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

We come to bury ChatGPT, not to praise it

danmcquillan.org

Post reply on HN
Print view View original

171–180 of 425 posts

Re: We come to bury ChatGPT, not to praise it

#171

Post by Kamq »

emodendroket wrote: post #151
tambourine_man wrote: post #121

I’ve seen this argument over and over again: it’s just predicting the next word, it’s not creating anything new, etc. How do we know we’re not doing the same or similar in our brains? We can’t even define creativity, intelligence, let alone tell how it works. Predicting the next word is very much part of an intelligent discourse. What if it’s all just a “computational guessing game” all the way down with perhaps a sp…

That would seem to be the “stochastic parrot” view denounced. But it seems hard to believe. Haven’t you ever had the experience of knowing exactly what you want to say but struggling to find the word for it? How could that be if we’re just randomly selecting words like a souped-up Markov chain?

> Haven’t you ever had the experience of knowing exactly what you want to say but struggling to find the word for it? How could that be if we’re just randomly selecting words like a souped-up Markov chain?

All words in the normal cache have a low probability, so you have to hit disk to get the full word list.

Re: We come to bury ChatGPT, not to praise it

#172

Post by WaxProlix »

AussieWog93 wrote: post #136
efitz wrote: post #102

The BS that ChatGPT generates is especially odious because it requires so much effort to detect if you're not a SME in whatever output it's giving. Plus, the output, when wrong, is subtly wrong- it's usually not obvious BS, it's credible BS. If you are not competent in the particular area you're asking about, you likely don't have the skills to recognize BS. It also is a time saver, doing work that most people find u…

>if you're not a SME Small and Medium-sized Enterprise? What's SME in this instance?

subject matter Expert .

Re: We come to bury ChatGPT, not to praise it

#173

Post by edgefield »

fenomas wrote: post #143

It's so weird to me when apparently tech-savvy people say ChatGPT "lies", or call it a bullshit generator and so on. I mean - when you ask StableDiffusion to draw a dog astronaut, everyone gets that the image it returns is made-up, right? Nobody expects the AI to return only "true" images of existing things - it was trained on fictional images as well as photographs, and people understand that it can imagine new thin…

I think a fairer analogy is always expecting a Google search to return factual, objective, and high quality information. Similar to Chat GPT, the results are often mixed and depend on the way you structure your search, the content area, etc.

Re: We come to bury ChatGPT, not to praise it

#174

Post by somenameforme »

WalterSear wrote: post #2

> If it produces plausible articles or computer code it means the inevitable hallucinations are becoming harder to spot. How functional does the code have to be before it's no longer a hallucination? What if we are all not much more than only drawing "on the (admittedly vast) proportion of [our experiences] ingested at training time?

Imagine yourself, for a minute, 50,000 years ago in the past. By many estimates this was just before humans had developed the ability for normal speech. The bleeding edge of technology is "poke him with the pointy end" and the epitome of human speech and literature is *angry grunting sound*. Even if somehow one was able to insert the entire quantifiable knowledge of humanity into an e.g. ChatGPT style system you're not really going to get anything more out of it than what you put into it. Ask the computer for how to develop nuclear energy and you're going to get back *confused grunting sound*.

You're not really going to get from there (as a machine) to man on the moon, Shakespeare, and nuclear energy through anything like a normal recombination of what's already known. Yet, somehow, humanity did. And extremely fast. The time frame from then to now is but 1000-2000 human generations. And that with endless war, fallible memory, dark ages, knowledge being lost (or burnt), and so on endlessly. A "intelligent" computer system, without such flaws, ought be able to replicate our progress in a negligible amount of time. But whatever technology this may be that I'm appealing to, I don't really see it on our current path with natural language search/recombination.

Re: We come to bury ChatGPT, not to praise it

#175

Post by lolinder »

fenomas wrote: post #143

It's so weird to me when apparently tech-savvy people say ChatGPT "lies", or call it a bullshit generator and so on. I mean - when you ask StableDiffusion to draw a dog astronaut, everyone gets that the image it returns is made-up, right? Nobody expects the AI to return only "true" images of existing things - it was trained on fictional images as well as photographs, and people understand that it can imagine new thin…

Whether something is an error mode doesn't depend on the design of the tech, it depends on the use case it's being applied to.

I frequently point out that GPT generates falsehoods because I'm constantly reading stuff from people who try to use ChatGPT as a knowledge retrieval engine. In that context, generating bullshit is an error mode, and the correct answer isn't to try to fix ChatGPT, it's to raise awareness of its inherent limitations.

So, I agree with you wholeheartedly that generating bullshit is a feature, not a bug. I just want the rest of the world to realize that.

Re: We come to bury ChatGPT, not to praise it

#177

Post by delhanty »

Gatsky wrote: post #56

> However, as I spell out in my book, the concept of AGI is inseparable from the kind of hierarchy of intelligence that has underpinned ideas of innate supremacy since the days of empire and colonialism. I think this line discredits the entire article, which was already a few obvious points blown at gale force into the reader’s ear.

Indeed

It seems that, like ChatGPT, the author is also a bullshit generator, and therefore, by his own definition, unintelligent.

Re: We come to bury ChatGPT, not to praise it

#178

Post by kazinator »

alfalfasprout wrote: post #112

It's a tool that isn't going away. But even Yann Lecun has pointed out that "A house cat has way more common sense and understanding of the world than any LLM". That's fundamentally the real issue. LLMs are far from even animal level "cognition" in their ability to solve novel problems. A lot of people seem to vastly underestimate the number of novel problems they solve on a regular basis. In practice, what we're see…

> but the risks of putting out brand-damaging copy

There are ways to mitigate that, like slow initial roll-out to a small population. Or something.

Re: We come to bury ChatGPT, not to praise it

#179

Post by zyklonix »

tambourine_man wrote: post #121

I’ve seen this argument over and over again: it’s just predicting the next word, it’s not creating anything new, etc. How do we know we’re not doing the same or similar in our brains? We can’t even define creativity, intelligence, let alone tell how it works. Predicting the next word is very much part of an intelligent discourse. What if it’s all just a “computational guessing game” all the way down with perhaps a sp…

This is in fact the whole premise of Jeff Hawkins' book "On intelligence". Prediction is at the core of how we learn/act. Since we are born, we are just predicting everything that will happen and our brain learns when we see/hear/sense something that we can't predict.

Re: We come to bury ChatGPT, not to praise it

#180

Post by muzani »

tambourine_man wrote: post #121

I’ve seen this argument over and over again: it’s just predicting the next word, it’s not creating anything new, etc. How do we know we’re not doing the same or similar in our brains? We can’t even define creativity, intelligence, let alone tell how it works. Predicting the next word is very much part of an intelligent discourse. What if it’s all just a “computational guessing game” all the way down with perhaps a sp…

I love the philosophy behind Gustafson's Law: the problem size increases in response to computational resources.

You get a simple computer, you can drop bombs more accurately. Then you make word processors. Then text based simulation aka games. Then turn books into digital format. Then whole online libraries, ecommerce, social media, cryptocurrency, and so on.

Seeing a LLM as purely language based might be missing the forest for the atoms.

Post reply on HN