Godot Game Engine No Longer Accepts AI Code 70
The Godot Foundation will stop accepting AI-authored code, agent-submitted pull requests, and AI-generated text in contributor communications after maintainers were overwhelmed by low-effort submissions. "It is time for us to recognize that these problems aren't going away and therefore 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 Godot Foundation said in a blog post. Contributors may still use AI for limited "menial things" if they disclose it, but humans must understand, own, and be able to fix the code they submit. PC Gamer reports: The Foundation says the pileup of Godot pull requests pending review isn't all bad: It's a sign that interest in using and contribution to Godot is increasing. But the influx of contributions authored or submitted by AI is sapping the projects' maintainers of their willingness to confront the "already tedious" work of reviewing pull requests. "If your feedback on PRs 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," the Foundation said.
As the problem becomes increasingly unsustainable, the Godot Foundation says it's in the process of updating its contribution policies, focusing on "adding barriers to low-effort slop" contributions, encouraging maintainers to review code, developing new contributors into future maintainers, and crucially, requiring that all contributions come from humans who are accountable for their code -- and fixing it if it fails. "AI cannot take responsibility, and we can't trust heavy users of AI to understand their code enough to fix it," the Foundation said.
The Foundation says we can expect Godot's contributing policy to soon include explicit rejections of AI-authored code, noting that contributors should only use AI assistance for "menial things" and must disclose its use. Additionally, the Foundation will reject any AI-generated text in human-to-human communications, saying it's "a basic principle of respect" -- though it says machine translations "are still acceptable" if the original text was human-authored. "Things change every day with respect to the current suite of AI tools available," the Foundation said. "We will continue taking a conservative approach in our policies towards them, but we will re-evaluate as things evolve."
As the problem becomes increasingly unsustainable, the Godot Foundation says it's in the process of updating its contribution policies, focusing on "adding barriers to low-effort slop" contributions, encouraging maintainers to review code, developing new contributors into future maintainers, and crucially, requiring that all contributions come from humans who are accountable for their code -- and fixing it if it fails. "AI cannot take responsibility, and we can't trust heavy users of AI to understand their code enough to fix it," the Foundation said.
The Foundation says we can expect Godot's contributing policy to soon include explicit rejections of AI-authored code, noting that contributors should only use AI assistance for "menial things" and must disclose its use. Additionally, the Foundation will reject any AI-generated text in human-to-human communications, saying it's "a basic principle of respect" -- though it says machine translations "are still acceptable" if the original text was human-authored. "Things change every day with respect to the current suite of AI tools available," the Foundation said. "We will continue taking a conservative approach in our policies towards them, but we will re-evaluate as things evolve."
Oh my, are we waiting? (Score:3, Funny)
Ohhhh-hhhh-hh-ooo
Re:Oh my, are we waiting? (Score:4, Informative)
Yes: [wikipedia.org]
The name "Godot" was chosen in reference to Samuel Beckett's play Waiting for Godot, as it represents the never-ending wish of adding new features in the engine, which would get it closer to an exhaustive product, even though it never would.
Re: (Score:2)
To update their policies and procedures.
I have Zero Mostel's face stuck in my mind now, covering his cheeks and moaning. Ever see the play?
Re: (Score:2)
For at least a year (maybe more) I thought that it was go-dot, kind of a play on dotnet. Especially with the work on adding c# support.
I so dumb!
Re: (Score:1)
Re: (Score:2)
what is godot game engine and why should I care?
Re: (Score:2)
Google is RIGHT THERE dude. Theres no need to be so lazy.
The open source world is going to fork bomb (Score:1)
Re: (Score:2)
Re: (Score:1)
Who's paying for all of these tokens?
Re: (Score:1)
Re: (Score:2)
Em... "brief storm"? If the "AI companies" profit both from the generation of slop, and from the defense against slop? I think they own both sides.
Re: (Score:3)
Those godot forks dont go far. There was one over..... shit maybe it was the code of conduct or something like that. It had 4, maybe 5 conrtributors. The Godot engine it self has 34 separate teams (Ie "core", "network", "physics") each with their own leads that do pull request merging and eventual pushing shit up towards juan who fills the role of basically linus torvalds for Godot). Its a shockingly big operation these days, and yeah. There was a reasonably larger fork over technical issues but it wasnt a
Spot on... (Score:5, Interesting)
reject any AI-generated text in human-to-human communications, saying it's "a basic principle of respect"
I cannot agree more with this sentiment. It feels outright insulting to asked to read LLM output in a context where it is *supposed* to be human feedback. Tell me what you would have told the LLM to say, I can take it from there. I don't need you to LLM it up, because it will bury your point in a bunch of crap.
Could it provide useful info? Maybe, but I can do that myself if so. I want *your* thought on something, however incomplete it might be.
Re: Spot on... (Score:2)
You are quite correct in that copy and pasting LLM responses to humans is disrespectful.
In this particular instance though, the Godot maintainers have been disrespectful to humans attempting to contribute for several years.
In other words - yes, we should decry honest devs getting swamped with AI crap. No, the Godot devs had it coming and do not deserve our pity, despite being OSS.
Re: (Score:3)
Oh nonsense. Godots a huge project with literally thousands of contributors. Theres always a handful of people who are serial garbage submitters in any large project (Ask the kernel people about this, they have a terrible time dealing with this stuff) who throw giant wobblies when their buggy nonsense gets politely declined. Theres also those contributions that arent necessa
Re:Spot on... (Score:4, Insightful)
Software development is changing. I started coding in BASIC in the 1980's and have been coding now for over 40 years, over 30 years professionally. I'm good at what I do, but the AI is faster. Claude can churn out code faster than I can, and it's often better, catching some conditions I would have missed. That said, it often messes up, misses the mark, or goes in directions that aren't right for the larger context in which the code exists.
Today, professional software development is best done by AI with skilled human guidance and review.
Rejecting AI generated code in today's environment is trying to turn back time. On the other hand, rejecting a submission where there is no human who can "understand, own, and be able to fix the code they submit" makes perfect sense. There is a big difference between asking an AI to generate a fix and blindly submitting the first thing it spits out, versus having an extended session with an AI, correcting it where it goes wrong, vetting and testing the patch with human review and testing, then submitting the PR.
Re: (Score:3)
Oh, and here's a fun prompt I like to throw
Re:Spot on... (Score:4, Interesting)
AI is the asbestos we are building into the walls of modern software. It's going to be a royal mess to clean up eventually.
Re: (Score:2)
Today, professional software development is best done by AI with skilled human guidance and review.
[Citation needed]
Re: Spot on... (Score:2)
I suspect tbag is true, or will be true.
The problem I see is tbat viewed from the outside, excellent code look like the dog shit that comes out of "please make xyz work with abc".
And since you can't tell the difference, you kind of have to treat everything AI generated the same.
Professional developers use AI for boilerplate (Score:3)
Today, professional software development is best done by AI with skilled human guidance and review.
Those that have reviewed and debugged AI generated code are the ones who are banning it.
Competent and professional software developers only use AI for boilerplate code. Ie extremely well known algorithms. UI layout code, and other such things that are extremely professionally well documented and somewhat menial. They know to stay away from more complex or novel problems.
Re: (Score:3)
Don't get me wrong, my gut says the same thing you did - don't make me read your AI summary, just tell me direct (bet you didn't know my gut could read). But I work with an Indian guy who uses it in emails to clean up his imperfect English. And I've seen people frustrated because they didn't know how to word something, and an AI cleared it up.
And if we hav
Re: (Score:1)
get good, or get out
Re: (Score:2)
I recently made an Android app with Gemini AI. I'm an embedded C guy, not done any Android development, and I just wanted a simple app for personal use so I thought I'd try it.
It made the app, eventually. I had to walk it through some parts, and repeat requests when it said it was done but hadn't actually done what I asked. Some research was needed to guide it, such as asking it to use a specific API call that properly aggregated data over a day, rather than trying to read all the raw readings and add them
Re: (Score:2)
Re: (Score:2)
Fitbit became Google Health and lost badges. Badges are little awards you get for passing certain milestones, like daily steps, or lifetime distance walked. This app brings them back.
Re: (Score:2)
Re: (Score:1)
Well, what about someone who isn't very good at putting their ideas into reasonably structured writing?
Then they probably also have problems putting their ideas into reasonably structured code.
Using AI generated summaries may be permissible if one is translating to another language they are not an expert with.
I've used AI to spell and grammar check summaries. Never to rewrite for style or tone.
Re: Spot on... (Score:2)
AI generated summaries maybe be an improvement on slashdot
Re: (Score:1)
AI generated summaries maybe be an improvement on slashdot
You sure its not already done, using the "Commy" AI writing agent ...
"Hey Commy!, Please summarize this historical event stylistically, and consistent with, 1970s Soviet propaganda and revisionist history. Please substitute millionaire with billionaire since too many useful idiots are now millionaires."
Re: (Score:2)
If people find it helpful, let them be helped.
And do note that when you let it change your grammar, it changes style and tone. Consider the preposition with which you ended, incorrectly, your penultimate sentence.
Re: (Score:2)
Permissible? I'm sorry Your Majesty, I did now know with whom I spoke.
If people find it helpful, let them be helped.
People are complaining because they are finding AI slop not so helpful.
And do note that when you let it change your grammar, it changes style and tone.
Fixing a grammar error rarely does so.
Re: (Score:2)
" It feels outright insulting to asked to read LLM output in a context where it is *supposed* to be human feedback."
More insulting or less insulting than having your human feedback rejected because someone claims it is LLM output?
What's this criterion does is provide non-falsifiable cover for rejecting anything.
Re: (Score:2)
What's this criterion does is provide non-falsifiable cover for rejecting anything.
Do they need cover to reject anything? In my projects, I reserve the right to reject anything, for any reason, solely on the grounds that they are my projects, and if someone doesn't like it, they can fork off (their own repository).
The "already tedious" work of reviewing pull reque (Score:1)
Godot is ...not a good game engine. It is in a very tiny niche of devs who attempt to avoid costs at all ... costs and still get stuff done. Almost no games made with Godot have ever amounted to anything. I am sorry, but it is true. We seriously considered Godot after the Unity runtime debacle a few years back, but we were put off by the Godot attitude, which seems to be... for lack of a better expression... "fuck you". The quip about the "already tedious" work of reviewing pull requests fits perfectly. The
Re: (Score:2)
Any chance you want to share what you were working on? You've piqued my curiosity. Or if you worked on any titles I may have played (as if you'd know what I've played...).
Re: (Score:3)
Re: (Score:2)
That was my point. We *did* try to contribute. We thought our contributions were reasonable. We were rebuffed. So, yeah, sure, two sides to a story.
Re: (Score:2)
Your statement has no merit, because it misses the point.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
What constitutes 'amounted to anything' here?
Seems suspiciously like a goalpost you're going to move if someone actually looks and provides any counterexamples (e.g. Slay the Spire 2, Cassette Beasts, Brotato, Cruelty Squad).
Rejections (Score:2)
Re: (Score:3)
And if you look at it like that, Godot is basically saying "we don't want to change our processes in the face of technology that has already forced them to change".
While I appreciate stubbornness, sometimes
Re: (Score:2)
"stubborness" is Godot in a nutshell. They might prevail with this in the future at some point. I do hope they will. We do need more options. If they do, we'll see changes. Unless this happens - stay clear.
AI merge is too dangerous (Score:2)
Or to flush that idea out a bit more, an AI to sift through them, organize, categorize, summarize, and merge or close when appropriate.
Not merge, that is too dangerous. The author of slop should not have merge permissions.
The problem with the AI generated code is that the human involved is likely not a competent reviewer. Such low quality submitters are trying to offload their self review and cleanup of code to others. That is negligent and likely wasting project resources.
Re: (Score:2)
Re: (Score:3)
Low effort to "write" code generates lots of slop (Score:2)
Who could ever have predicted that. The thing is, this does not even need to be low-quality code (although most LLM code is), the low quality prompting will already kill its usefulness totally and make it negative-worth "contributions".
In a more general sense, whenever some tech becomes something the masses can make, contributions to real projects need to establish gate-keeping, reputation and qualification standards, because most people have no understanding of quality or even the basics of real engineerin
You kind of have to (Score:4, Interesting)
Re: (Score:2)
Thats actually a really good point. We are far from untangling the legal mess this stuff brings. Fully AI genned code does not have copyright proection at all (My boss was a bit shocked when I pointed this out to him after he vibecoded an internal tool. I also found about 10 different serious vunerabilities in that mess after about 10 minutes of inspection lol. Hey man, this AI shit is existential threat territory for me......) and partial AI genned stuff is ...... still somewhat undefined copyright legalit
Re: (Score:2)
That is nonsense.
The copyright is with the author. Regardless what tool he uses.
The "AI code can not be copyrighted" decisions are about: The AI itself can not hold the copyright, or the creator of the AI/LLM can't have copyrights.
In other words: there is no code under copyright of Claude or Gemini. Or the owners/creators of Claude or Gemini.
Re: (Score:2)
People say the GPL is viral. So slap some GPL code next to the public domain output of the LLM and you're fine.
The more detailed truth is, that the public domain code stays public domain itself, but usually is not useful without the GPL components, that require the user to adhere to the GPL. If you are really able to rip out the unchanged LLM output, you can use it freely. That's similar to adding let's say BSDL code to the project. The GPL does not really infect it, but when using the combined project the
Re: (Score:1)
Re: (Score:2)
Nope. GPL software has absolutely no problem to be combined with public domain code.
This is a Opened on May 14 (Score:1)
I've been contributing to godot projects this year. When this PR came up [github.com], I thought it was interesting. It was closed as not planned.
Protect godot from AI slop PRs #119449
Closed as not planned
Description
@kimjune01
kimjune01
opened on May 14
7 of 197 PRs in the last 2 weeks would have been auto-closed before review (e.g. #119405 WIP: Add Web C# export pipeline pieces, #119379 Add offscreen display server rendering path).
Install a GitHub Action:
Full writeup
/quote.
This was partially generated by a clanker. The human that sent me cares about your repo. We just wanna help.
ScummVM has a similar policy (Score:2)
This closely mirrors the recent AI policy decision by the ScummVM group; we too were seeing an uptick in automated AI generated PRs, particularly by GSOC (Google Summer of Code) applicants. That in itself is concerning, since there seems to be more and more spurious AI-generated applications each year.
https://github.com/scummvm/scu... [github.com]
Disturbance in The Force (Score:2)
but humans must understand, own, and be able to fix the code they submit.
I wonder how many software shops would have to close their doors if this were adopted industry-wide.
Godot got to a good start--without AI (Score:2)