Skip to content

governance: Add AI/LLM contribution policy #10057

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
jennings wants to merge 1 commit into
base: main
Choose a base branch
from
Open

governance: Add AI/LLM contribution policy #10057

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
37 changes: 37 additions & 0 deletions docs/contributing.md
View file
Open in desktop

@PhilipMetzger PhilipMetzger Aug 26, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

nit: I think this should be here and in a separate document, something like AI_POLICY which we then also add to the website.

Original file line number Diff line number Diff line change
Expand Up @@ -19,6 +19,43 @@ again. In other words, the Google CLA is not scoped to a specific project. If
you sign the Google CLA, it applies to other Google open-source projects as
well (including future projects).

### Use of AI and LLM tools

You are welcome to use generative AI and LLM tools ("AI") to help create
contributions as long as you review and edit what it produces to follow the
project guidelines. However, please do not use AI and agents to communicate with
the community, including Discord, issues, and pull requests.

1. **You are the author.** AI is a tool you may use while creating a
contribution, but you are the author. If a substantial part of your
contribution is generated by AI, please disclose this by adding an
`Assisted-by` trailer at the end of the commit message. Examples:

```
Assisted-by: Claude Opus 4.8
Assisted-by: Kimi K3
Assisted-by: GPT 5.6 Sol
```

2. **You must understand what you are contributing.** If you do not understand
what AI has generated, please do not contribute it. If you would like help
understanding code that is already part of the project, please ask in Discord
or GitHub Discussions.

3. **We want to talk to you, not your AI.** When communicating with the
community, do not paste the output from AI or allow an AI agent to post for you.
This includes, but is not limited to:

- Creating pull requests and responding to feedback
- Creating or replying to GitHub issues
- Posting messages in Discord and GitHub Discussions

Avoid using AI to generate commit messages. Writing the commit message
yourself ensures you understand and agree with what you're writing.

It is acceptable to use AI to translate to English, but please also include
your writing in its original language.

### Commit guidelines
Comment on lines +58 to 59

@PhilipMetzger PhilipMetzger Aug 26, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

nit: This has currently no teeth, since it doesn't provide a way for the project to react to this beside commenting. If we want to have this as a policy there should be a way for core contributors and maintainers to effectively ban or close PRs which don't conform to this.

@jennings jennings Aug 26, 2026

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm glad to consider other opinions, but I made a conscious choice to omit that for now, because:

  • I don't think anything's stopping us from closing a PR even if we don't mention that here.
  • If we do close a PR citing the policy, the author saying, "But the policy doesn't say you'll close the PR" seems a pretty weak counter-argument to me.
  • We're pretty low-touch in general, so I'd rather avoid creating an exact process until we need it

@PhilipMetzger PhilipMetzger Aug 26, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm glad to consider other opinions, but I made a conscious choice to omit that for now, because:

  • I don't think anything's stopping us from closing a PR even if we don't mention that here.
  • If we do close a PR citing the policy, the author saying, "But the policy doesn't say you'll close the PR" seems a pretty weak counter-argument to me.
  • We're pretty low-touch in general, so I'd rather avoid creating an exact process until we need it

In my opinion, a policy like this is not written against the current state of the project (and the still low-touch sentiment people have), its written for the future and any issue which comes along from the AI direction and for that we should empower people we trust to do the right thing.

I would like you to trust me on this since I the last few policies were written from me and that at a large personal cost.

  • If we do close a PR citing the policy, the author saying, "But the policy doesn't say you'll close the PR" seems a pretty weak counter-argument to me.

And that's how you shouldn't do it (as in explicitly writing "we're going to close PRs and hide comments etc..."), but just encoding the right to act on it.

@senekor senekor Aug 26, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Stephens reasoning makes sense to me. If experience shows that enforcement isn't working, the policy can be adjusted.

I would like you to trust me on this since I the last few policies were written from me and that at a large personal cost.

"Trust me" isn't an argument. Decisions in open-source aren't made based on authority. And what's with the comment about personal cost? I can't see how that's relevant to the topic at hand.


Unlike many GitHub projects (but like many VCS projects), we care more about the
Expand Down