Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

AI isn’t good enough

skventures.substack.com

Post reply on HN
Print view View original

131–140 of 374 posts

Re: AI isn’t good enough

#131

Post by coldtea »

RandomLensman wrote: post #125
auggierose wrote: post #122

Earlier quoted context omitted.

Anyone who has worked a bit with a top LLM thinks that they learn world models. Otherwise, what they are doing would be impossible. I've used them for things that are definitely not on the web, because they are brand new research. They are definitely able to apply what they've learnt in novel ways.

If they learn world models, those world models are incredible poor, i.e., there is no consistency of thought in those world models. In my experience, things outside coding quickly devolve into something more like "technobabble" (and in coding there is always a lot of made-up stuff that doesn't exists in terms of functions etc.).

>If they learn world models, those world models are incredible poor, i.e., there is no consistency of thought in those world models

Incredibly poor compared to ours, but thousands of times better than what "AI" we had before.

Re: AI isn’t good enough

#132

Post by js8 »

beefield wrote: post #78
usrbinbash wrote: post #67

Earlier quoted context omitted.

> This entire piece is based on one massive, unsupported assertion, which is that LLM progress will cease. Which is countered by...the assertion that it won't? LLMs won't get intelligent. That's a fact based on their MO. They are sequence completion engines. They can be fine tuned to specific tasks, but at their core, they remain stochastic parrots. > I want to know only one thing, which is what gives him the confide…

> LLMs won't get intelligent. That's a fact based on their MO. I kind of agree. However, I see a real possibility that in the near future LLM behaviour would be practically indistinguishable from intelligent/sentient behavior. And at that point we (or at least I) are facing some really interesting/difficult questions, namely how do you know an intelligent looking thing actually is intelligent (or sentient). How do yo…

Any sufficiently advanced intelligence is indistinguishable from an LLM?

Cute, but practically speaking, I would prefer the former.

Re: AI isn’t good enough

#133

Post by coldtea »

velvetz wrote: post #104
beefield wrote: post #78

Earlier quoted context omitted.

> LLMs won't get intelligent. That's a fact based on their MO. I kind of agree. However, I see a real possibility that in the near future LLM behaviour would be practically indistinguishable from intelligent/sentient behavior. And at that point we (or at least I) are facing some really interesting/difficult questions, namely how do you know an intelligent looking thing actually is intelligent (or sentient). How do yo…

Off-topic and might sound strange but I find it intriguing that you wrote behavior and behaviour in these two different forms in the same sentence:)

Quite easy to happen if like many of us you were taugh British English, and then you remember that you're on a US forum, and everybody around you uses the US spelling for things (and you get to read the US variants all the time in other comments).

Re: AI isn’t good enough

#134

Post by glenstein »

tlarkworthy wrote: post #119
glenstein wrote: post #115

Earlier quoted context omitted.

>Which is countered by...the assertion that it won't? No it's countered by principled restraint in not making an affirmative claim one way or the other. I've heard this referred to as the overconfident pessimism problem. Which is that normal, well founded scientific discipline and evidence-based restraint go out of the window when people declare, without evidence that they know certain advances won't happen. Because…

AI has been around the corner since the 1950s, this is the historical evidence for the pessimistic stance against over optimistic predictions. LLMs are a huge stride forward, but AI does not progress like Moore's law. LLM have revealed a new wall. Combining multi agents is not working out as hoped.

Perhaps without intending to, you've cited a pretty appropriate example of overconfident pessimism. Philosopher Hubert Dreyfus is most responsible for this portrayal of AI research in the '50s and '60s. He made a career of insisting that advances in AI would never come to pass, famously predicting that computers couldn't become good at chess because it required "insight", and routinely listing off what he believed were uniquely human qualities that couldn't be embodied in computers, always in the form of underdefined terms such as intuition, insight, and other such terms.

Many of the things AI does now are exactly the type of things that doomsayers explicitly predicted would never happen, because they extrapolated from limited progress in the short term to absolute declarations over the infinite timeline of the future.

There's a difference between the outer limits of theoretical possibility on the one hand, and instant results over the span of a couple new cycles, and it's unfortunate that these get conflated.

Re: AI isn’t good enough

#135

Post by auggierose »

RandomLensman wrote: post #125
auggierose wrote: post #122

Earlier quoted context omitted.

Anyone who has worked a bit with a top LLM thinks that they learn world models. Otherwise, what they are doing would be impossible. I've used them for things that are definitely not on the web, because they are brand new research. They are definitely able to apply what they've learnt in novel ways.

If they learn world models, those world models are incredible poor, i.e., there is no consistency of thought in those world models. In my experience, things outside coding quickly devolve into something more like "technobabble" (and in coding there is always a lot of made-up stuff that doesn't exists in terms of functions etc.).

I see them more as creative artists who have very good intuition, but are poor logicians. Their world model is not a strict database of consistent facts, it is more like a set of various beliefs, and of course those can be highly contradictory.

Re: AI isn’t good enough

#136

Post by nottorp »

atleastoptimal wrote: post #30

"we are already at the tail end of the current wave of AI." Is that really true? Generative AI spending this year just so far is 4x greater than last year, and the results of that spending haven't been released yet. How can they conclude this? Most "predictions" about the future I've seen assume that ChatGPT will stay the same and not improve for the next 10 years, or only marginally. Why they make this assumption, I…

> Generative AI spending this year just so far is 4x greater than last year

I'm sure they spent a lot of money looking for gold in the long rushes long before the gold was all gone too.

Re: AI isn’t good enough

#137

Post by nine_k »

ambrozk wrote: post #18

This entire piece is based on one massive, unsupported assertion, which is that LLM progress will cease. Or, as the author puts it, "we are at the tail end of the first wave of large language model-based AI... [it] ends somewhere in the next year or two with the kinds of limits people are running up against." I want to know only one thing, which is what gives him the confidence necessary to say that. If that one stat…

The current crop of LLMs is like Paleozoic megafauna, or like Egyptian Pyramids. It takes a few relatively simple approaches and stretches them wildly, using colossal computing resources.

Live systems in nature seem to solve similar problems with way less compute available. There should be better architectures.

Also, as somebody said, every exponential growth curve is a lower part of a sigmoid. LLMs will plateau at some level. That level may be impressively high though.

Re: AI isn’t good enough

#138

Post by RandomLensman »

coldtea wrote: post #131
RandomLensman wrote: post #125

Earlier quoted context omitted.

If they learn world models, those world models are incredible poor, i.e., there is no consistency of thought in those world models. In my experience, things outside coding quickly devolve into something more like "technobabble" (and in coding there is always a lot of made-up stuff that doesn't exists in terms of functions etc.).

> If they learn world models, those world models are incredible poor, i.e., there is no consistency of thought in those world models Incredibly poor compared to ours, but thousands of times better than what "AI" we had before.

Not sure that matters much as they are only for low risk stuff without skilled supervision, so back to advertising, marketing, cheap customer support, etc.

Re: AI isn’t good enough

#139

Post by coldtea »

savolai wrote: post #101
beefield wrote: post #78

Earlier quoted context omitted.

> LLMs won't get intelligent. That's a fact based on their MO. I kind of agree. However, I see a real possibility that in the near future LLM behaviour would be practically indistinguishable from intelligent/sentient behavior. And at that point we (or at least I) are facing some really interesting/difficult questions, namely how do you know an intelligent looking thing actually is intelligent (or sentient). How do yo…

It seems to me that the real question here is what is true human intelligence. Ai has made it plain to see, by being able to replicate it so convincingly, that much of what we have considered intelligence has been pattern matching or acting as complex parrots. There is much more to the abilities of human body-mind-emotional-experiential being, but it is only slowly becoming mainstream. (Edit: Of course there are also…

>One salient question in this is: will we reach a level of intelligence where we become beings capable of actual collaboration that doesn’t waste so much effort in conflicts, or one that is capable of living in harmony within its environment?

That's more about ethics and an wish for moral behavior and conflict aversion, than about intelligence.

Intelligence (human and AI) could just as well opt for conflict and evil, if this helps it get the upper hand for its own private goals and interests.

Simply put, the interests of the collective, are not necessarily the interests of the individual intelligence.

(Even assuming there was a single, easy to agree upon, "interest of the collective" for most problems).

Re: AI isn’t good enough

#140

Post by RandomLensman »

auggierose wrote: post #135
RandomLensman wrote: post #125

Earlier quoted context omitted.

If they learn world models, those world models are incredible poor, i.e., there is no consistency of thought in those world models. In my experience, things outside coding quickly devolve into something more like "technobabble" (and in coding there is always a lot of made-up stuff that doesn't exists in terms of functions etc.).

I see them more as creative artists who have very good intuition, but are poor logicians. Their world model is not a strict database of consistent facts, it is more like a set of various beliefs, and of course those can be highly contradictory.

That maybe sufficient for advertising, marketing, some shallow story telling etc., it is way too dangerous for anything in the physical sciences, legal, medicine, ...
Post reply on HN