created.

[oxfordmale]

This exactly illustrates the problem of ChatGPT. I asked the same question, phrased slightly differently, and it said no. It also not unlikely my question was used to train CHATGPT further.

[9dev]

Does it, really? I’m pretty sure if you’d ask some people, there are going to be some that’d answer „no“ too.

Besides, you’re shifting the goalposts as you go, altering your arguments as it suits your view. The parent comment just disproved your entire point how LLMs cannot combine different fields - and now you just pick a different angle altogether. Maybe take a step back and reconsider your opinions?

[pcthrowaway]

ChatGPT is GPT3.5 isn't it? The other person was using GPT4.

[jasmer]

Different answers at different times, seems very human to me :)

[divan]

Yet, this doesn't answer my question. Human brain is obviously has so much more in it (visual cortex for starters, grid cells etc), and, in terms of neural network, much more sophisticated architecture. But still, there is a big probability that what we call "knowledge" and "reasoning" and "conciousness" are just a result of this sophisticated architecture. I.e. there is no special magical thing for "reasoning" that next generations of prediction models can't replicate.

There is a faboulous book by Jeff Hawkins "On Intelligence" (2004) that explores this. I think main premise of it still holds true: brain is "just" a highly sophisticated hierarchical tissue whose main job is to extract patterns and make predictions. Fundamentally it doesn't seem very different from what LLMs are.