@tante I feel a little hope.

@tante

Wow!

You're arguing, in part, that copyright infringement is problematic.

If code that contains infringed material is found on Codeberg, that will cause problems.

That seems like a perfectly reasonable point to make. I sure cannot argue with that.
(remembering the years of SCO vs IBM).

Putting aside all the other issues that people have with AI generated code, this is a very important point to raise.

Thank you!

@tante Communities are free to set their own standards.

@tante this is important for another reason: AI shops want you to label what you did with their bots so they can filter it out of their training set. If you're forced to hide your machine-generated contributions then they're more likely to end up in the training sets and poison the future generations of the models with their own outputs.

@tante So proud of @Codeberg e.V. doing the right thing here. I'm a Codeberg user since 2020 and it's so nice to see how they developed into something much bigger without compromising their values.

@tante This reinforces my decision to move my projects to Codeberg. They get it.

@tante it's good, I just wish they had followed Zig's lead here and gone scorched earth against slop

@tante That’s a win!

@tante I do in fact not know much about Codeberg. But I think it's great to ban AI. If somebody has bad code in their projects, they at least can not put it on AI even if it is.

@tante sorry, the "how will people enforce this" argument is real, dismissing it will not make it disappear.

I am a regular Codeberg user, left GitHub in 2019, and co-maintain Libre projects that saw very little if none at all AI in development.

And still I think this decision will backfire badly.

I, like most of Codeberg users, was not even aware that this discussion was in place, did not vote. This only passed because of the anti-AI Luddite crowd. Which, mind you, has a point, I have very conflicting feelings about AI myself, and I think that the current trajectory is quite worrying. But this is just a knee-jerk reaction that has clearly not been thought of well enough.

I applaude the philosophy, or the intent if you want, of not having Codeberg turn into another SlopHub. But this is fairy-tale wishful thinking. There is no way to reliably detect AI-generated code at scale in 2026.
I can already smell the mutual witch hunts across projects and "camps".

Anyway, I hope I'm wrong.

#codeberg #ai #llm #drama #opensource #ludd #luddism #antiai #libre #foss #floss #slop #github

@tante

As far as I'm concerned if a non profit (or any others) wants to ban slop submissions, they can ban slop... if you want to write slop you can send it to a slop friendly enshitifier.

@tante I don't often comment on things because I hardly know anything, but with this I simply feel that this is an important provision in the evolution of the digital human, or humanity or whatever this is we've gotten ourselves into... like a sanctuary for code principles...

@tante

7. You must not share projects that mostly consist of code written by "generative AI"-tools (including services such as *Claude*, *OpenAI Codex*). Such projects having an unclear copyright status (see requirements § 2 (1) 1 and § 2 (1) 3) and furthermore have little safeguards to ensure that they do not include harmful code (c.f. § 2 (1) 5).

Is the clause that has been added. I really wish the Codeberg aup was better written. So now we have no public domain code, no private projects, no code they think is mostly AI generated. Bit of a mine field there...

Not that there's no place for a project like this it's just never going to be that big which makes me sad.

@tante Good news. The sad news is that with only slop free contributions it will be a prime target for AI scraper bots.

@tante ...just wondering if AI-assisted coding is also banned? At what point becomes AI vibe code and when is it still human-made with help of AI-tooling?