←back to thread

Define policy forbidding use of AI code generators

(github.com)
491 points todsacerdoti | 4 comments | 25 Jun 25 23:26 UTC | HN request time: 0.476s | source
Show context
Havoc ◴[25 Jun 25 23:41 UTC] No.44382839[source]▶
I wonder whether the motivation is really legal? I get the sense that some projects are just sick of reviewing crap AI submissions
replies(6): >>44382854 #>>44382954 #>>44383005 #>>44383017 #>>44383164 #>>44383177 #
SchemaLoad ◴[25 Jun 25 23:44 UTC] No.44382854[source]▶
This could honestly break open source, with how quickly you can generate bullshit, and how long it takes to review and reject it. I can imagine more projects going the way of Android where you can download the source, but realistically you can't contribute as a random outsider.
replies(5): >>44382866 #>>44382874 #>>44383174 #>>44383418 #>>44385273 #
api ◴[25 Jun 25 23:48 UTC] No.44382866[source]▶
Quality contributions to OSS are rare unless the project is huge.
replies(1): >>44382922 #
1. loeg ◴[25 Jun 25 23:54 UTC] No.44382922[source]▶
Historically the opposite of quality contributions has been no contributions, not net-negative contributions (random slop that costs more in review than it provides benefit).
replies(2): >>44383133 #>>44387502 #
2. lmm ◴[26 Jun 25 00:34 UTC] No.44383133[source]▶
No it hasn't? Net-negative contributions to open source have been extremely common for years, it's not like you need an LLM to make them.
replies(1): >>44383393 #
3. loeg ◴[26 Jun 25 01:23 UTC] No.44383393[source]▶
I guess we've had very different experiences!
4. LtWorf ◴[26 Jun 25 13:58 UTC] No.44387502[source]▶
Nah. I've had a lot of bad contributions. One PR deleted and readded all of the lines in the project, and the entire test suite was failing.

The person got upset at me for saying I could not accept such a thing.

There's other examples.