HN Explorer

Ask HN: How are you attributing your AI usage when developing software?

4 points by hartleybrody ↗ HN
I have started a convention of prefixing a commit message with "prompt:" for commits that are entirely AI generated (which I have read through and approved) and then I make my own manual changes and iterations in subsequent commits.

I feel like this is a reasonable attempt at disclosure, but it also seems like it's lacking. I usually include a summary, not the entire prompt (how would I included tool use?) I don't say which models or "agent" was used (cursor sidebar chat vs calude code, etc). Not sure if this matters though?

How have you been managing this in your own workflows?

4 comments

[ 3.1 ms ] story [ 24.5 ms ] thread
not_your_vase
I don't think it's any different than attributing the compiler or the text editor. I don't do either of those, and I also don't attribute LLM generated code.
ailabs_hq
Do we need to unless you want to indemnify yourself
mikewarot
At the top of my README.md

IMPORTANT: This is almost entirely written by ChatGPT 5, not me

kamphey
I don't think it matters much. Sometimes I do consulting where I'm coding at the same time as the client is there with me. There's lots of little questions and tons of testing involved as I do coding in Google Apps Script in Google Sheets. And I'll quickly say "oh I think ChatGPT can figure this out faster than me" and just prompt it at the same time. Then copy paste and edit and figure out if it's right code for that part.

Not only do they not care, they think it's pretty cool that I can essentially save them an hour here or there.