docs: AI contribution policy and PR disclosure levels#430
docs: AI contribution policy and PR disclosure levels#430Just-Insane wants to merge 5 commits intodevfrom
Conversation
CONTRIBUTING.md
Outdated
| ## AI-Assisted Development | ||
|
|
||
| We expect and appreciate authentic engagement in our community. | ||
| We welcome AI-assisted contributions. Tools like GitHub Copilot, Cursor, Claude, ChatGPT, and others are legitimate parts of a modern development workflow, and we don't ask you to hide their use. |
There was a problem hiding this comment.
hmmm that might be understood as us endorsing ai generated code.
i liked the previous version more.
There was a problem hiding this comment.
agree, we should stay neutral.
There was a problem hiding this comment.
yeah the new version somehow sounds aggressively pro-ai.
like we would endorse it, idk if that's what we want to do
There was a problem hiding this comment.
Disagree, this language is used to explicitly take a stance that can be pointed to when people make complaints.
The last section does need to be re-written though.
There was a problem hiding this comment.
i think i just have to disagree. it’s too pro. we should be more neutral.
CONTRIBUTING.md
Outdated
| - **Add genuine value.** Wrapping a prompt in a PR without understanding or verifying the output is not a contribution. Maintainers can tell the difference and will close PRs that don't meet the bar. | ||
|
|
||
| Make sure you have added value based on your personal competency to your contributions. Just taking some input, feeding it to an AI and posting the result is not of value to the project. We reserve the right to rigorously reject seemingly AI generated low-value contributions. | ||
| When opening a pull request, **disclose your level of AI use** using the checklist in the PR template. This isn't gatekeeping — it's transparency that helps reviewers calibrate their review depth. A heavily AI-generated PR will get more scrutiny; that's expected and fine. |
There was a problem hiding this comment.
i would keep the part with adding value based on personal competency in there.
There was a problem hiding this comment.
There was a problem hiding this comment.
ig i just liked the old phrasing more, idk
Clarified guidelines on AI-assisted contributions, emphasizing transparency in AI usage.
|
Not a maintainer but a user of Sable. Im rather opposed of the use of generative AI. It includes a lot of issues (code quality, performance, and usage of licensed code are the big one). Sable has been doing fine without it, I dont really see what it would bring or add to the project (except the issues listed above). |
Code quality, testing, reviews, etc are out of scope for this PR and will be addressed separately. Additionally, there is AI generated code currently in Sable. |
dozro
left a comment
There was a problem hiding this comment.
seems fine now, even tho i would’ve kept the heading as “restrictions on gen ai” but seems fine :3
|
i think this works now |
|
wait why was this closed?? |
Because I am done with this community after the last couple days. This place is toxic and I will not put up with that. I'm sure you all can figure out a path forward, but I will no longer be interacting with this community. If anyone wants to keep in touch, my mxid is: |
|
This may be okay to merge? But I can't support/update it going forward. You can merge it if you think it is useful. |
Summary
Updates the "Generative AI Usage" section in
CONTRIBUTING.mdto append a note about disclosing AI use in PRs. Also updates the PR template with a four-level AI disclosure checklist and fixes the stale Cinny link.Changes
CONTRIBUTING.md.github/PULL_REQUEST_TEMPLATE.mdType of change
AI assistance level
Checklist: