Leah Crane covers a fascinating intellectual property question for NewScientist (5 January 2019, paywall):
As artificial intelligence algorithms play an increasing role in media production, questions of ownership are becoming fuzzy.
Creating something with an AI takes three steps. First, someone codes the algorithm itself, then it must be fed masses of data to teach it to recognise and mimic patterns, and finally the AI produces some sort of output.
When it comes to copyright, the big question is who owns that output: the person who built the algorithm, the person who picked the training data or the person who selected the specific output?
In some ways, this isn’t a new conundrum: think of bands arguing over who should own the rights to a particular song when one member wrote the chord sequence and another the solo, says Tom Lingard, an intellectual property and technology lawyer.
Elevating current AI systems to the status of bandmate is probably going too far. Both artists and lawyers say they are more like word-processing programs: if nobody types into one, there can be no essay. The software might check your spelling, but the thing that makes an essay unique is the writer.
This is really all about the definition of Art, isn’t it? That’s a notoriously difficult subject, as exemplified by the American judiciary’s travails in defining the difference between art & porn. The name of that case is Jacobellis v. Ohio, and it’s my impression that the case revolves around the end product, production methods considered irrelevant.
Nor can this article decide if the end product controls, or if the process matters. I ran across this same divide in a BBS debate decades ago, in which a student of guitar (flamenco, if memory serves) was debating with a synthesizer musician concerning the nature of excellence in music, and neither would gave way. The former was fixated on the performance experience, while the latter was concerned about the final product.
So which really defines art?
Is it a product, tangible or intangible, of little appreciable functionality, which expresses an idea? Or is it how the artist creates that product? Or some combination?
I dunno.
I’m tempted to take the mechanical approach. Is an “AI” (insert tirade about how it’s not AI because there’s no independent agency involved, it’s ML or even something less definable) a tool of the artist who’s choosing to create this piece of art, akin to a paintbrush? A super paintbrush? This pushes aside the dubious thought that the “AI” owns it, and pushes it on the agent motivating the actions that create the art. The writer of the algorithm, assuming they are separate from the motivator, also is deprived of ownership rights. This would make sense as the maker of the paintbrush also has no ownership rights to the art made using the paintbrush[1].
But there’s a problem with this reductionism, as Crane points out:
It is likely that the question of AI copyright will be answered by some future lawsuit that sets a precedent and trickles through the courts. If the ruling is that AI art cannot be copyrighted, it could kill the genre entirely as artists refocus on work that can pay their rent. Yet if the law decides that such art can be protected, it could damage other methods and industries as AIs flood copyright offices with millions of applications and simply wait for someone else to infringe them.
I’ve often said that computers are multipliers while people are merely adders, but I hadn’t envisioned this particular sticky wicket. Although I’m not sure why anyone would wish to sue the typical artist, whose net worth is often negative. On the other hand, would all this art be salable? The economics and legal seem to be twisting before me on this subject, which may, in itself, mark good art.
I suppose one approach is to ask whether a specific piece of art was produced with the knowledge of another piece of art, and, if so, it’s infringement, otherwise not. Another approach is to simply acknowledge that art produced via “AI” is not copyrightable. While Crane suggests that this will kill the industry, I’m not convinced. I think a lot of people buy art because it’s pretty, not because it was done by any particular artist. Oh, I’m not denying the existence of collectors who use authorship as their criterion of collection, but I’m not convinced that they are a majority of the market for art.
But, unless one is willing to create parallel copyright systems for art produced with and without the assistance of “AIs”, this appears to be quite the intellectual puzzle.
And, yet, it pales against the problems that would be brought about by the true AI artist. Keep in mind that, by AI, I mean the entity has self-agency, that is, it prefers to direct itself in its activities. Now, assuming that such a capability does not impair its abilities to create great masses of art, and I think that’s a fairly large and problematic assumption to make, this could really bring hell to the art and copyright markets of the world. Think of someone OCD, with no need to sleep, and practically swimming in its medium.
1 This reminds me of the odd story I had from the late Jeff Prothero, aka Cynbe ru Taren, concerning a kerfuffle he had with the creator of a computer language, name unremembered, who claimed ownership rights to the programs created by others to be compiled by the compiler for his language. Cynbe claimed he took him to court over the ownership rights and won. Thus, a compiler is a paintbrush.
