5

The image below is generated in real time by a Generative Adversarial Network trained on existing works of art (try reloading the page). The process is described in their paper which also demonstrates that it is indistinguishable by humans from art generated by contemporary artists and shown in top art fairs. They make no claim of copyright, freely provide the code that produces it and similar techniques have been used to produce many forms of work that would be copyrightable if produced by a human.

Can we say if it would be possible to successfully claim/defend copyright on such work in any way? Has anyone actually managed to defend such a claim? I am aware of the different decisions that have been made around AI as an inventor in patent law, but copyright seems very different. This question is somewhat prompted by this answer, but the question is different. Any jurisdiction would be interesting.

Click to see: Generative Adversarial Network produced work of art

Improve this question

5 Answers 5

Reset to default
3

I am reasonably sure that works like this would be Public Domain, i.e. not copyrighted in the first place. In Germany, copyright (or more precisely, Author's Right) can only be held by a natural person, which is a legal term of art that essentially means "human being". There are two groups of natural persons involved here: the programmers and the creators of the works which are part of the training set. However, I can't see how either group could have a reasonable claim of copyright.

Note: there have been similar discussions about GitHub Copilot, an AI that writes code for you. It is my understanding that the process used by GitHub Copilot is roughly comparable to the process used here. With GitHub Copilot, there have been instances where significant snippets which are part of the training set have appeared in the output with only minor alterations. In this case, it is likely that the original author will have copyright over that portion of the output.

However, IFF a human being were actually involved in selecting specific works, then there is probably a copyright based on the creative decision of selecting this particular work and rejecting all the others. This is similar to the classic textbook example of a driftwood sculpture: the creative process here is not creating the sculpture but choosing to pick up this particular piece of driftwood instead of the hundreds of others on the beach.

So, if you simply generate these works and publish all of them, there is no copyright. If you generate a large batch and then select a certain number, then the person who made the selection might hold a copyright.

Improve this answer
1
  • 1
    There are always decisions involved in running a program. The 1st decision is to execute it. But in the case of AI, there are decisions on the set of training inputs, etc.
    – grovkin
    Commented Jan 28, 2022 at 8:46
2

The work is public domain

To have original copyright a work must have a qualified person as author: “qualified person means an Australian citizen or a person resident in Australia.”

The Regulations extend coverage to works published in international jurisdictions “as if the foreign work or subject‑matter were made or first published in Australia.” So, the still require a “qualified person” but the definition would be changed to substitute the foreign nationality/residence.

Australian courts have held that AI can be listed as inventors (but not owners) of patents, the Patents Act does not require an inventor to be a person the way the Copyright Act does for an author. Current case law, Telstra Corp Ltd v Phone Directories Co Pty Ltd [2010] FCAFC 149, found that a predominantly computer created telephone directory was not subject to copyright due to the absence of an author.

Improve this answer
1

Since the "contemporary artists" have not been dead for 70 years, their works are copyrighted and may be used as training data only with an appropriate license, which should specify terms under which derived works can be produced and whether royalties must be paid in return.

For practical purposes, that means it makes no real difference if the trainer, operator or prompt gives acquire partial copyright, their contribution is dwarfed by that of the artists that produced the original works.

Improve this answer
3
  • 1
    If a human art student looks at copyrighted works and then learns from them then what is happening is that a biological neural network is being trained to produce art based on those works. If we construct an artificial version of that same process, one would think that the rights of the original artists would remain the same.
    – user4574
    Commented Sep 18, 2023 at 21:53
  • There is a massive debate in the art scene on works "in the style of x", so training biological neural networks to produce art based on existing works is by no means uncontroversial -- and that is if you accept the premises that art is a product that is decoupled from its context, that artificial neural nets are the same as biological ones, that training on a specific set is the same as special training on top of general life experiences, and that a production process that has no feedback loop back from society as a whole is comparable to one that does -- all of which I'd question.
    – Simon Richter
    Commented Sep 19, 2023 at 4:07
  • The closest thing to an existing thing is probably sampling, and if you look at the output of sampling artists like The Prodigy, you can see that the list of credits on their productions is ...extensive.
    – Simon Richter
    Commented Sep 19, 2023 at 4:10
1

Nobody has a copyright

A few years ago, there was a discussion if a monkey can have copyright, but the case settled. Now, USPTO enacted a rule together with the library of congress on 16th march 2023, that only a human can create a copyright interest. See their rule (emphasis mine):

II. The Human Authorship Requirement

In the [Copyright] Office's view, it is well-established that copyright can protect only material that is the product of human creativity. Most fundamentally, the term “author,” which is used in both the Constitution and the Copyright Act, excludes non-humans. The Office's registration policies and regulations reflect statutory and judicial guidance on this issue.

[...]

III. The Office's Application of the Human Authorship Requirement

[...]

When an AI technology determines the expressive elements of its output, the generated material is not the product of human authorship.[31] As a result, that material is not protected by copyright and must be disclaimed in a registration application.[32]

Improve this answer
2
  • "only a human can have copyright interest" is not the same as "only a product of human authorship is protected by copyright," which is what the quoted material actually says.
    – phoog
    Commented Jul 23, 2023 at 7:24
  • @phoog I should have typed create. Now better?
    – Trish
    Commented Jul 23, 2023 at 9:33
0

U.K. legislation allows for computer-generated work to be the subject of copyright. A work is computer generated when the work is "generated by computer in circumstances such that there is no human author of the work."

Further,

In the case of a literary, dramatic, musical or artistic work which is computer-generated, the author shall be taken to be the person by whom the arrangements necessary for the creation of the work are undertaken.

However, it is widely recognized that this leaves ambiguity about who this would be in any given instance.

Improve this answer

You must log in to answer this question.

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