Skip to content

AI Contributions Policy

Blender is software made by humans for humans. The success and growth of the project depends on a developer community that both maintainers and new contributors want to participate in, where the work is rewarding, real connections are made between people, and a collaborative effort leads to better results. That culture is one of Blender’s greatest strengths.

In recent years, AI tools have made it easier to generate and contribute code, documentation, and designs. However, quality is often poor, and reviewers and maintainers already have limited time to handle every contribution. It is important to focus resources on newcomers who can learn from reviews and grow into long-term contributors. Reviewers’ time should also be protected from submissions that ultimately waste it. Additionally, some maintainers do not wish to review or engage with AI assisted contributions, and that choice should be respected.

This policy is designed to safeguard the core principles of Blender Development’s culture. In addition to the general contribution guidelines, AI assisted contributions must also follow the rules and (unless otherwise agreed with maintainers) the guidelines defined below. This policy applies to code, documentation and design contributions for Blender development, as well as bug reports and comments. It does not apply to Blender extensions.

If maintainers either do not wish to review AI assisted contributions or strongly suspect a contribution does not follow this policy, they reserve the right to close the pull request or issue without review with a link to this policy. For significant contributions, it is therefore highly recommended to check with maintainers of the relevant module(s) beforehand.

Rules

  • You must take responsibility for your contribution. Contributing to Blender means vouching for the quality, license compliance, and utility of your submission. All contributions, whether human-written or tool generated, must meet the project’s standards for inclusion.
  • For Git commits, only humans can be commit authors or listed as Co-authored-by. You are always the author and fully accountable for the entirety of your contributions.
  • You must understand and manually review code, designs, bug reports, descriptions and comments you submit. For code contributions, you must carefully test the resulting functionalities and fixes. Never pass along tool generated output without fully understanding it.

Guidelines

  • Primarily think of tools as a way to improve the quality of your contributions and the Blender project, rather than a way to work faster or go beyond your expertise. If a tool does part of the work for you, that means you have more time for careful testing, and thinking through design and corner cases. Additionally, these tools can be prompted to perform code reviews, check for consistency with guidelines, run detailed tests, etc. For this reason, the quality bar for AI generated contributions is higher than other contributions.
  • Your own effort should provide significant value beyond the output of tools. If you are merely acting as an intermediary between the tool and reviewer, this is not a good use of the reviewer's time. They could have been using the tool directly or could have chosen different priorities.
  • Write your own descriptions and comments in pull requests, issues, forum posts and chat. This helps you understand the code and think through the decisions, assumptions and consequences. If you need to additionally pass along tool generated text verbatim, clearly indicate it. Edits for spelling, grammar or clarity do not need to be indicated. If you translate from another language, it is recommended to include both the English and original text.
  • For large scale changes or code you are unfamiliar with, describe the methodology used to implement tool-assisted pull requests. Whether it's AI or a more traditional tool or script, explaining the tools and process is educational, and is important for the reviewer to understand what to pay attention to and what may have been missed.
  • Do not solve Good First Issues with AI. These are meant as a learning experience and require time investment from reviewers, which is wasted on AI.
  • If generated code is not clearly an extension to or refactoring of existing code, expect extra scrutiny and potential rejection due to copyright risk.