←back to thread

LLM Inevitabilism

(tomrenner.com)
1613 points SwoopsFromAbove | 2 comments | 15 Jul 25 04:35 UTC | HN request time: 0.001s | source
Show context
mg ◴[15 Jul 25 05:41 UTC] No.44568158[source]▶
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 shipping, no waiting. This was the future. It was inevitable.

Yesterday I wanted to rewrite a program to use a large library that would have required me to dive deep down into the documentation or read its code to tackle my use case. As a first try, I just copy+pasted the whole library and my whole program into GPT 4.1 and told it to rewrite it using the library. It succeeded at the first attempt. The rewrite itself was small enough that I could read all code changes in 15 minutes and make a few stylistic changes. Done. Hours of time saved. This is the future. It is inevitable.

PS: Most replies seem to compare my experience to experiences that the responders have with agentic coding, where the developer is iteratively changing the code by chatting with an LLM. I am not doing that. I use a "One prompt one file. No code edits." approach, which I describe here:

https://www.gibney.org/prompt_coding

replies(58): >>44568182 #>>44568188 #>>44568190 #>>44568192 #>>44568320 #>>44568350 #>>44568360 #>>44568380 #>>44568449 #>>44568468 #>>44568473 #>>44568515 #>>44568537 #>>44568578 #>>44568699 #>>44568746 #>>44568760 #>>44568767 #>>44568791 #>>44568805 #>>44568823 #>>44568844 #>>44568871 #>>44568887 #>>44568901 #>>44568927 #>>44569007 #>>44569010 #>>44569128 #>>44569134 #>>44569145 #>>44569203 #>>44569303 #>>44569320 #>>44569347 #>>44569391 #>>44569396 #>>44569574 #>>44569581 #>>44569584 #>>44569621 #>>44569732 #>>44569761 #>>44569803 #>>44569903 #>>44570005 #>>44570024 #>>44570069 #>>44570120 #>>44570129 #>>44570365 #>>44570482 #>>44570537 #>>44570585 #>>44570642 #>>44570674 #>>44572113 #>>44574176 #
bambax ◴[15 Jul 25 07:35 UTC] No.44568844[source]▶
The problem with LLM is when they're used for creativity or for thinking.

Just because LLMs are indeed useful in some (even many!) context, including coding, esp. to either get something started, or, like in your example, to transcode an existing code base to another platform, doesn't mean they will change everything.

It doesn't mean “AI is the new electricity.” (actual quote from Andrew Ng in the post).

More like AI is the new VBA. Same promise: everyone can code! Comparable excitement -- although the hype machine is orders of magnitude more efficient today than it was then.

replies(5): >>44568939 #>>44568982 #>>44569154 #>>44569340 #>>44569371 #
mettamage ◴[15 Jul 25 09:10 UTC] No.44569340{3}[source]▶
> everyone can code!

I work directly with marketers and even if you give them something like n8n, they find it hard to be precise. Programming teaches you a "precise mindset" that one doesn't have when they aren't really thinking about tech professionally.

I wonder if seasoned UX designers can code now. They do think professionally about software. I wonder if it's at a deep enough granularity such that they can simply use natural language to get something to work.

replies(2): >>44569445 #>>44574164 #
petra ◴[15 Jul 25 09:28 UTC] No.44569445{4}[source]▶
Can an LLM detect a lack of precision and point it to you ?
replies(3): >>44569512 #>>44569524 #>>44569554 #
1. TheOtherHobbes ◴[15 Jul 25 09:50 UTC] No.44569554{5}[source]▶
Sometimes, yes. Reliably, no.

LLMs don't have enough of a model of the world to understand anything. There was a paper floating around recently about how someone trained an ML system on orbital dynamics. The result was a system that could calculate orbits correctly, but it completely failed to extract the underlying - simple - math. Instead it basically frankensteined together its own system of epicycles which solved a very narrow range of problems but lacked any generality.

Any coding has the same problems. Sometimes you get lucky, sometimes you don't. And if you strap on an emulator and test rig and allow the machine to flail around inside it, sometimes working code falls out.

But there's no abstracted model of software development as a process in there, either in theory or practise. And no understanding of vague goals with constraints and requirements that can be inferred creatively from outside the training data.

replies(1): >>44575715 #
2. antonvs ◴[15 Jul 25 20:55 UTC] No.44575715[source]▶
> LLMs don't have enough of a model of the world to understand anything.

This is binary thinking, and it's fallacious.

For your orbital mechanics example, sure, it's difficult for LLMs to develop good models of the physical world, in large part because they aren't able to interact with the world directly and have to rely on human texts to describe it to them.

For your software development example, you're making a similar mistake: the fact that their strongest suit is not producing fully working systems doesn't mean that they have no world model, or that their successes are as random as you seem to think ("Sometimes you get lucky, sometimes you don't," "sometimes working code falls out.")

But if you try, for example, asking an LLM to identify a bug in a program, or ask it questions about how a program works, you'll find that from a functional perspective, they exhibit excellent understanding that strongly implies a good world model. You may be taking your own thought processes for granted too much to realize how good they are at this. The idea that "there's no abstracted model of software development as a process in there" is hard to reconcile with the often superhuman responses they're capable of, when you use them in the scenarios they're most effective at.