Godot Bans AI Contributions: Broken Trust

What happened?

On July 1, 2026, the Godot team, the most popular open source game engine after Unity and Unreal, announced a rewrite of its contribution policy to ban almost all use of artificial intelligence. According to The Register, the measure responds to an overwhelming number of pull requests (PRs) that maintainers describe as 'demoralizing'. In their announcement, maintainers stated: 'AI cannot assume responsibility, and we cannot trust heavy AI users to understand their code enough to fix it.' The decision was not sudden: since early 2025, Rémi Verschelde, a Godot maintainer, had already pointed out that AI-generated PRs were an 'increasingly frustrating waste of time'. A game studio using Godot described these PRs as 'garbage' and 'total chaos'. The accumulation of these low-quality contributions led the team to conclude that the problem would not go away on its own, prompting the drastic action.

Context and background

The phenomenon of 'vibe coding' —letting AI write code without understanding it— has been on the rise since the popularization of tools like GitHub Copilot, ChatGPT, and Claude. In 2025, a GitClear report revealed that AI-generated code introduced more bugs and was harder to maintain than human-written code. Open source projects like Mozilla and TensorFlow had already reported similar issues, but none had taken such restrictive measures as Godot. Godot's case is particularly illustrative because its community is relatively small (about 1,200 active contributors), making PR review a bottleneck. Maintainers, mostly volunteers, were overwhelmed by PRs that seemed plausible at first glance but, upon examination, contained logical errors, security vulnerabilities, or simply did not follow project conventions. Infosys chairman Nandan Nilekani warned at a recent conference that 'there is much more to the software development lifecycle than just coding,' referring to debugging, testing, and long-term maintenance, tasks that AI cannot yet perform reliably.

The new policy in detail

The change includes several drastic measures, detailed in the official announcement from the Godot team:

• Explicit permission for newcomers: New contributors (with three or fewer merged PRs) must obtain prior authorization from maintainers to propose new features or major refactorings. This aims to exclude 'vibe coders' and AI agents, fostering a group of contributors who understand the codebase and are willing to communicate with the team.
• Mandatory human communication: AI agents or bots are banned from discussion channels, except for language translation. Maintainers explained: 'We need to ensure that people who choose to review PRs feel their time is well spent.'
• Ban on substantial AI-generated code: Only minimal assistance (autocomplete, regex, search and replace) is allowed. Any use of AI must be declared in the PR thread. The policy states that 'AI-generated code that constitutes a substantial part of a contribution will not be accepted.'
• Auto-ban for violations: PRs generated by autonomous agents or 'vibe coding' will result in automatic expulsion from the GitHub repository. This measure was already applied informally but is now formalized.

Why is this important?

Godot is not an isolated case. The trend toward 'vibe coding' is causing quality, security, and maintainability issues across the open source ecosystem. A 2025 Stanford University study found that AI-generated PRs had a 30% lower acceptance rate than human ones but consumed twice the review time. Godot's decision sets a precedent for other projects struggling with contamination from AI-generated PRs. Projects like Linux, Kubernetes, or the Android kernel could adopt similar measures, especially since their maintainers have also expressed frustration. Moreover, Godot's policy reflects a paradigm shift: from uncritical acceptance of AI to a more critical stance that prioritizes quality and accountability. As the team noted, 'AI cannot assume responsibility,' a key point in free software where every contribution must be maintained by the community.

Immediate and long-term consequences

In the short term, regular Godot contributors will see relief in the review burden. Maintainers will be able to dedicate more time to real improvements and less to filtering garbage. However, the policy could discourage new contributors who rely on AI to learn or prototype. In the long term, the measure could slow AI-driven innovation in the Godot ecosystem, but it will also ensure that code is maintainable and accountable. For the market, Godot's decision could influence other open source projects and how companies adopt AI in development. For example, companies like Google and Microsoft, which promote AI tools, could face pressure to improve the quality of generated contributions. In the gaming field, Godot is used by indie and educational studios; a cleaner codebase will benefit the long-term stability of games developed with the engine.

What readers should know

If you contribute to Godot, make sure you understand every line of code you propose. The policy requires declaring any use of AI, and violations can lead to auto-ban. For developers in general, the lesson is that AI is a tool, not a substitute for understanding. Maintainers' trust is the most valuable resource in free software. If you are a 'vibe coder,' look for projects that accept that style, or better yet, learn to code. As a Godot maintainer said: 'Code doesn't maintain itself; someone has to understand it and fix it when something goes wrong.' Godot's decision will likely mark a before and after in the relationship between AI and open source development.