Godot’s AI Ban Is About People, Not Just Code
An open source AI ban in a project like the Godot game engine is a contribution policy that restricts or forbids substantial AI-generated code, automated pull requests, and AI-written communication in order to protect code quality, maintain human accountability, and keep community review work sustainable for volunteer maintainers.
The Godot Foundation is rewriting its contribution policy to prohibit AI-authored code, autonomous agent pull requests, and AI-generated text in maintainer conversations. On paper, that sounds like a conservative technical choice. In practice, it is an aggressive defense of community culture. Godot’s maintainers are overwhelmed by a backlog of pull requests that has “become a meme in the community,” much of it padded by low-effort AI slop. The move says, bluntly, that open source is not a free playground for bots. It is a social contract between humans who donate time, attention, and care—and that contract breaks when contributors outsource both thinking and conversation to machines.
The New Godot Contribution Policy: AI Out, Accountability In
Godot’s updated contribution policy draws a bright line: no autonomous AI agents, no “vibe coding,” and no substantial AI-generated code. AI use is limited to “menial things” such as code completion, regex, or find-and-replace, and any AI involvement must be disclosed directly in the pull request discussion. AI-written text is also banned from human-to-human communication, with the Foundation calling it a “basic principle of respect” that maintainers talk to a person, not a model.
This is not a symbolic gesture. Autonomous AI agent use or vibe-coded pull requests already trigger an automatic ban from Godot’s GitHub repository, and the new rules formalize that stance. All pull requests must still be reviewed and approved by a human before merging, a requirement now made explicit. In other words, Godot is rejecting the fantasy of a semi-automated codebase and doubling down on a simple expectation: if you contribute, you are personally responsible for understanding and defending every line you submit.
Why AI Slop Demoralizes Maintainers
Maintainer burnout in open source is not new, but AI has poured fuel on it. Godot’s contribution pipeline has grown sharply, yet the pool of qualified reviewers remains small. Reviewing pull requests is demanding work, and maintainers say they can no longer keep pace with a backlog swollen by AI-authored submissions. Worse, much of that content is, in their words, “AI slop code” and “low-effort slop” that threatens the coherence of the engine.
The quality problems are familiar: wild variance, accountability gaps, and contributors who rely so heavily on AI that they do not understand the code well enough to fix it when it breaks. But the deeper wound is psychological. “AI contributions have the added pain of being demoralizing,” the Foundation notes, because feedback on a pull request does not change a model’s future behavior, and heavy AI users cannot be trusted to absorb that feedback on the model’s behalf. Maintaining an open source project already means giving away nights and weekends; doing that for a machine that will never learn or care feels like a bad trade.
Protecting the Mentorship Pipeline, Not Gatekeeping
Godot’s maintainers are explicit: pull request review has always been tedious, but it was worth the effort because it doubled as mentorship. Reviewers were not only improving code; they were training future maintainers and reviewers. That equation collapses when the “contributor” is a language model. Feedback goes into a void, and there is no growing human on the other side of the comment thread.
To restore that pipeline, the Foundation is tightening expectations on humans. New contributors—anyone with three or fewer merged pull requests—are now barred from submitting new features or significant refactors without explicit maintainer permission. They are expected to build trust on bug fixes and documentation first. At the same time, every contributor must disclose AI use and stay within narrow assistance boundaries. This is not gatekeeping for its own sake; it is triage. “We need to take steps to reduce the burden on maintainers while ensuring we still have a pipeline to mentor new contributors to become future maintainers,” the Foundation writes.
A Broader Open Source Reckoning With AI Hype
Godot is not the first project to confront this problem, and it will not be the last. Other open source communities, such as the Zig programming language, have adopted zero-tolerance policies for AI-assisted contributions on similar mentorship grounds, arguing that review time should invest in the person submitting the code. Terminal projects like Ghostty and long-running libraries like curl have also restricted or closed parts of their contribution pipelines over AI-fueled review burden and false bug reports. Together, these moves amount to an emerging open source AI ban pattern: where AI increases volume without improving relationships, volunteers push back.
The underlying tension is simple. AI promises acceleration; open source survives on sustainable, human-centered maintenance. Projects run on volunteer labour and a social contract in which people give time because they feel part of something and because their contributions are seen by other humans. AI agents do not participate in that contract; they consume its outputs without strengthening the community that generates them. Godot’s Foundation calls its approach “conservative” and says it will re-evaluate its policy as tools evolve. For now, the message to developers is clear: if you want a say in open source engines, show up as a person, not a prompt.
