[ADMIN] Adding an MDAnalysis AI tools policy. #5210

IAlibay · 2026-01-18T21:30:28Z

A policy for AI generated tools.

LLM / AI generated code disclosure

A likely-LLM powered tool in google docs was used for spell checking and pointing out gramatical issues.

PR Checklist

[-] Issue raised/referenced?
[-] Tests updated/added?
Documentation updated/added?
[-] package/CHANGELOG file updated?
Is your name in package/AUTHORS? (If it is not, add it!)
LLM/AI disclosure was updated.

Developers Certificate of Origin

I certify that I can submit this code contribution as described in the Developer Certificate of Origin, under the MDAnalysis LICENSE.

📚 Documentation preview 📚: https://mdanalysis--5210.org.readthedocs.build/en/5210/

Adds a markdown document that defines MDAnalysis' current stance on AI tools for contributions.

codecov · 2026-01-18T21:43:57Z

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 92.72%. Comparing base (4c12f15) to head (1615ac7).
⚠️ Report is 2 commits behind head on develop.

Additional details and impacted files

@@           Coverage Diff            @@
##           develop    #5210   +/-   ##
========================================
  Coverage    92.72%   92.72%           
========================================
  Files          180      180           
  Lines        22475    22475           
  Branches      3190     3190           
========================================
  Hits         20841    20841           
  Misses        1177     1177           
  Partials       457      457

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

RMeli

Thanks @IAlibay for getting this started!

First pass. I'll post comments about content in the coming days.

AI_POLICY.md

Co-authored-by: Rocco Meli <r.meli@bluemail.ch>

orbeckst

I'd like to also state that we're open to discussions and expect the policy to be periodically reviewed.

I also have one smaller edit.

Thank you very much for drafting @IAlibay !

orbeckst · 2026-01-19T01:07:59Z

AI_POLICY.md

+AI assitance is deemed acceptable. However, if code generate exceeds minimal, sporadic amounts (e.g. repeated or large multi-line blocks),
+it would be considered fully AI-generated and, as defined in section #1, is not acceptable.
+
+As per section #0, where possible please state that you are using AI assistance via an IDE.


Remove "where possible" as this puts the burden on us. Instead, it should be a contributors responsibility to know when they are using AI features.

Suggested change

As per section #0, where possible please state that you are using AI assistance via an IDE.

As per section #0, please state when you are using AI assistance via an IDE.

With IDEs, my main worry is that it's just easy to forget - like I'm a vim guy, but sometimes I use vscode and then the auto complete happens but I forget about it.

It's still on the user to know. I don't think the policy is the place to say "we know, you're human, so we are putting qualifiers on our requirements". We should say what we want to happen.

A checklist will make this easier.

I think this may be something to discuss on Friday. For now, I have pushed a different variant of that text which hopefully may offer a middle ground.

AI_POLICY.md

tylerjereddy

FWIW, the sklearn PR template currently has this:

AI usage disclosure

I used AI assistance for:

Code generation (e.g., when writing an implementation or fixing a bug)
Test/benchmark generation
Documentation (including examples)
Research and understanding

AI_POLICY.md

BradyAJohnston · 2026-01-19T20:30:28Z

FWIW, the sklearn PR template currently has this:

I really like this approach. It gives people more of a chance to acknowledge usage of AI in different aspects where they might initially say "no AI" for a single response. Potentially saves us some time in review.

IAlibay · 2026-01-19T20:39:27Z

P.S. I'm going to merge in grammatical corrections, but not content just yet - that way everyone gets a chance to comment on the same thing.

Correcting some typos found during review.

orbeckst · 2026-01-28T18:11:14Z

In addition to the suggested disclosure categories from #5210 (review)

Code generation (e.g., when writing an implementation or fixing a bug)
Test/benchmark generation
Documentation (including examples)
Research and understanding
Tooling, Packaging, Continuous Integration

I'd also add a category on tooling/CI. Even if this is more a coredev thing to touch, we should be clear ourselves.

IAlibay · 2026-01-29T12:45:13Z

In addition to the suggested disclosure categories from #5210 (review)

Code generation (e.g., when writing an implementation or fixing a bug)

Test/benchmark generation

Documentation (including examples)

Research and understanding

Tooling, Packaging, Continuous Integration

I'd also add a category on tooling/CI. Even if this is more a coredev thing to touch, we should be clear ourselves.

I'll be honest, this makes for a really large disclosure list in the PR template. Is it really needed? It's not like we allow for much nuance in the current draft of the policy.

AI_POLICY.md

orbeckst · 2026-01-29T16:13:22Z

I'll be honest, this makes for a really large disclosure list in the PR template. Is it really needed? It's not like we allow for much nuance in the current draft of the policy.

That's a fair comment. Given that we'll have a larger discussion soon, I can certainly see not including something like the disclosures in the first round and start with a strict policy.

Longer term I am convinced that we will have to adapt a more nuanced view, simply because for routine tasks, genAI use is just so much more efficient when wielded by a knowledgeable person. I would not want to ham-string our own coredevs. I assume that in the future we'll adopt a more nuanced view.

Perhaps it helps if I offer my current (and evolving) views on these individual points. For all these points I expect someone using AI to actually understand the issue and they should have been able to do it themselves. We reserve the "humans are right"-prerogative for discussions and decisions. With these caveats, I don't see AI use for Research/Understanding a problem. For boiler-plate docs I don't see an issue, as long as we're not generating whole novels with original content, and as long as submitters actually proof-read what they submit. I'd also argue that updating boiler-plate code in workflows and tooling can be more efficiently done with AI help — all of this areas where we are stretched quite thin.

Test and benchmark generation is a bit more involved. We have to ask what this policy should do: Is it supposed to shield coredevs from having to deal with low-quality and low-effort code contributions or is it supposed to maintain license integrity of these parts of the code? The answer to this question is important in how to approach it.

orbeckst · 2026-01-30T22:26:53Z

@MDAnalysis/coredevs we decided to merge an initial version of the AI policy by Tuesday 2026-02-03 (next week) — see today's business meeting notes for details.

The initial version is a baseline and will evolve in the future. Please review and comment and if possible approve (with the understanding that we will come back and update, based on discussion on this PR and elsewhere).

orbeckst · 2026-01-30T22:27:38Z

(@IAlibay volunteered to be in charge of making sure that the PR gets merged. 🙏 )

orbeckst

I would like to see a couple of changes incorporated (version + date, statement that the document will change, contact details) before approving.

Other suggestions should be uncontroversial (spelling) or are really just suggestions.

AI_POLICY.md

Co-authored-by: Oliver Beckstein <orbeckst@gmail.com>

IAlibay · 2026-02-01T02:56:20Z

I think all your comments have been addressed @orbeckst

orbeckst

Thank you, @IAlibay — v1.0 lgtm!

RMeli

LGTM. A few minor comments, but it can be merged IMO. Thanks for all the work!

AI_POLICY.md

RMeli · 2026-02-02T08:14:56Z

AI_POLICY.md

+
+### 5. Human reviewers are required
+
+All code merged into MDAnalysis repositories must be reviewed by a human reviewer. Instructions / suggestions from human reviewers always take precedence over those of non-human reviewers.


Should we suggest/ask of not adding Copilot as a reviewer, or that is something we would be OK having?

I don't have a strong view on this, personally folks can do what they want, I just don't want to be second guessed by AI.

AI_POLICY.md

Co-authored-by: Rocco Meli <r.meli@bluemail.ch>

Add AI tools policy.

1d69a74

Adds a markdown document that defines MDAnalysis' current stance on AI tools for contributions.

IAlibay requested a review from a team January 18, 2026 21:30

RMeli reviewed Jan 18, 2026

View reviewed changes

AI_POLICY.md Outdated Show resolved Hide resolved

AI_POLICY.md Outdated Show resolved Hide resolved

AI_POLICY.md Outdated Show resolved Hide resolved

Apply suggestions from code review

d1cf662

Co-authored-by: Rocco Meli <r.meli@bluemail.ch>

orbeckst requested changes Jan 19, 2026

View reviewed changes

tylerjereddy reviewed Jan 19, 2026

View reviewed changes

AI_POLICY.md Outdated Show resolved Hide resolved

AI_POLICY.md Outdated Show resolved Hide resolved

RMeli reviewed Jan 19, 2026

View reviewed changes

AI_POLICY.md Outdated Show resolved Hide resolved

Typos found by @tylerjereddy

aa74113

Correcting some typos found during review.

orbeckst reviewed Jan 29, 2026

View reviewed changes

AI_POLICY.md Outdated Show resolved Hide resolved

orbeckst assigned IAlibay Jan 30, 2026

orbeckst requested changes Jan 30, 2026

View reviewed changes

IAlibay and others added 10 commits February 1, 2026 01:35

Update AI_POLICY.md

9720d25

Co-authored-by: Oliver Beckstein <orbeckst@gmail.com>

Try to address orbeckst's review

b84a3de

Update AI_POLICY.md

f425322

Co-authored-by: Oliver Beckstein <orbeckst@gmail.com>

Some slight changes

900ba30

some review changes

7e95791

Update AI_POLICY.md

709c1d3

Co-authored-by: Oliver Beckstein <orbeckst@gmail.com>

Update AI_POLICY.md

d143c4c

Co-authored-by: Oliver Beckstein <orbeckst@gmail.com>

Update AI_POLICY.md

634ab6c

Co-authored-by: Oliver Beckstein <orbeckst@gmail.com>

Update AI_POLICY.md

3ed7106

Co-authored-by: Oliver Beckstein <orbeckst@gmail.com>

A clarify org vs single repo

d4b30c1

IAlibay requested a review from orbeckst February 1, 2026 02:56

orbeckst approved these changes Feb 2, 2026

View reviewed changes

RMeli approved these changes Feb 2, 2026

View reviewed changes

IAlibay and others added 2 commits February 2, 2026 09:27

Update AI_POLICY.md

af8c3cf

Co-authored-by: Rocco Meli <r.meli@bluemail.ch>

Update AI_POLICY.md

1615ac7

Co-authored-by: Rocco Meli <r.meli@bluemail.ch>

IAlibay merged commit 6b8d417 into develop Feb 2, 2026
14 of 21 checks passed

	As per section #0, where possible please state that you are using AI assistance via an IDE.
	As per section #0, please state when you are using AI assistance via an IDE.


		### 5. Human reviewers are required

		All code merged into MDAnalysis repositories must be reviewed by a human reviewer. Instructions / suggestions from human reviewers always take precedence over those of non-human reviewers.

[ADMIN] Adding an MDAnalysis AI tools policy. #5210

[ADMIN] Adding an MDAnalysis AI tools policy. #5210

Conversation

IAlibay commented Jan 18, 2026 • edited by github-actions bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

LLM / AI generated code disclosure

PR Checklist

Developers Certificate of Origin

Uh oh!

codecov bot commented Jan 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

RMeli left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

orbeckst left a comment

Choose a reason for hiding this comment

Uh oh!

orbeckst Jan 19, 2026

Choose a reason for hiding this comment

Uh oh!

IAlibay Jan 19, 2026

Choose a reason for hiding this comment

Uh oh!

orbeckst Jan 30, 2026

Choose a reason for hiding this comment

Uh oh!

IAlibay Feb 1, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

tylerjereddy left a comment

Choose a reason for hiding this comment

AI usage disclosure

Uh oh!

Uh oh!

Uh oh!

Uh oh!

BradyAJohnston commented Jan 19, 2026

Uh oh!

IAlibay commented Jan 19, 2026

Uh oh!

orbeckst commented Jan 28, 2026

Uh oh!

IAlibay commented Jan 29, 2026

Uh oh!

Uh oh!

orbeckst commented Jan 29, 2026

Uh oh!

orbeckst commented Jan 30, 2026

Uh oh!

orbeckst commented Jan 30, 2026

Uh oh!

orbeckst left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

IAlibay commented Feb 1, 2026

Uh oh!

orbeckst left a comment

Choose a reason for hiding this comment

Uh oh!

RMeli left a comment

Choose a reason for hiding this comment

Uh oh!

IAlibay commented Jan 18, 2026 •

edited by github-actions bot

Loading

codecov bot commented Jan 18, 2026 •

edited

Loading