melissawm/open-source-ai-contribution-policies

What it solves

Open source maintainers and contributors face uncertainty about how AI-generated code and content should be handled. This project collects and organizes the actual policies that different open source projects have adopted, so people can see the range of approaches and find precedents.

How it works

It is a simple table in the README listing each project, a link to its policy, and several attributes: whether AI/LLMs are allowed, whether disclosure is required, whether there's a copyright statement, whether a human must be in the loop, and extra notes. The classification is admitted to be imperfect and only a general guide. The project invites pull requests to add more policies from other projects.

Who it's for

Open source maintainers deciding on their own AI-contribution policies, contributors who want to know what to expect before submitting AI-assisted work, and anyone studying how the open source community is responding to generative AI.

Highlights

• A growing list of policies from dozens of projects (e.g., Adwaita, Apache Airflow, CPython, curl, Django, Firefox, Godot, and many more).
• Each entry captures key details: allowance, disclosure requirement, copyright language, and human-in-the-loop requirement.
• Links to the actual policy documents for each project.
• Includes a note that the classification is not definitive and serves only as a general guide.
• Open to contributions: users can submit pull requests with policies from other projects.
• Provides similar project links for broader context (e.g., CHAOSS AI Alignment, RedMonk analysis).