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

I genuinely don't understand why some people are still bullish about LLMs

twitter.com

Post reply on HN
Print view View original

471–480 of 1001 posts

Re: I genuinely don't understand why some people are still bullish about LLMs

#471

Post by kristiandupont »

TZubiri wrote: post #416
Animats wrote: post #178

Earlier quoted context omitted.

> inbuilt into the models of all LLMs is that they'll fabricate a lot. Still the elephant in the room. We need an AI technology that can output "don't know" when appropriate. How's that coming along?

I always see this, and I always answer the same. This exists, each next token has a probability assigned to it. High probability means "it knows", if there's two or more tokens of similar probability, or the prob of the first token is low in general, then you are less confident about that datum. Of course there's areas where there's more than one possible answer, but both possibilities are very consistent. I feel LLM…

>can we stop pretending with the generic name for ChatGPT?

What? I use several LLM's, including ChatGPT, every day. It's not like they have it all cornered..

Re: I genuinely don't understand why some people are still bullish about LLMs

#472

Post by Dylan16807 »

nemothekid wrote: post #328
jasondigitized wrote: post #188

Earlier quoted context omitted.

Anyone who doesn't understand this either isn't required to use to utility it provides or has no idea how to prompt it correctly. My wife is a bookkeeper. There are some tasks that are a pain in the ass without writing some custom code. In her case, we just saved her about 2 hours by asking Claude to do it. It wrote the code, applied the code to a CSV we uploadrd and gave us exactly what we needed in 2 minutes.

> Anyone who doesn't understand this either isn't required to use to utility it provides or has no idea how to prompt it correctly. Almost every counter-criticism of LLMs almost boil down to 1. you're holding it wrong 2. Well, I use it $DAYJOB and it works great for me! (And $DAYJOB is software engineering). I'm glad your wife was able to save 2 hours of work, but forgive me if that doesn't translate to the trillion…

Boiling down to a couple cases would be more useful if you actually tried to disprove those cases or explain why they're not good enough.

> It's strange you don't see the inherent irony in your post. Instead of your wife just directly uploading the dataset and a prompt, she first has to prompt it to write code. There are clear limitations and it looks like LLMs are stuck at some sort of wall.

What's ironic about that? That's such a tiny imperfection. If that's anything near the biggest flaw then things look amazing. (Not that I think it is, but I'm not here to talk about my opinion, I'm here to talk about your irony claim.)

Re: I genuinely don't understand why some people are still bullish about LLMs

#473

Post by alabastervlog »

cratermoon wrote: post #420
gilbetron wrote: post #109

I get so confused on this. I play around, test, and mess with LLMs all the time and they are miraculous. Just amazing, doing things we dreamed about for decades. I mean, I can ask for obscure things with subtle nuance where I misspell words and mess up my question and it figures it out. It talks to me like a person. It generates really cool images. It helps me write code. And just tons of other stuff that astounds me…

People were impressed by Eliza in 1964.

I’m reminded of how I always think current cutting edge good examples of CG in movies looks so real and then, consistently, when I watch it again in 10 years it always looks distractingly shitty.

Re: I genuinely don't understand why some people are still bullish about LLMs

#474

Post by fergie »

gilbetron wrote: post #109

I get so confused on this. I play around, test, and mess with LLMs all the time and they are miraculous. Just amazing, doing things we dreamed about for decades. I mean, I can ask for obscure things with subtle nuance where I misspell words and mess up my question and it figures it out. It talks to me like a person. It generates really cool images. It helps me write code. And just tons of other stuff that astounds me…

I feel like both standpoints are true. Yes the tech is miraculous, and yes it has a very long way to go.

Re: I genuinely don't understand why some people are still bullish about LLMs

#475

Post by kragen »

chii wrote: post #426
nonethewiser wrote: post #355

Earlier quoted context omitted.

> What's worse, people are treating them as authoritative. … I've both seen online and heard people quote LLM output as if it were authoritative. Thats not an LLM problem. But indeed quite bothersome. Dont tell me what Chatgpt told you. Tell me what you know. Maybe you got it from ChatGPT and verified it. Great. But my jaw kind of drops when people cite an LLM and just assume it’s correct.

> when people cite an LLM and just assume it’s correct. people used to say the exact same thing with wikipedia back when it first started.

These are not similar. Wikipedia says the same thing to everybody, and when what it says is wrong, anybody can correct it, and they do. Consequently it's always been fairly reliable.

Re: I genuinely don't understand why some people are still bullish about LLMs

#476

Post by ddejohn »

agentultra wrote: post #398
gilbetron wrote: post #109

I get so confused on this. I play around, test, and mess with LLMs all the time and they are miraculous. Just amazing, doing things we dreamed about for decades. I mean, I can ask for obscure things with subtle nuance where I misspell words and mess up my question and it figures it out. It talks to me like a person. It generates really cool images. It helps me write code. And just tons of other stuff that astounds me…

It’s not that impressive to me, as a programmer. The crisis in programming hasn’t been writing code. It has been developing languages and tools so that we can write less of it that is easy to verify as correct. These tools generate more code. More than you can read and more than you will want to before you get bored and decide to trust the output. It is trained on the most average code available that could be sucked…

I can't believe I had to dig this deep to find this comment.

I have yet to see an AI-generated image that was "really cool".

AI images and videos strike me as the coffee pods of the digital world -- we're just absolutely littering the internet with garbage. And as a bonus, it's also environmentally devastating to the real world!

I live nearby a landfill, and go there often to get rid of yard waste, construction materials, etc. The sheer volume of perfectly serviceable stuff people are throwing out in my relatively small city (stuff humans consume and dispose. I hope people are noticing just how much more full of trash the internet has become in the last few years. It seems like it, but then I read this thread full of people that are still hyped about it all and I wonder.

This isn't even to mention the generated text... it's all just so inane and I just don't get it. I've tried a few times to ask for relatively simple code and the results have been laughable.

Re: I genuinely don't understand why some people are still bullish about LLMs

#477

Post by kristiandupont »

worthless-trash wrote: post #366
cogman10 wrote: post #161

Earlier quoted context omitted.

> Wah, it can't write code like a Senior engineer with 20 years of experience! No, that's not my problem with it. My problem with it is that inbuilt into the models of all LLMs is that they'll fabricate a lot. What's worse, people are treating them as authoritative. Sure, sometimes it produces useful code. And often, it'll simply call the "doTheHardPart()" method. I've even caught it literally writing the wrong algor…

I saw someone saying 80% of doctors believe that LLM's are trustworthy consultation partners. Code created by LLM's doesnt compile, hallucinated API's.. invalid syntax and completely broken logic, why would you trust it with someones life !

>I saw someone saying 80% of doctors believe that LLM's are trustworthy consultation partners.

See, now that is something I don't know why I should trust: a random person on the internet citing a statistics that they saw someone else say.

Re: I genuinely don't understand why some people are still bullish about LLMs

#478

Post by oakpond »

I'm also not bullish on this. In the sense that I don't think LLMs are going to get 10x better, but they are useful for what they can do already.

If I see what Copilot suggests most of the time, I would be very uncomfortable using it for vibe coding though. I think it's going to be... entertaining watching this trend take off. I don't really fear I'm going to lose my job soon.

I'm skeptical that you can build a business on a calculator that's wrong 10% of the time when you're using it 24/7. You're gonna need a human who can do the math.

Re: I genuinely don't understand why some people are still bullish about LLMs

#479

Post by geysersam »

cogman10 wrote: post #161
gilbetron wrote: post #109

I get so confused on this. I play around, test, and mess with LLMs all the time and they are miraculous. Just amazing, doing things we dreamed about for decades. I mean, I can ask for obscure things with subtle nuance where I misspell words and mess up my question and it figures it out. It talks to me like a person. It generates really cool images. It helps me write code. And just tons of other stuff that astounds me…

> Wah, it can't write code like a Senior engineer with 20 years of experience! No, that's not my problem with it. My problem with it is that inbuilt into the models of all LLMs is that they'll fabricate a lot. What's worse, people are treating them as authoritative. Sure, sometimes it produces useful code. And often, it'll simply call the "doTheHardPart()" method. I've even caught it literally writing the wrong algor…

> What's worse, people are treating them as authoritative.

So what? People are wrong all the time. What happens when people are wrong? Things go wrong. What happens then? People learn that the way they got their information wasn't robust enough and they'll adapt to be more careful in the future.

This is the way it has always worked. But people are "worried" about LLMs... Because they're new. Don't worry, it's just another tool in the box, people are perfectly capable of being wrong without LLMs.

Re: I genuinely don't understand why some people are still bullish about LLMs

#480

Post by eptcyka »

gilbetron wrote: post #109

I get so confused on this. I play around, test, and mess with LLMs all the time and they are miraculous. Just amazing, doing things we dreamed about for decades. I mean, I can ask for obscure things with subtle nuance where I misspell words and mess up my question and it figures it out. It talks to me like a person. It generates really cool images. It helps me write code. And just tons of other stuff that astounds me…

For exploring topics in a shallow fashion is fine with LLMs, doing anything deep is just too unreliable due to hallucination. All models I’ve talked to desperately want to give a positive answer, and thus will often just lie.
Post reply on HN