My point in bringing up copyright is that the law is decidedly not settled, and that for the purposes of GnuCash policy, further consideration is needed.
A quick search online shows me numerous articles exploring the question of copyright with regard to AI-generated content. Many point out that copyright of AI content depends on a number of unquantified variables, including: * the overall proportion of AI to human created content; * whether the human performed higher functions on the AI content; * and even the jurisdiction in which the claim of copyright is made. Furthermore (to return to the issues raised by GnuCash policy), several sources noted that with regard to open source projects, the issue revolves around licensing, rather than copyright--a distinction that I admit I don't truly comprehend. The implication in these articles is that, while AI code might not be copyrightable, it might still be licensable. I think, again, to turn back to the GnuCash policy, the fundamental copyright issue surrounding the use of AI-generated code in the project is that there is no way for the human prompter to be certain that the code was not lifted from another copyrighted work. This results from the commercial LLMs using unspecified sources for generation. I wrote the policy the way I did because my sense of the literature suggested that some leeway might be called for, and this may still work as an initial policy. As I have read further in this literature, however, I'm leaning much more towards John's initial implied policy banning such contributions. Without a transparent view into what corpus was used in an LLM, it is impossible to determine intellectual ownership. David T. P. S. I agree with Patrick about your scenario. Unless and until you can demonstrate that your code meets current definitions of copyrightable works, your assertion of GPL status sits on shaky ground. I'll note that, as is true with copyright, software licensing really only becomes an issue when you are attempting to share or sell your work to others. I don't know whether you're trying to do either with your AI code. On August 14, 2026 4:31:29 PM GMT+05:30, David Kirkby via gnucash-user <[email protected]> wrote: >On Fri, 14 Aug 2026 at 08:31, David T. via gnucash-user < >[email protected]> wrote: > >> I personally don't have the legal expertise to judge the conclusions of >> the copyrightability of AI code, and while I respect John's greater >> knowledge and experience in this area, I question whether anyone can tell >> where these legal boundaries lie, and where they will end up. > > >I have asked ChatGPT to develop me code I would release for GPL. It then >inserts comments at the start to say the code is released under the GPL. >So, whilst like yourself I don't have a legal background, I don't think >copyright should be an issue. > >I think asking for AI prompts to be recorded has some merit, but when >logged into an AI system it will remember things you wrote ages ago. > >If we are not at the point where AI generated code has less bugs than human >generated code, we soon will be. > >I notice the need to get AI to comment its code properly. It's also worth >getting a different AI engine to scrutinise the code generated by the >first. > >Dave >_______________________________________________ >gnucash-user mailing list >[email protected] >To update your subscription preferences or to unsubscribe: >https://lists.gnucash.org/mailman/listinfo/gnucash-user >----- >Please remember to CC this list on all your replies. >You can do this by using Reply-To-List or Reply-All. _______________________________________________ gnucash-user mailing list [email protected] To update your subscription preferences or to unsubscribe: https://lists.gnucash.org/mailman/listinfo/gnucash-user ----- Please remember to CC this list on all your replies. You can do this by using Reply-To-List or Reply-All.
