[openstreetmap/openstreetmap-website] How should we deal with zero-effort AI PRs? (Issue #6954)

[Mateusz Konieczny]

matkoniecz left a comment (openstreetmap/openstreetmap-website#6954)

> Or should we concentrate our efforts on developers who are genuinely trying to be helpful, and have made at least some demonstrable effort first?

well, obviously yes

the question is probably rather whether LLM PRs should get any review at all and whether their authors should be banned 

If you feel guilty/worried about closing them with no review - I would say that it is a correct response to useless spam.

Because PR posting unreviewed LLM output is simply spam and should be treated as such.

> I'm not sure what the best approach is for these zero-effort PRs?

close as soon as spotted, without wasting time on review?

I am dealing with those also in https://github.com/openstreetmap/id-tagging-schema and in https://github.com/openstreetmap/id

In iD tagging schema numbers there are manageable, but I am getting closer to just closing them on sight with no explanation.

So far I was making minimal comments and then closing PR after getting no response few days later. But it looks like a waste of time, and in all blatant cases just closing them with no comment would be better.

And iD tagging schema is lucky as runaway LLMs are quite obvious, in iD it is often less obvious.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/6954#issuecomment-4180626746
You are receiving this because you are subscribed to this thread.

Message ID: <openstreetmap/openstreetmap-website/issues/6954/4180626746 at github.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/rails-dev/attachments/20260402/972f2b90/attachment.htm>