0

The UK Government says this about the relevance of substantial part to copyright law:

Infringement is where someone uses the whole or a substantial part of your work without your permission and none of the exceptions to copyright apply.

Substantial part

A substantial part is not defined in copyright law but has been interpreted by the courts to mean a qualitatively significant part of a work even where this is not a large part of the work. Therefore, it is quite likely that even a small portion of the whole work will still be a substantial part.

What does "qualitatively significant part of a work" part of a work mean practically? Are we able to at least put some bounds, such as "Fewer than X words will definitely not be, and more than Y definitely will be" or anything? The best answer would probably reference cases which were found in opposite directions.

The particular case I am thinking about is if an Artificial Intelligence Large Language Model generated part of ones creative work, what factors would one consider in assessing if the amount was significant? The work could be an English/human readable work or computer code. It would seem that with such text based systems one could quantify the extent of the work that was taken quite objectively.

Improve this question
4
  • I would assume that the definition you quoted is all the legal definition there is. Are you looking for case law where some specific examples where deemed to be or not to be a substantial part? What else do you hope to see in an answer?
    – quarague
    Commented Jul 3 at 8:14
  • @quarague As I say in the question that woudl probably be the best I could hope for, but in the absence of that some authoritative writing I guess. Ot seems liek the sort of thing that comes up.
    – User65535
    Commented Jul 3 at 9:35
  • If your LLM creates the very same output as a book, then this begs the question: Do you have a license to reproduce the book? And did you infringe the copyright of the book by training? The letter is up in several courts at the moment.
    – Trish
    Commented Jul 3 at 13:05
  • @Trish Indeed, but that is a different question.
    – User65535
    Commented Jul 3 at 14:59

1 Answer 1

Reset to default
2

The interpretation precludes any quantitative rule, like word count. Rather, its a judgement call about whether its the important bits of the work. For instance, a scientific paper will have lots of details about previous work, methods etc, but the meat is the results and conclusions. So even if the results and conclusions are quantitively only a small fraction of the work they are qualitatively the most significant part.

Improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .