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

Sparks of Artificial General Intelligence: Early Experiments with GPT-4

arxiv.org

Post reply on HN
Print view View original

111–120 of 244 posts

Re: Sparks of Artificial General Intelligence: Early Experiments with GPT-4

#111

Post by crooked-v »

beoberha wrote: post #75

I know enough about how neural nets work to be absolutely blown away at how good the GPT are. I only skimmed the paper, but even chatGPT showed a lot of these “sparks”, IMO. We are certainly a long way off from any semblance of general intelligence, but for a model that just tries to predict the next word, I’m dumbfounded at how good it is.

> We are certainly a long way off from any semblance of general intelligence

Part of me is starting to think that the only thing we're really missing at this point to start seeing that is to have one of these models that can modify itself with its output and thereby have a mechanism to 'learn' or 'remember' things.

Re: Sparks of Artificial General Intelligence: Early Experiments with GPT-4

#113

Post by jwdunne »

HopenHeyHi wrote: post #80
cowl wrote: post #71

Earlier quoted context omitted.

So you predict the top comments For a claim would be: 1. Dismisal 2. Trivialism 3. Non Well Formed Claim 4. I accept the claim 5. Disprove by counter example Are you sure you have not forgotten any tactic of debate to include in you prediction? I predict that you Prediction will result probably in these actions: 1. upvoted 2. downvoted

6. Meta comment for karma whoring 7. Like 6, but funnier A. Joke thread pile on B. Reprimands from humorless C. Dejected mods having to clean it all up

Hang on, let me cover B:

tHiS sHoUlDn’T bE oN hAcKeRnEws

Re: Sparks of Artificial General Intelligence: Early Experiments with GPT-4

#114

Post by js8 »

siva7 wrote: post #95
js8 wrote: post #83

I don't accept that something is AGI unless it can solve general instances of SAT (satisfiability problem, not the school test). Also recognizing (formulating from the task) an instance in the first place would help too. To me, these are hallmarks of reason, and not available in LLMs, in fact probably impossible just with pattern recognition.

That's fine. It's maybe not AGI IQ 180 to meet your demands but if we're honest it's pretty close.

Feynman had a really nice story about how he was into puzzles when he was at Princeton. It took him a while to solve the new ones, but eventually he learned all the well-known instances so he could answer instantly. It made him a genius in other people's eyes.

All I want from AGI is to demonstrate that it can solve a straightforward logic problems (puzzles, if you will), that it provably didn't see before. Or at least recognize it is being indirectly given such task. So far, evidence suggests it is not capable of that.

Re: Sparks of Artificial General Intelligence: Early Experiments with GPT-4

#115

Post by pzo »

worrycue wrote: post #94
hislaziness wrote: post #79

For whatever reason we seem to have set a very high expectation from AI as compared to NI (Natural Intelligence). I remember reading "There is a considerable overlap between the intelligence of the smartest bears and the dumbest tourists."

We get our expectations from fiction. AIs in shows like Star Trek are precise and accurate - the perfect complement to the unreliability of humans. That’s what we want.

That's true and it's a high bar because it seems many people would expect AI to be at least as the smartest of human ever lived.

However if the AI is the same smart as the most dumb human or human with mental disability would we then consider those humans as no intelligent at all or not qualifying as homo sapiens anymore?

If AI can be the same as good as even 'dumb' human it's already a big achievement because can still provide some value and because AI can be scaled so you can still have billions of dumb AIs - already millions of users are interacting with chatGPT daily

Re: Sparks of Artificial General Intelligence: Early Experiments with GPT-4

#116

Post by throw310822 »

gwoolhurme wrote: post #88

From the intro: "we believe that it could reasonably be viewed as an early (yet still incomplete) version of an artificial general intelligence (AGI) system." What does that mean? If we take it as fact, so if it is an early version of AGI, Microsoft is using this thing to push subscriptions to all their services? This thing that is potentially the greatest thing humanity has made, an artificial living thing, and it's…

It's a product, and it's not that far away from that of their competitors. And there are a lot. Just a few weeks ago, Yann LeCun said that llms are not particularly interesting or innovative from the research point of view.

Re: Sparks of Artificial General Intelligence: Early Experiments with GPT-4

#117

Post by IanCal »

cowl wrote: post #71
raincole wrote: post #67

My prediction for the top comments of this thread (paraphrased) 1. It's just Microsoft's advertisement 2. No it's just a very effective pattern matching algorithm 3. Please define intelligence first otherwise it's nonsense 4. I welcome our machine overlord 5. Lmao I asked it to do $thing and it failed I'd like to know if GPT-4 can predict the top comments of this thread?

So you predict the top comments For a claim would be: 1. Dismisal 2. Trivialism 3. Non Well Formed Claim 4. I accept the claim 5. Disprove by counter example Are you sure you have not forgotten any tactic of debate to include in you prediction? I predict that you Prediction will result probably in these actions: 1. upvoted 2. downvoted

6. Discussion around the idea that adds something of value.

Re: Sparks of Artificial General Intelligence: Early Experiments with GPT-4

#120

Post by atleastoptimal »

To me it's really crazy that there is a public UI (ChatGPT) that lets people use GPT-4. If OpenAI had the attitude of Google they would have just gone "Yeah we created a language model that's light years ahead of anything else, look how cool it is, but sorry due to public safety you will never get to use it. Bye now!" I feel that the public accessibility of these large language models is a fluke. Being able to use it for almost free feels like cheating reality.
Post reply on HN