[Chirono]

This paper, and most other places i’ve seen it argued that language models can’t possibly be conscious, sentient, thinking etc, rely heavily on the idea that llms are ‘just’ doing statistical prediction of tokens.

I personally find this utterly unconvincing. For a start, I’m not entirely sure that’s not what I’m doing in typing out this message. My brain is ‘just’ chemistry, so clearly can’t have beliefs or be conscious, right?

But more relevant is the fact that llms like ChatGPT are only pre-trained on pure statistical generation, followed by further tuning through reinforcement learning. So ChatGPT is no longer simply doing pure statistical modelling, though of course the interface of calculating logits for the next token remains the same.

note: i’m not saying i think llms are conscious. I don’t think the question even makes much sense. I am saying all the arguments that i’ve seen for why they aren’t have been very unsatisfying.

[goatlover]

> I personally find this utterly unconvincing. For a start, I’m not entirely sure that’s not what I’m doing in typing out this message. My brain is ‘just’ chemistry, so clearly can’t have beliefs or be conscious, right?

Your brain is part of an organism who's ancestors evolved to survive the real world, not by matching tokens. As such, language is a skill that helps humans survive and reproduce, not a tool used to mimic human language. Chemistry is the wrong level to evaluate cognition at.

Also, you can note the differences between how actual neurons work compared to language models as other posters have mentioned.

[Chirono]

Of course they’re different. But so what? That’s not exactly proof of anything, unless you’re suggestion biological neurons are the only configuration in the universe capable of thought? Maybe that’s true, but it seems unlikely to me.

The pressure of natural selection can lead to the phenomenon of consciousness. Why not the process of training llms? Perhaps developing the machine equivalent of consciousness helps that particular configuration of weights survive the otherwise destructive process of gradient descent.

[goatlover]

I'm saying that when humans use language, it's about stuff in the world and ourselves. The words have references or uses that we call meaning. When an LLM models language, it is approximating the patterns of language use we have produced that have been made available to it. It doesn't understand what the words it's producing are about (no external references), only how to produce those words in patterns that are meaningful to us.

What if we fed an LLM a bunch of crazy nonsense instead? It would model the patterns in the word use and then give us answers based on the nonsense it was fed. But it wouldn't understand that it's actually nonsense that doesn't apply to the real world.

[roflyear]

OK well - sure - even if that is how we work, then language models are much worse at it than we are.

They are better than us at some things already, but do I think they will be better than us at EVERYTHING?

No.

[Chirono]

out of interest, is there anything specific you think humans will always be better at than AI?

[roflyear]

Hmmm, unsure. Maybe insight? Creativity? I would not call the art models creative, they are copying things similarly to how GPT-3 does. Very impressive but I don't think that is "creativity."