Generative AI policy

[browniebroke]

As the use of generative AI is becoming more common in the industry, more code is being produced and I think that poses a risk to the project sustainability as we (maintianers) are on the receiving end of this.

A number of open source project are starting to clarify their posture:

• Python: https://devguide.python.org/getting-started/generative-ai/
• Django (WIP): Extended the GitHub pull request template to require AI assistance disclosure. django/django#19594
• Octoprint: https://github.com/OctoPrint/OctoPrint/blob/master/CONTRIBUTING.md#on-ai-use-when-contributing-to-octoprint

I'm starting to see some pull requests coming through which are looking very much like AI output, sometimes accompanied with a couple of pages in the PR description. I'm not against using LLMs to help developers be more productive, but sending a barely reviewed PR to maintainers is a bit disrespectful of the volunteers time. Having a clear, upfront policy would make closing/rejecting these PR much easier.

I'm happy to put this together, we could start with a section in our contributing page, which is already linked in our PR template.