[frabcus]

Germany copyright law feels like the relevant thing here - Codeberg is a non-profit based in Germany.

Best quick English language overview of status that I could find: https://www.twobirds.com/en/insights/2026/germany/when-can-a...

It looks like Codeberg want only copyrighted material in their service, so it is reliable in the future that e.g. licenses must be followed (e.g. GPL), and copyright doesn't suddenly get declared as being of the model owner, and it isn't a copy of something else.

That is a cautious reasonable position - in early days of LLM coding (3 years ago!) indemnity from model companies was a major issue globally because of the lack of clarity of the law around this. The US specifically has settled on it being (effectively?) public domain. But I don't think that is fully settled, and it certainly isn't settled in international copyright law.

The goal of the vague "mostly" in the Codeberg change is to ensure there is enough human input to the code they host, to be reasonably sure under German copyright law it is copyright of the person sharing it.

[graemep]

> copyright doesn't suddenly get declared as being of the model owner

That does not seem to be likely. The article you link indicates it is the prompter and provider of any other inputs (in the case discussed the photographer who took the photo the LLM modified) who might have claims.

> The US specifically has settled on it being (effectively?) public domain.

My understanding is that sufficient human input makes it copyright. This is not greatly different from German law.

In both cases something that is vibe coded in the strict sense of having only vague prompts will not be covered by copyright, something with real human input will.

That leads to a problem with "mostly" there. Does it mean mostly vibe coded, or mostly AI written? AI written under close guidance is covered by copyright so should be fine.

This does make it important to keep prompt history and record human edits of generated code as that is what will prove sufficient human input.

Also, its not just German law that matters to users. For many users law in their jurisdiction is what matters.

[FinnKuhn]

German copyright works a bit different from US copyright, as it requires a "personal intellectual creation", but there have already been multiple court cases that decided AI generated content is not protected by German copyright as it doesn't meet this requirement. You can find the original (German) verdict here as it goes into some more legal details on this: https://www.gesetze-bayern.de/Content/Document/Y-300-Z-BECKR...

So from an open source perspective I don't see any legal risk here as the worst that could happen is that there is no legal protection of the code, which would only give users more and not less rights. Also not sure what they mean by "unclear" copyright status. At least from what I can tell it seems pretty clear that there is generally no copyright on AI output.

[graemep]

> German copyright works a bit different from US copyright, as it requires a "personal intellectual creation"

Can you explain how it is different from US law? It sounds closer to US law than to, for example, British law.

> You can find the original (German) verdict

I cannot read German. So you are saying Bird & Bird's explanation of the ruling is wrong? Or is it contradicted by other cases?

[FinnKuhn]

> Can you explain how it is different from US law? It sounds closer to US law than to, for example, British law.

Yes, there are some similarities. The "personal intellectual creation" requirement I think is also similarly used in the US where only genuine intellectual creations by the author count as works. Typing "generate an image of a flower" doesn't clear that bar. If you however write very detailed long description of the image then it could clear that hurdle. The verdict says that the personal creative input needs to dominate the result if you sue someone over it, it is on you to proof that your prompt meet this hurdle, which "vibecoders" are not going to meet I believe.

However, in Germany copyright always belongs to the creator (a natural person), while in the US it can also be held by a company. In practice, this means you can't sell your copyright, only license it, although even an exclusive license has limits. So an AI provider never holds copyright over generated output either way: it either belongs to the creator or the output has none. German copyright law aims to protect the creator while US copyright law focuses more on economic aspects.

So the result is that either the author has the copyright (so just like any other code) or there is no copyright. Either way I don't see what issue uploading it to Codeberg could cause for them as it should be fine either way.

> You can find the original (German) verdict

I tried to explain the argument from the verdict in more detail above, otherwise number 18-22 are the relevant part. See here for the translation of those: https://claude.ai/share/4d52fb18-a375-4ce7-9d7f-35e3d52971aa The Bird & Bird article seems to pretty much say the same from what I can tell, just in more detail.

[frabcus]

Given a German court talked about not having the prompt logs making it hard in a specific case to prove it had sufficient human input, maybe someone could do an open source hosting service where the Claude Code logs were fully uploaded with each commit, so that could be later proven? It would be cool anyway to have such a service, so people could learn how to do LLM coding better from the examples.

[bxff]

If I had to guess this would make distillation 'attacks' easier, and I am not sure if it's allowed under most AI labs ToS. Anthropic goes pretty hard against keeping Claude Codes session histories local, opting to delete them if they aren't touched in the last 30 days by default.

[veeti]

Once you look at how German anti-nuclear lobbying set Europe back decades on its energy independence and made us dependent on authoritarian hellholes of the world, it all makes sense. They just want to do the same with AI.

[andor]

It’s probably pointless to argue, but there was never a lot of nuclear power in Germany to begin with. This is just turning energy policy into culture war.

[kalessin]

Not a lot is inaccurate given the carbon intensity of coal and its importance in Germany's energy mix. Anyway, I think the nuclear debate is quite off-topic here.

[veeti]

Yes, when you spend decades arguing "nuclear bad" (or "AI bad"), it turns out you get left in the past as Jurassic Park.

[InsideOutSanta]

Jurassic Park is science fiction, though.

[veeti]

Unfortunately the truth is far worse, and some of us have to share a union with these fools ;)

[flohofwoe]

    > user: veeti
    > about: Just trolling.

Well, at least you're upfront about it ;)

[lionkor]

Open source code that is AI authored is worthless. If I need an LLM authored, idk, terminal emulator, I will simply ask my AI or use one of the tens of thousands out there.

[samatman]

Like Ghostty, you mean? The best terminal emulator out there?

[veeti]

I use Linux every day, where AI coauthored analysis and code has already contributed to harden its security. Like it or not, any project that categorically rejects AI contributions or engages in vague activist type purity spiral testing over "vibe coding" seems like a security liability at best.

[graemep]

It depends what you mean by "AI authored". If you mean one-shotted vide code, then yes. If you mean someone providing detailed instructions for architecture and functionality and reviewing the code, then it does have a lot of value.