Artificial Intelligence Usage Policy

(Applicable to Google Summer of Code Contributors and Applicants)

1. Purpose

This policy defines the acceptable use of Artificial Intelligence (AI) tools within our organization during all phases of participation in Google Summer of Code (GSoC), including pre-application contributions, proposal writing, development, and reporting.

Our goal is to allow AI as a productivity and learning accelerator, while ensuring that all accepted contributors demonstrate genuine understanding, authorship, and responsibility for their work.

2. Scope

This policy applies to:

GSoC proposals
Pre-proposal contributions (issues, discussions, pull requests)
Accepted project work (code, documentation, design)
Weekly reports and final evaluations
Any communication or artifact submitted to the organization

3. Permitted Use of AI Tools

Contributors may use AI tools for:

Brainstorming and refining ideas
Improving clarity, grammar, or structure of text
Learning unfamiliar concepts or APIs
Debugging assistance
Code explanation or refactoring suggestions

AI is considered an assistant, not an author.

4. Mandatory Disclosure

AI usage must be explicitly disclosed in all relevant contexts, including but not limited to:

GSoC proposals
Pull requests and commit messages
Reports and documentation

At minimum, contributors must state that AI was used. Whenever possible, contributors should also explain how the AI was used (e.g., for debugging, rewriting, brainstorming).

Failure to disclose AI usage is considered a policy violation.

5. Authorship & Understanding Requirements

All submitted work must:

Be fully understood by the contributor
Be explainable on request by mentors or reviewers
Reflect active human decision-making

The following are not acceptable:

Fully AI-generated proposals
Code written by AI that the contributor cannot explain
Submissions copied from AI outputs without review or comprehension

AI must not be used as a replacement for thinking, design decisions, or learning.

6. Contribution Quality Standards

When contributing code (including pre-proposal PRs):

Large, untested, or unexplained submissions (e.g., thousands of lines of code) will not be accepted
Contributors are expected to submit small, atomic, reviewable changes whenever possible
Code must be tested, justified, and aligned with project goals

AI-assisted code does not exempt contributors from these standards.

7. Review, Clarification, and Enforcement

If a mentor or reviewer believes that:

AI has been used improperly, or
The contributor does not understand their own submission,

A clarification round will be initiated, giving the contributor an opportunity to:

Explain their work
Revise the submission
Correct disclosure issues

Repeated misuse, inability to demonstrate understanding, or intentional deception may result in:

Proposal rejection
Pull request rejection
Removal from consideration or the program, at mentor discretion

8. Mentor Oversight

Mentors are expected to:

Actively assess contributor understanding
Ask clarification questions when AI usage is suspected
Ensure AI is being used as a productivity aid, not a dependency

Mentor judgment is final in matters related to AI usage compliance.

9. Final Note

Using AI tools responsibly is encouraged. Using AI to avoid learning, understanding, or accountability is not.

you can check here our policy