On 2026-05-28 05:01, Lars Noodén via libreplanet-discuss wrote:
On 5/26/26 16:09, Jean Louis wrote:
Your statement "The US Copyright Office declared that LLM output is
non- copyrightable." is wrong, it is in the wrong context and without
the context!
In Part II of the report on Copyright and Artificial Intelligence,
https://www.copyright.gov/ai/
it concludes with the observation that LLM output, and AI output in
general, is ineligible for copyrights unless that output is modified by
a human for the purposes of creative expression:
"As described above, in many circumstances these outputs
will be copyrightable in whole or in part—where AI is
used as a tool, and where a human has been able to
determine the expressive elements they contain. Prompts
alone, however, at this stage are unlikely to satisfy
those requirements."
https://www.copyright.gov/ai/Copyright-and-Artificial-Intelligence-Part-2-Copyrightability-Report.pdf
Yes, I agree, it says same what I said and meant. But you quote me and
follow with citation, is it to support my argument or to counter against
it?
I think same as copyright office, as it is common sense. If human has
created anything, no matter the tool, it is copyrighted.
But if you ask computer to display the picture of popular painter, and
then you blur the picture, or distort it a bit, how can that be
copyrightable? That is same analogy to LLM creation of works without
substantial human input.
Thus the problem raised in Andy's original message, that LLMs can be
and are being misused to strip licensing (and attribution) from
projects' code bases.
"can be" I agree, but not mainly. If I am mistaken, okay, just sensitive
on generalization and blatant statements which are not helpful to
public.
And you do not need LLM to strip licensing and attribution. So why blame
the tool for human actions?
The code goes into the large language model with a license and a
copyright holder.
Not generally.
There are millions of LLMs out there, so can you be specific and say
which large language model you specifically talk about? I am sure you do
not talk about OLMO models (just as example)? So do you?
Calling that infringement a training set doesn't change its nature.
Then the code comes back out of the model as slop, and while slop in
general is ineligible for copyright status the original code remains
strongly protected by copyright, whether plagiarized by an LLM or not.
I agree on that statement, but please, many people run locally LLMs, so
please not assume we do that. LLMs do not do that. It is human doing. Do
not anthropomorhize LLMs like having some intention to do that.
And, again, as mentioned in an earlier reply, there is the further
problem of the LLMs separating users from projects and vice versa.
Wow, you are talking like we are already in Odysee 2001... Hal...
separate user from the project.
It is not LLM separating any user from project!
"COPE joins organisations, such as WAME and the JAMA
Network among others, to state that AI tools cannot
be listed as an author of a paper."
Exactly. So your statements are contradictory. First you speak of LLM
like human living being, then you say it cannot be designated as
author...
It would not be a stretch to apply a similar conclusion to code which
has been mangled by an LLM.
It was not LLM. It was human.
--
Jean Louis
_______________________________________________
libreplanet-discuss mailing list
[email protected]
https://lists.libreplanet.org/mailman/listinfo/libreplanet-discuss
