[Biopython] Generative AI policy for contributions to Biopython

[Iddo Friedberg]

I like the NumPy/SciPy policy.

Their outline resonates well with my intuition of how to (1) avoid slop /
malicious code; (2) avoid agentic submission (3) mitigate copyright
violations (4) using disclosures to understand the AI component of any PR.

I suggest to also add the following requirement: submissions should include
meaningful inline documentation / docstrings, that is NOT AI generated.
This will require the submitter to prove code understanding AND ensure that
communication is not via AI. This is on top of a meaningful (read: "long")
documentation and reasoning as part of the PR.

My humble $0.02

./I

On Wed, Oct 21, 2026 at 12:51 PM Roo Sczesnak <andrewscz at gmail.com> wrote:

> +1 to adopting a more moderate AI contribution policy, following
> NumPy/SciPy's example. IMHO, a blanket ban on all AI tool usage is
> heavy-handed. It is entirely possible to write great code with AI tools. A
> blanket ban won't stop anyone from using the tools, it just forces them to
> obscure their usage.
>
> On Wed, Oct 21, 2026 at 4:42 AM Peter Cock <p.j.a.cock at googlemail.com>
> wrote:
>
>> On Sun, Oct 11, 2026 at 11:48 AM Peter Cock <p.j.a.cock at googlemail.com>
>> wrote:
>> >
>> > ... I understand that Michiel favours a more
>> > permissive policy like NumPy or SciPy adopted, based on technical
>> > merits and still expecting human understanding of the code. ...
>> >
>> > Would any of you like to put forward an alternative AI policy draft
>> > pull request based on NumPy and SciPy please (I find the Python
>> > project text too light on detail), including our consensus that the
>> > communications at least should be human written (but possibly computer
>> > translated into English). I think we need something concretely enough
>> > that we can review PRs consistently. One of the advantages of a no-AI
>> > policy is the line for what is or isn't acceptable is much easier to
>> > follow.
>>
>> As I noted on my no-AI PR yesterday, BioPerl is looking at adopting
>> the CoMaps/Forejo AI policy:
>>
>> https://github.com/bioperl/bioperl-live/issues/407#issuecomment-6275498900
>> https://codeberg.org/comaps/Governance/src/branch/main/AI_USAGE.md
>>
>> It has a lot in common with the NumPy/SciPy policy (both based on
>> SymPy), both stress human communications and no autonomous agents
>> (which if followed would limit some of the worst aspects of slop PRs),
>> but these are more explicit about copyright (but not explicitly about
>> attribution):
>>
>> https://numpy.org/doc/stable/dev/ai_policy.html
>> https://docs.scipy.org/doc/scipy/dev/conduct/ai_policy.html
>>
>> NumPy/SciPy have a minimal PR template which includes an "AI
>> Generation Disclosure" point. CoMap's PR template is closer to ours
>> with checkboxes:
>>
>>
>> https://codeberg.org/comaps/comaps/src/branch/main/.forgejo/PULL_REQUEST_TEMPLATE.yml
>>
>> If Biopython is going to adopt a moderate stance on AI contributions
>> (which I remain opposed to in principle), these are a reasonable
>> starting point. I can see arguments in following either (consistency
>> with sister project BioPerl, or Biopython's main dependency library,
>> NumPy), or drawing on both.
>>
>> Again, would anyone like to put together an alternative AI policy PR
>> along these lines?
>>
>> Peter
>> _______________________________________________
>> Biopython mailing list  -  Biopython at biopython.org
>> https://mailman.open-bio.org/mailman/listinfo/biopython
>>
> _______________________________________________
> Biopython mailing list  -  Biopython at biopython.org
> https://mailman.open-bio.org/mailman/listinfo/biopython
>


-- 
Iddo Friedberg

++++++++++[>+++>++++++>++++++++>++++++++++>+++++++++++<<<<<-]>>>>++++.>
++++++..----.<<<<++++++++++++++++++++++++++++.-----------..>>>+.-----.
.>-.<<<<--.>>>++.>+++.<+++.----.-.<++++++++++++++++++.>+.>.<++.<<<+.>>
>>----.<--.>++++++.<<<<------------------------------------.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.open-bio.org/pipermail/biopython/attachments/20261021/8c70e151/attachment-0001.htm>