Godot’s Ban in Plain Terms: AI Can’t Be a Maintainer
The Godot Engine’s new open-source AI policy is a set of contribution rules that prohibit AI-generated code submissions, autonomous AI agents creating pull requests, and AI-generated text in human-to-human communication, in order to protect reviewer capacity, code quality standards, and the human mentorship that keeps volunteer-driven projects sustainable. The Godot Foundation has updated its contribution guidelines to require that all substantial code contributions be human authored and disclosed if AI tools were involved in the process. This is not a minor tweak; it is an explicit statement that AI-generated code contributions, as they commonly appear today, are incompatible with how the project wants to be maintained. By drawing a hard line now, Godot is rejecting the idea that more pull requests are automatically good and arguing instead that the only PRs that matter are those humans can stand behind over time.
The Backlog, the “AI Slop,” and Why Maintainers Snapped
Godot’s ban did not come out of nowhere; it followed months of maintainers describing AI-generated contributions as “increasingly draining and demoralizing.” The number of qualified reviewers is small, reviewing pull requests is demanding, and the maintainers admit they “can’t keep up with everything coming in.” At the same time, AI has lowered the effort needed to file a PR, inflating the total volume while the review capacity stayed flat. That flood included “AI slop code” and “low-effort slop” from vibe coding and autonomous agents, often nonsensical or misaligned with the engine’s architecture. When the queue becomes a meme, review stops feeling like collaboration and starts feeling like unpaid triage. According to one statement, “AI contributions have the added pain of being demoralizing” because feedback no longer builds a future maintainer; it disappears into a machine.
Accountability, Not Algorithms, Is the Real Technical Problem
The Foundation’s stance is not anti-tool; it is anti-unowned code. Their rules explicitly allow AI assistance for “menial things” such as code completion, regex, or find-and-replace, but ban AI-generated substantial pieces of code. The rationale is blunt: quality variance, accountability gaps, and the reality that heavy users of AI often do not understand the code well enough to fix it when it breaks. AI cannot take responsibility for security, performance, or long-term maintenance; that burden falls entirely on human reviewers who are already scarce. LLMs also “can’t learn from specific feedback and thus can’t benefit from maintainers providing feedback,” which destroys the core feedback loop that sustains open-source communities. In a project that must remain coherent for years, unaccountable automation is not free productivity—it is technical debt outsourced to volunteers.
Respecting Human Review: Communication and New Contributor Limits
Godot did something many projects shy away from: it tied technical rules to social expectations. The updated policy bans AI-generated text in human-to-human communication, framing it as a “basic principle of respect” that volunteers who review issues, PRs, or proposals “do not want to talk to a machine.” That may sound harsh, but it acknowledges that open-source work is powered by relationships, not tickets. To relieve pressure on reviewers, the Foundation now treats anyone with three or fewer merged pull requests as a new contributor and blocks them from submitting new features or major refactors without explicit maintainer permission. The goal is to route newcomers toward bug fixes and documentation first, building trust incrementally. PR review has become “the largest bottleneck” for the engine, and these barriers are designed to filter out low-effort AI slop and refocus limited attention on humans who want to learn and stick around.
What Other Open-Source Projects Should Learn From Godot’s Crackdown
Godot is clear that it will “re-evaluate as things evolve,” taking a conservative approach to AI tools but not treating the current ban as permanent doctrine. Still, its move surfaces a tension that every open-source project now faces: automation may increase throughput, but it can erode code quality standards and the social contract that keeps reviewers engaged. Human review is already mandatory in other large projects, such as the Linux kernel, and Godot’s clampdown suggests more communities will tighten their open-source AI policy as AI-generated contributions spread across GitHub. Godot is not the first to grapple with this and “most definitely won’t be the last.” The lesson is not that AI has no place in open source; it is that AI must serve human maintainers, not drown them. Projects that fail to put people first will discover that endless PRs are worthless if no one is willing to review them.
