←back to thread

AI tooling must be disclosed for contributions

(github.com)
728 points freetonik | 2 comments | 21 Aug 25 18:49 UTC | HN request time: 0s | source
Show context
Waterluvian ◴[21 Aug 25 19:07 UTC] No.44976790[source]▶
I’m not a big AI fan but I do see it as just another tool in your toolbox. I wouldn’t really care how someone got to the end result that is a PR.

But I also think that if a maintainer asks you to jump before submitting a PR, you politely ask, “how high?”

replies(16): >>44976860 #>>44976869 #>>44976945 #>>44977015 #>>44977025 #>>44977121 #>>44977142 #>>44977241 #>>44977503 #>>44978050 #>>44978116 #>>44978159 #>>44978240 #>>44978311 #>>44978533 #>>44979437 #
sheepscreek ◴[21 Aug 25 21:12 UTC] No.44978159[source]▶
We keep talking about “AI replacing coders,” but the real shift might be that coding itself stops looking like coding. If prompts become the de facto way to create applications/developing systems in the future, maybe programming languages will just be baggage we’ll need to unlearn.

Programming languages were a nice abstraction to accommodate our inability to comprehend complexity - current day LLMs do not have the same limitations as us.

The uncomfortable part will be what happens to PRs and other human-in-the-loop checks. It’s worthwhile to consider that not too far into the future, we might not be debugging code anymore - we’ll be debugging the AI itself. That’s a whole different problem space that will need an entirely new class of solutions and tools.

replies(2): >>44978213 #>>44978727 #
1. ryoshu ◴[21 Aug 25 21:17 UTC] No.44978213[source]▶
All we need to do is prompt an LLM with such specificity that it does exactly what we want the machine to do.
replies(1): >>44980878 #
2. kentm ◴[22 Aug 25 03:54 UTC] No.44980878[source]▶
Good idea! We can have some sort of standard grammar that we use to prompt the LLM such that it deterministically gives us the result we ask for. We then constrain all prompts to match that grammar. Some sort of language describing programs.