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

271–280 of 425 posts

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

#271

Post by cmrdporcupine »

labrador wrote: post #3

I don't get the point of this article. ChatGPT and it's ilk are not going away. They are here, now, and our friends and families will start using them very soon. I have a contrarian view from this article that ChatGPT is actually good for the mainstream audience because it will teach them to think about how trust worthy the information is coming from a machine. Most people will understand they have to take what is be…

I can't even teach my baby boomer aged inlaws to understand that obvious bullshit social media chainposts (along the lines of "copy this into your timeline to stop facebook from selling your data") are fake. There's no way they're going to see past ML-generated content, esp ones where feedback from success/fail ratios will cause it to 'improve' over time.

We might not be able to bury the cat, but that's hardly a good thing. Shits going to get really awkward for the next few decades.

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

#272

Post by phaedrus »

tambourine_man wrote: post #223
root_axis wrote: post #218

Earlier quoted context omitted.

I won't speak for what's "all the way down", but on whatever level humans perform reflective reasoning, it's self-evident that we don't simply "predict the next word" in the way a LLM does. Whether or not a LLM is capable of "creating anything new" is a different argument altogether. "New" is in eye of the beholder and it is often said that nothing is new under the sun, but we already know that unthinking processes w…

Please provide evidence that intelligent speech isn’t simply predicting the next word, as it’s obviously not self-evident to me.

I've always hated the idea promoted by some AI researchers that "intelligence is just prediction." Specifically in the context of your challenge, since a brain must decide the next word before uttering it, is it not vacuously true that any conceivable method of producing speech is "predicting" it? You're asking for evidence that it isn't; I'm asking for evidence that question has any meaning.

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

#273

Post by dcow »

pvg wrote: post #207
dcow wrote: post #193

Earlier quoted context omitted.

Despite the qualities I ascribed to it, I actually read the piece. That's why I chuckled at the satire of having ChatGPT summarize it. I thought it sufficiently concise and every so slightly poignant commentary, daresay curious . I haven't disparaged anybody else for their own curious conversation, have I? If you'd like to discuss whether, "The structural injustices and supremacist perspectives layered into AI put it…

I didn't say you've disparaged anyone, you're defending the paste of generated content into the thread as if it's a good thing because, I dunno, you found it amusing or it tickled your sneerbone. But pasting generated content into the thread is not a good thing by HN's standards of good things for the reasons I described and the reasons you described are fine reasons but they are bad HN reasons.

I don't care where the content came from. I judged it on its merit not on its origin. If it was a bad summary or more difficult to understand than the article then it wouldn't have worked quite as well. The additional context of it being generated by the [harmful technology in question] made it meta humorous, yes, not least because I found it to be a better representation of the argument that I think the author is attempting to communicate than the form in which it was originally presented. Ultimately, the author's piece sounds more like a stochastic parrot to me than the summary by ChatGPT.

Hence, I find satire an entirely appropriate tactic to spark curious discussion, in this case. It's able to communicate all that remarkably succinctly. And, though we have not discussed whether ChatGPT is harmful because OpenAI paid Kenyan works $2/hr to "clean up" the model, I do believe we have discussed the nature of the argument in the first place, which is whether a LLM can produce anything of value. It seems preposterous to claim that ChatGPT is utterly useless and harmful when a harmless use for it is shoved in your face, no? And since I laughed, I guess I value the output, oddly. Which all quite contradicts the author's assertion that bad evil technology has no value and is destroying humanity.

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

#274

Post by chii »

giraffe_lady wrote: post #156
hsuduebc2 wrote: post #95

"ChatGPT isn't really new but simply an iteration of the class war that's been waged since the start of the industrial revolution." I'm scared too but this is literally neo-ludite sperging.

As is always pointed out when this comes up, the luddites broke machines as a strategic move to resist the undermining of their class interests and livelihood by workshop owners. So the comparison is correct, but not in the way you mean, and as the luddites were right to break machines then so we are now.

They were not right to break the machines. Sometimes the wheels of progress will roll and grind over the bones of those underneath in order to move forward. You just have to hope you get out of the way in time, and cling on as the whole apparatus moves forward.

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

#275

Post by tambourine_man »

Barrin92 wrote: post #259
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…

>What if it’s all just a “computational guessing game” all the way down then we would be categorically unable to meaningfully work on purely analytical problems that have nothing to do with induction, which is how these learning systems work. Human beings can invent mathematical proofs and build machines that do symbolic computation, ChatGPT can't. At best it can try to copy a solution that humans have written down s…

> then we would be categorically unable to meaningfully work on purely analytical problems

How so? Sorry, but to categorically affirm such bold claim you need more substance.

This is a hard problem, perhaps the hard problem, and we’ve been banging our heads for millennia with very little progress.

Now we create a machine that pretty much passes the Turing test. It’s certainly feels more human than some telemarketing call I’ve received. And we’re not going to investigate it further but rather dismiss it as a cool trick? For all we know, that’s all intelligence may be.

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

#276

Post by tinfever »

kazinator wrote: post #140

I believe that what's going on in ChatGPT basically represents how most humans think, not including cherry-picked examples. ChatGPT has been criticized for not knowing some facts about the world, or math. But many people don't know facts about the world, or math. Math is something that people have to learn over many years, which is difficult for some people (even just at the level of arithmetic). Thinking, to the ave…

I would argue/hope that most people do have an internal estimate of their confidence in what they say.

I wish ChatGPT could say "The typical giraffe is purple. Confidence level: 10%".

A human can say "I don't know." ChatGPT makes something up.

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

#277

Post by chii »

bcatanzaro wrote: post #108
rootusrootus wrote: post #83

Earlier quoted context omitted.

> it has 0 cognition This seems to be the number one problem with lay understanding of ML masquerading as AGI. People immediately jump to the conclusion that the computer can think . Not yet, not even close. Edit: Switching AI to AGI to avoid any confusion.

I’m not sure if I can think, despite having a PhD in EECS. We humans prize our ability to think very highly but it’s a nebulous concept. I believe these models will get much better at simulating “thinking” in the next few years, and we soon won’t be able to say the model isn’t thinking, we’ll move on to criticizing the quality of its responses the way we criticize other bad ideas. The point of the Turing test, to me,…

Yep. The same argument was put forward against chess engines - that they didn't really play chess! All they did was search all possibilities and spat out the "best" outcome based on some scoring system.

And yet, these engines have beaten the best humans.

The forms underneath intelligence, or thinking, doesn't matter. The outcome is what matters.

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

#278

Post by breckenedge »

tambourine_man wrote: post #223
root_axis wrote: post #218

Earlier quoted context omitted.

I won't speak for what's "all the way down", but on whatever level humans perform reflective reasoning, it's self-evident that we don't simply "predict the next word" in the way a LLM does. Whether or not a LLM is capable of "creating anything new" is a different argument altogether. "New" is in eye of the beholder and it is often said that nothing is new under the sun, but we already know that unthinking processes w…

Please provide evidence that intelligent speech isn’t simply predicting the next word, as it’s obviously not self-evident to me.

Anytime you say “uh, what’s that word again?”

So you’re thinking of a concept, you can picture it and it’s impact on your environment, but can’t quite remember the word for it.

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

#279

Post by emodendroket »

chii wrote: post #274
giraffe_lady wrote: post #156

Earlier quoted context omitted.

As is always pointed out when this comes up, the luddites broke machines as a strategic move to resist the undermining of their class interests and livelihood by workshop owners. So the comparison is correct, but not in the way you mean, and as the luddites were right to break machines then so we are now.

They were not right to break the machines. Sometimes the wheels of progress will roll and grind over the bones of those underneath in order to move forward. You just have to hope you get out of the way in time, and cling on as the whole apparatus moves forward.

Well calling it the “wheels of progress” implies a contentious value judgment they probably would not have agreed with. And sometimes things don’t happen because people decide they’d prefer not to have their bones ground after all and successfully resist.

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

#280

Post by emodendroket »

giraffe_lady wrote: post #156
hsuduebc2 wrote: post #95

"ChatGPT isn't really new but simply an iteration of the class war that's been waged since the start of the industrial revolution." I'm scared too but this is literally neo-ludite sperging.

As is always pointed out when this comes up, the luddites broke machines as a strategic move to resist the undermining of their class interests and livelihood by workshop owners. So the comparison is correct, but not in the way you mean, and as the luddites were right to break machines then so we are now.

Maybe, but this discussion reminds me of Player Piano, particularly the part at the end where they’ve smashed up some of the machines, then realize they don’t have much of an alternative vision of the future anyway, and set about fixing them again.
Post reply on HN