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

LLM Inevitabilism

tomrenner.com

Post reply on HN
Print view View original

211–220 of 1001 posts

Re: LLM Inevitabilism

#211

Post by 01HNNWZ0MV43FF »

petetnt wrote: post #143
mg wrote: post #70

In the 90s a friend told me about the internet. And that he knows someone who is in a university and has access to it and can show us. An hour later, we were sitting in front of a computer in that university and watched his friend surfing the web. Clicking on links, receiving pages of text. Faster than one could read. In a nice layout. Even with images. And links to other pages. We were shocked. No printing, no shipp…

There’s always a distinct lack of the names in the posts like this. What was the library that was being changed to what? You say it had ”no good documentation”, but it clearly has some sort of documentation considering the LLM did such a good job on the rewrite. Do you understand the ”large library” now?

It would be so funny if the library is like, curl

Re: LLM Inevitabilism

#212

Post by globular-toast »

Did anyone even read the article? Maybe you should get an LLM to bullet point it for you.

The author isn't arguing about whether LLMs (or AI) is inevitable or not. They are saying you don't have to operate within their framing. You should be thinking about whether this thing is really good for us and not just jumping on the wagon and toeing the line because you're told it's inevitable.

I've noticed more and more the go to technique for marketing anything now is FOMO. It works. Don't let it work on you. Don't buy into a thing just because everyone else is. Most of the time you aren't missing out on anything at all. Some of the time the thing is actively harmful to the participants and society.

Re: LLM Inevitabilism

#213

Post by AndyKelley »

mg wrote: post #70

In the 90s a friend told me about the internet. And that he knows someone who is in a university and has access to it and can show us. An hour later, we were sitting in front of a computer in that university and watched his friend surfing the web. Clicking on links, receiving pages of text. Faster than one could read. In a nice layout. Even with images. And links to other pages. We were shocked. No printing, no shipp…

You speak with a passive voice, as if the future is something that happens to you, rather than something that you participate in.

Re: LLM Inevitabilism

#214

Post by globular-toast »

killerstorm wrote: post #197
sircastor wrote: post #172

The hardest part about inevitablism here is that the people who are making the argument this is inevitable are the same people who are the people who are shoveling hundreds of millions of dollars into it. Into the development, the use, the advertisement. The foxes are building doors into the hen houses and saying there’s nothing to be done, foxes are going to get in so we might as well make it something that works fo…

"put your money where your mouth is" is generally a good thing.

Except "the money" in this case is just part of funds distributed around by the super rich. The saying works better when it's about regular people actually taking risks and making sacrifices.

Re: LLM Inevitabilism

#215

Post by Karrot_Kream »

cmdli wrote: post #204
chadcmulligan wrote: post #149

Earlier quoted context omitted.

Any code thats easy to define and tedious I just get AI's to do it now, and its awesome. Saves me so much work, though you have to read the code, it still puts in odd stuff sometimes.

How much of the code you are writing is tedious? If its a significant amount, the framework you are using could use some improvement.

Maybe?

In some cases, definitely. Then good luck making the business case to improve the framework or swap and refactor around a different framework. (Or you can do what I do during the more motivated/less busy times in my life: find undisturbed unpaid time to do it for your team.)

In other cases improving the framework comes at the cost of some magic that may obscure the intent of the code.

The nice thing about LLM code is that it's code. You're not monkey patching a method. You're not subtly changing the behavior of a built-in. You're not adding a build step (though one can argue that LLM generated code is akin to a separate build step.) You're just checking in code. Other contributors can just read the code.

Re: LLM Inevitabilism

#216

Post by rgoulter »

imiric wrote: post #112
Animats wrote: post #46

There may be an "LLM Winter" as people discover that LLMs can't be trusted to do anything. Look for frantic efforts by companies to offload responsibility for LLM mistakes onto consumers. We've got to have something that has solid "I don't know" and "I don't know how to do this" outputs. We're starting to see reports of LLM usage having negative value for programmers, even though they think it's helping. Too much eff…

> Look for frantic efforts by companies to offload responsibility for LLM mistakes onto consumers. Not just by companies. We see this from enthusiastic consumers as well, on this very forum. Or it might just be astroturfing, it's hard to tell. The mantra is that in order to extract value from LLMs, the user must have a certain level of knowledge and skill of how to use them. "Prompt engineering", now reframed as "con…

A couple of typical comments about LLMs would be:

"This LLM is able to capably output useful snippets of code for Python. That's useful."

and

"I tried to get an LLM to perform a niche task with a niche language, it performed terribly."

I think the right synthesis is that there are some tasks the LLMs are useful at, some which they're not useful at; practically, it's useful to be able to know what they're useful for.

Or, if we trust that LLMs are useful for all tasks, then it's practically useful to know what they're not good at.

Re: LLM Inevitabilism

#217

Post by salviati »

AndyKelley wrote: post #213
mg wrote: post #70

In the 90s a friend told me about the internet. And that he knows someone who is in a university and has access to it and can show us. An hour later, we were sitting in front of a computer in that university and watched his friend surfing the web. Clicking on links, receiving pages of text. Faster than one could read. In a nice layout. Even with images. And links to other pages. We were shocked. No printing, no shipp…

You speak with a passive voice, as if the future is something that happens to you, rather than something that you participate in.

Isn't it kind of both?

Did luddites ever have a chance of stopping the industrial revolution?

Re: LLM Inevitabilism

#218

Post by wisty »

cmdli wrote: post #204
chadcmulligan wrote: post #149

Earlier quoted context omitted.

Any code thats easy to define and tedious I just get AI's to do it now, and its awesome. Saves me so much work, though you have to read the code, it still puts in odd stuff sometimes.

How much of the code you are writing is tedious? If its a significant amount, the framework you are using could use some improvement.

Sometimes frameworks are a little too magic. Think raw sql vs a magic orm. No I'm not saying don't use an orm, but when everything ends up as magic meta configuration it's sometimes too much. Sometimes making things a little explicit can make it more flexible going forward.

Even if the framework is good, an llm can read the docs faster than you. Probably it's important to understand things in a lot of cases, but sometimes you just need to get it working without really reading the framework source or docs.

Re: LLM Inevitabilism

#219

Post by chadcmulligan »

cmdli wrote: post #204
chadcmulligan wrote: post #149

Earlier quoted context omitted.

Any code thats easy to define and tedious I just get AI's to do it now, and its awesome. Saves me so much work, though you have to read the code, it still puts in odd stuff sometimes.

How much of the code you are writing is tedious? If its a significant amount, the framework you are using could use some improvement.

Yeah its not a huge amount, but its a start. eg - just got it to make me a class in Lua with names for all the colours. It went and named all the colors and did a very nice job (Claude) - it would have taken me ages to go and find the names, sort them out etc, and I was avoiding the work, cause its tedious. I've got it to make me windows controls and data structures, parsers all well defined stuff.

I think the problem comes about when it doesn't know the context you're in - give me a list of colour names is well defined, and I assume the LLM's would have read a million pages with this done, so its easy for it to do this. Doing something more exotic that it hasn't seen a lot, then you'll get weird results.

Re: LLM Inevitabilism

#220

Post by fendy3002 »

trashchomper wrote: post #108
fendy3002 wrote: post #68

Earlier quoted context omitted.

LLMs need significant optimization or we get significant improvement on computing power while keeping the energy cost the same. It's similar with smartphone, when at the start it's not feasible because of computing power, and now we have one that can rival 2000s notebooks. LLMs is too trivial to be expensive EDIT: I presented the statement wrongly. What I mean is the use case for LLM are trivial things, it shouldn't…

Calling LLMs trivial is a new one. Yea just consume all of the information on the internet and encode it into a statistical model, trivial, child could do it /s

well I presented the statement wrongly. What I mean is the use case for LLM are trivial things, it shouldn't be expensive to operate
Post reply on HN