Live data from Hacker News

Rodney Brooks on GPT-4

spectrum.ieee.org

91–100 of 412 posts

Re: Rodney Brooks on GPT-4

#91

Earlier quoted context omitted.

Firstly, if humans created sapience, we created it with all the interacting elements that evolution used plus all the things that we used. Secondly, maybe sapience isn't as big of a deal as we thought compared with all the other things that evolution did. Remember that biological entities have to figure out survival, reproduction etc. Sapience emerges as a byproduct but the selective pressure is towards those things…

Alright, I give. What the heck is "sapience"?

Generally people use that word to mean "wisdom" or at least "some reasoning ability distinct from what animals have".

Edit to add: From a personal perspective I don't see any reason to think humans have qualitatively different reasoning abilities from animals, or a unique "soul" or anything like that, so the term "sapience" doesn't really have a special resonance with me like it might for someone who thinks those things. That may affect some judgements here I don't know.

Re: Rodney Brooks on GPT-4

#92

Annoyed at all these N=1 articles from prominent thinkers about this stuff. Especially from scientists - can these sorts of folks please more carefully quantify, how often it’s “wrong” and then from that decide whether or not to “calm down”. Right now I suspect we hear from the outliers on both ends of the spectrum here. People who either see AGI happening tomorrow and the more dismissive crowd. But aside from what w…

Anytime I ask these things something (bard, gpt etc), 33% of the answer is genius, 33% misleading garbage, 33% filler stuff that’s neither here or there The problem is distinguishing between these parts requires me to be be an expert in the area I’m inquiring about - and then why the heck do I need to ask some idiot bot for answers to questions that I already know an answer to? I don’t know who finds these things use…

Yeah I think the use cares for these things are far narrower than the boosters & hype cycle think.

If you could have unlimited interns for $0 (let's pretend it doesn't cost tons and tons of compute) that don't shutup, hallucinate & lie, and also do good work, in varying degrees.. how many would you want?

These things are probably going to be great for lots of blackout - propaganda, political marketing, flooding the zone with BS of unlimited iterations of messaging. Basically things that can be A/B tested to death, where veracity is of zero importance, and you have near limitless shots on goal to keep iterating.

Re: Rodney Brooks on GPT-4

#93
post #33

> The large language models are a little surprising. I’ll give you that. I think this is the key point about LLMs that kind of explains the wide and polarized views on whether it understands or parrots, whether it can think or is the precursor to thinking or is a dead-end, whether it will catastrophically destroy the world, or “merely” make it steadily worse with bullshit, or just put a few industries out of a job. A…

I think the main issue is that since we don't have access to the implementation of GPT-4, there's no way for researchers to know how it works, so either they're not saying anything or they're just making things up. GPT-4 is "multimodal" and RLHF'd, so it was trained with some tasks other than next word prediction. I don't remember if it's been trained for code correctness (by running unit tests etc.), but other model…

That's not much of a problem, as it only affects the "GPT" LLM.

The same researchers can always study comparable in behavior LLMs like Llama and its descendants.

Re: Rodney Brooks on GPT-4

#94

Earlier quoted context omitted.

If LLMs are an understanding of the world it would mean humans in a few decades found a way to create sapience with many orders of magnitude fewer interacting elements than evolution did. I find that doubtful, at least in light of the fact every other way we've replicated biological computation requires many more computational elements.

Firstly, if humans created sapience, we created it with all the interacting elements that evolution used plus all the things that we used. Secondly, maybe sapience isn't as big of a deal as we thought compared with all the other things that evolution did. Remember that biological entities have to figure out survival, reproduction etc. Sapience emerges as a byproduct but the selective pressure is towards those things…

Good point. I would even add humans brain have to process a lot of sensory input: video, audio, touch, smell, taste. And communication channel for human is very inefficient - you have to translate your thoughs to language then to audio then someone alse doing the same in reverse. If you texting you have to slowly type something then someone else need to do text recognition in their brain.

LLM are like handicapped humans that are visual, hearing impaired, with long covid no taste or smell and can focus only thinking and very efficient communication channel as text/tokens sent via wire on the internet

Re: Rodney Brooks on GPT-4

#95
post #39

Earlier quoted context omitted.

Youre right but I think the point still stands. Many times it really is easier to verify than produce. Compilable code being a good example.

I dunno about you, but I spend a LOT of my time fixing code that was broken in some subtle way the compiler didn't catch If we're talking about generating and integrating sample code, it's great at that Anything more advanced and it's a footgun

I have very quickly picked up the habit of pasting snippets of my code into GPT-4 and simply asking "Why is this not working?" Almost every time, it succinctly explains the apparent purpose of the code, and how it is subtly wrong.

It's so good that I often do this preemptively to avoid a compile/deploy/test cycle.

Re: Rodney Brooks on GPT-4

#96

Earlier quoted context omitted.

Firstly, if humans created sapience, we created it with all the interacting elements that evolution used plus all the things that we used. Secondly, maybe sapience isn't as big of a deal as we thought compared with all the other things that evolution did. Remember that biological entities have to figure out survival, reproduction etc. Sapience emerges as a byproduct but the selective pressure is towards those things…

Alright, I give. What the heck is "sapience"?

You know, the capacity of though/intelligence/understanding the world?

As in "homo sapiens" (same latin root)

Re: Rodney Brooks on GPT-4

#97
Brooks has done something I appreciate a lot: he turned a phrase.

> stop confusing performance with competence

You can safely skip the rest of the article. That sentence gives you all you need, because you are competent.

If you want a little more meat:

> The example I used at the time was, I think it was a Google program labeling an image of people playing Frisbee in the park. And if a person says, “Oh, that’s a person playing Frisbee in the park,” you would assume you could ask him a question, like, “Can you eat a Frisbee?” And they would know, of course not; it’s made of plastic. You’d just expect they’d have that competence. That they would know the answer to the question, “Can you play Frisbee in a snowstorm? Or, how far can a person throw a Frisbee? Can they throw it 10 miles? Can they only throw it 10 centimeters?” You’d expect all that competence from that one piece of performance: a person saying, “That’s a picture of people playing Frisbee in the park.”

---

So I've calmed down. Now what? The problem isn't only that this train is flying off on a tangent: it's that it's off the rails. What rails should it be on?

The problem, as I see it, is narrative. As soon as we called it "AI", that wrote the Genesis of the Scripture of the cult. In this new religious movement, God is spelled L-L-M. Back here in reality, LLM isn't a God; or even a person at all.

That's the mistake: personification. A person can perform, but a performance can't person.

---

Narrative is a powerful tool. It's why we're so excited about Natural Language Processing in the first place. Ever since the very origins of software, the power of narrative has been so close, but always still just out of grasp. Do we even know what we are reaching for in the first place?

In a sense, we have a part of it: explicit definition. What Chomsky categorized "Context-Free Grammar", we have made into programming languages. What they are missing is implicit inference: context.

That's what LLMs do. They use inference to model the patterns that exist in written text. With that model, they can hallucinate more text that follows the same patterns: they can perform natural language.

So that's it, right? Problem solved! What's missing? explicit definition. We traded one problem for another. No one (so far) has figured out how to solve both in the same program. You can have definition, or, you can have inference. You can't have both.

This doesn't make any sense to us humans. We don't have any trouble at all doing both at the same time. We do it all the time! Do we actually do anything else? Unfortunately, LLMs are not humans.

---

The two approaches to language are diametrically opposed, but they work with the same domain. Approaching from either end of the spectrum, definition and inference explore together the wild universe that is story. That's the missing piece: once we figure out what story is made of, we should be able to put all three pieces together.

Re: Rodney Brooks on GPT-4

#98
Not really the thrust of the article, but that '50s picture of the family playing scrabble in a self-driving car, surrounded by text about trains... really makes me think that if I were that family, I'd still prefer to be playing scrabble on a train, rather than in a cramped self-driving car on a highway.

Re: Rodney Brooks on GPT-4

#99
Gpt is a great tool, it won’t be able to do complex tasks because it really isn’t that smart. If you tell it to do something relatively complex end to end, it will fail unless a plugin specifically supports it.

Re: Rodney Brooks on GPT-4

#100

At least run the model on the examples considered... Here's GPT3.5 > Can you eat a Frisbee? No, you cannot eat a Frisbee. A Frisbee is typically made of plastic, often polypropylene or similar materials, which are not meant for consumption. These materials are not digestible and can pose a choking hazard or harm your digestive system if ingested. It's important to only consume food and items that are safe and intende…

Funny. I asked GPT-4 what frisbees are most commonly made of, and it sayd polyethylene. I then asked if it was sure, and it corrected itself and said it was polypropylene. I asked again if it was sure, and it corrected itself saying it was low-density polyethylene. I kept this going for a while and it kept changing its answer. This does seem to agree with the author. The first answer was very convincing, but not what…

I checked the first answer with a Google search. The first answer was right.
Post reply on HN