←back to thread

LLM Inevitabilism

(tomrenner.com)
1611 points SwoopsFromAbove | 1 comments | 15 Jul 25 04:35 UTC | HN request time: 0.305s | 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 #
AndyKelley ◴[15 Jul 25 07:08 UTC] No.44568699[source]▶
You speak with a passive voice, as if the future is something that happens to you, rather than something that you participate in.
replies(7): >>44568718 #>>44568811 #>>44568842 #>>44568904 #>>44569270 #>>44569402 #>>44570058 #
1. tankenmate ◴[15 Jul 25 07:46 UTC] No.44568904[source]▶
I have a parallel to suggest; I know it's the rhetorical tool of analogous reasoning, but it deeply matches the psychology of the way most people think. Just like getting to a "certain" number of activated parameters in a model (for some "simple" tasks like summarisation) can be as low as 1.8 billion, once that threshold is breached the "emergent" behaviour of "reasonable", "contextual", or "lucid" text is achieved; or to put this in layman's terms, once your model is "large enough" (and this is quite small compared to the largest models currently in daily use by millions) the generated text goes from jibberish to uncanny valley to lucid text quite quickly.

In the same way once a certain threshold is reached in the utility of AI (in a similar vein to the "once I saw the Internet for the first time I knew I would just keep using it") it becomes "inevitable"; it becomes a cheaper option than "the way we've always done it", a better option, or some combination of the two.

So, as is very common in technological innovation / revolution, the question isn't will it change the way things are done so much as where will it shift the cost curve? How deeply will it displace "the way we've always done it"? How many hand weaved shirts do you own? Joseph-Marie Jacquard wants to know (and King Cnut has metaphorical clogs to sell to the Luddites).