Godot is drawing a hard line: the open source game engine will no longer accept AI-authored code contributions, after the Godot Foundation concluded the flood of AI “slop” pull requests is becoming unsustainable for maintainers and code reviewers.

In a blog post, the Godot Foundation said it is amending contribution guidelines to reject AI-authored code, pull requests submitted by AI agents, and AI-generated text in human-to-human communication. The foundation is explicit about the reason: “AI cannot take responsibility, and we can’t trust heavy users of AI to understand their code enough to fix it.” If you run, fund, or depend on open source infrastructure, this is the moment the industry’s tolerance for unaccountable automation gets replaced by policy with teeth.

This didn’t come out of nowhere. Back in February, Godot maintainers said they were deliberating how to address a rising tide of AI slop pull requests. Those requests had grown “increasingly draining and demoralizing” for the project’s code reviewers. In other words, the work wasn’t just extra. It was corrosive, because the reviewer effort was being spent sorting through low-effort submissions rather than moving the project forward.

The Foundation also frames the problem as more than pure workload. It argues the pile of pending pull requests is not all bad because it signals growing interest in using and contributing to Godot. But interest has a cost. The foundation said when review feedback is “just being absorbed by a machine and not going towards mentoring a potential future maintainer, it becomes much harder to justify spending your free time on PR review.” That is the core governance tension: open source needs a pipeline for new maintainers, not just a constant stream of code artifacts that do not cultivate responsibility.

So Godot’s response is structured like a policy designed for enforcement, not vibes. The Foundation says it is updating its contribution policies with an emphasis on adding “barriers to low-effort slop” contributions, encouraging maintainers to review code, and developing new contributors into future maintainers. Crucially, it says all contributions must come from humans who are accountable for their code and for fixing it if it fails. That human accountability requirement is what separates “AI assistance” from “AI authorship,” and it is what the Foundation thinks makes maintenance realistic.

The new rules will also include explicit rejections of AI-authored code, along with expectations about how contributors can use AI tools. The Foundation says contributors should only use AI assistance for “menial things” and must disclose its use. That disclosure requirement matters because it turns AI from an invisible wildcard into a declared input. In practice, it gives reviewers a chance to interpret code quality and intent, and it sets a standard for what “acceptable assistance” looks like.

On the communication side, Godot will reject any AI-generated text in human-to-human communications, calling it “a basic principle of respect.” That may sound cultural, but it has an enforcement logic too: human-to-human communication is where maintainers assess understanding, intent, and follow-through. If messages are generated wholesale, it becomes harder to tell whether a contributor actually knows what they wrote and can debug when reality hits.

The Foundation did add one meaningful exception for context: machine translations are still acceptable if the original text was human-authored. The Foundation also cautioned that tools and capabilities keep changing, saying it will keep taking a conservative approach in its policies towards AI, and “we will continue taking a conservative approach in our policies towards them, but we will re-evaluate as things evolve.” That language signals the Foundation is trying to lock in principles while leaving room for future adjustment.

Second-order, this matters well beyond Godot. Open source projects live or die by review bandwidth, and review bandwidth is a scarce resource. When AI submissions inflate the number of pull requests while reducing the likelihood of successful follow-through, the ecosystem can end up with an unhealthy equilibrium: lots of activity, less progress. By requiring human accountability and creating rejection criteria for AI-authored contributions, Godot is essentially redesigning incentives. It is telling contributors: participation is welcome, but maintenance-grade contribution requires responsibility, not automation.

If you are a board member, investor, foundation operator, or platform leader supporting open infrastructure, the takeaway is that governance is getting stricter, not softer. Godot’s policy shift is an early signal that the open source funding and stewardship question may soon turn from “how do we attract contributors?” to “how do we ensure contributors can own outcomes?” In a world where AI can generate code quickly, Godot is betting that what differentiates healthy projects is not speed. It is accountability.