Category: Intellectual Property

Are AI Prompts Authorship in Copyright Law?

The production of creative works by artificial intelligence (AI) provokes many responses—philosophical, cultural, economic, and legal. I have already argued against copyright protection for works created by AI, supporting the longstanding doctrine that copyright rights can only attach to works of human authorship. But one paragraph in a recent article by attorney Adam Adler raises a potentially difficult question as to whether human prompts directing an AI to produce work could ever constitute authorship of the resulting work?  Adler writes:

… proponents of AI art don’t have to look very hard to find the required creative contribution. The most prominent AI works are generated through trial and error using specially crafted word prompts. For example, Jason Allen, the winner of the Colorado State Fair, spent 80 hours crafting the prompts he used to generate the art and tested over 900 different prompts before settling on the winner. Given the sensitivity of AI art generators, one could argue that the selection and refinement of prompts (at least as they are used today) involves significant creative work, analogous to placing a camera or framing a shot. And because a human’s prompt selection informs the creation of the entire work, there would not be any obvious way to disentangle the creative and non-creative elements of the work

Whether Adler endorses the view that the prompts in this example should vest James Allen with the rights of authorship in the resulting image, he is probably correct that advocates for copyrightability of AI works will advance this argument. But is the position valid? If I ask a friend to paint a picture of a weeping willow by a brook, my broad description does not constitute even joint authorship in the painting, and that example is arguably no different than my recent playing around with DALL-E 2 writing prompts with an existing painting in mind—Henry Wallis’s “Chatterton” (1856).

Although the results were nothing like the original work (and I am admittedly a novice propter), the image on the left could, eerily enough, be passed off by a would-be forger as an early sketch in the development of the Wallis painting, and it was admittedly astounding to watch these, and other variations appear in a matter of seconds.

PROMPT: A painting by Henry Wallis of Chatterton wearing purple pants and a stained shirt, strewn across his deathbed in a garret.
PROMPT: A painting by Henry Wallis of the poet Chatterton wearing purple pants and a stained shirt, strewn across his deathbed in a garret, the bottle of poison on the floor.

But returning to the theme of this post, I maintain that I did not author these images or the other variations output by DALL-E 2. Authorship flows from the creative choices made by a human, and there would need to be a colorable nexus between my prompt writing and the selection and arrangement of the choices made in the work—in this case a visual work—upon its fixation.

Prompts by themselves may be protectable “literary works” under copyright law—not unlike computer code, which can be sufficiently creative while also serving a utilitarian function as a set of instructions. But the potential copyrightability of prompts themselves does not necessarily extend to protection of the resulting work—not even in the case of Allen writing complex prompts into the app Midjourney to produce the visual work he called “Théâtre D’opéra Spatial.”

Neither the 80 hours Allen spent nor the 900 different prompts he tested has any bearing on a potential claim of authorship in the resulting image because copyright does not protect “sweat of the brow.” Copyright also does not protect ideas or concepts; and the incident of copyrightability is agnostic with regard to the author’s intent, message, or methodology. Unless there is a lack (i.e., less than a modicum) of originality in the work, copyright attaches upon fixation, and without consideration as to how or why the work was made. But the Allen example implies a potential difficulty in the doctrine to which Adler alludes in his description.

Given the time and energy Allen spent on the prompts, we can assume he developed a somewhat complex set of instructions, and it is conceivable that there may be a point at which a creative arrangement of prompts could approach a defensible claim of human authorship in the resulting work output by the AI. But it’s tricky, and I am skeptical.

Ordinarily, the moment a work is fixed in a tangible medium, the human’s creative choices may be inferred, credited solely to the human, and the choices need not be explained. In the photojournalist’s image of the factual event, it is longstanding doctrine that the existence of the image itself is evidence that she made sufficient choices (even in a second or two) to meet the “modicum of originality” threshold, and copyright rights are vested in her without asking her to describe or defend the choices made.

AI production may frustrate this doctrine in the near future by providing a reason to ask how a vast amount of work has been produced and about the nature of the human involvement in its production. Then, even if prompts may be protectable literary works on their own, the consideration as to whether this confers authorship to the human in, for instance, a resulting visual work or music work output by the AI implies a case-by-case consideration of copyrightability that would be administrative chaos for the Copyright Office.

It is hard to imagine a generally applicable doctrine that would harmonize the human authorship requirement with a definable nexus between prompt writing and the resulting work, and this suggests that the law must hold that copyright does not attach where an AI has made any (and likely most) of the creative “choices” in the work being claimed.

This view is consistent with the purpose of art and the purpose of copyright—both of which are profoundly human constructs. Neither copyright’s utilitarian origins (i.e., the author must earn a living), nor its civil rights origins (i.e., the product of the author’s mind is naturally his property) has any meaning whatsoever to a machine, just as machine made “art,” in my view, will ultimately mean nothing to humans.

Art without human creators may be decorative, interesting to a point, useful, entertaining, or even conducive to computer science in other contexts, but the products themselves are bloodless in every sense. Making art and engaging with art is one of the most human of all activities—transcendent and spiritual for many—and I have no idea why it would ever be outsourced to computers. One might as well suggest that the Buddhist set his mobile device in front of the alter to chant for him while he does something else with his body, mind, and hands.

When Henry Wallis revealed “Chatterton” at the Royal Academy in 1856, it caused a stir—both because it was considered a masterwork made early in the painter’s career, and because its romantic yet grim subject matter was viewed by many as a comment on the poor treatment of artists. Chatterton’s suicide by arsenic at the age of seventeen was believed to be at least partly the result of his abject poverty due to the failure of publishers to pay him for his writing. His “Last Verses,” found with his lifeless body, contain an 18th century version of the artist who was supposed to live on “exposure” rather than compensation.

Farewell Bristolia's dingy piles of brick,
Lovers of Mammon, worshippers of trick!
You spurned the boy who gave you antique lays,
And paid for learning with your empty praise.

Two and a half centuries later, “Lovers of Mammon” have invested billions in technologies and business models designed to devalue creative work and infringe copyright rights at massive scale. And now, we enter the next phase, when machines are being “trained” on volumes of human-authored works to potentially replace humans in the production of literary works, visual arts, music, and perhaps eventually, performing arts. And, as usual, the technology is advancing apace without any consideration as to whether it can reasonably be called progress.

Illustration by: zdeneksasek

On Copyright, the EFF Will Say Anything to Scare You

Are you gifted in the art of bullshit, but the popular conspiracy theories just aren’t for you? Do you enjoy riling people up about threats that are demonstrably false, but you just can’t get comfortable with QAnon or election deniers? Well, maybe you should consider an exciting career with the Electronic Frontier Foundation writing about copyright law. It seems they’re always looking for eager young folks with good writing skills and no scruples. For instance, you can learn how to write blog posts like Stop the Copyright Creep, posted this week by Katherine Trendacosta, and which begins as follows:

In 2020, two copyright-related proposals became law despite the uproar against them. The first was the unconstitutional CASE Act. The second was a felony streaming proposal that had never been seen or debated in public. In fact, its inclusion was in the news before its text was ever made public. The only way to find it was when the 6,000-page year-end omnibus was published. We want to make sure that doesn’t happen again.

Here’s how that opening paragraph would read if it stated the facts and concluded with a more honest declaration of EFF’s position:

In 2020, two copyright–related proposals bills became law. despite the uproar against them. The first was the unconstitutional CASE Act. The second was a the Protect Lawful Streaming Act (PLSA). felony streaming proposal that had never been seen or debated in public. In fact, its inclusion was in the news before its text was ever made public. The only way to find it was when the 6,000-page year-end omnibus was published. We want to make sure copyright enforcement does not happen anymore. that doesn’t happen again.

That the CASE Act is unconstitutional is a highly speculative opinion which the EFF is free to argue in court one day, and it probably will. That there was much “uproar” against these bills outside the tiny community of anti-copyright wonks is hyperbole (In 2020? Really?). The implication that the felony streaming bill was quietly snuck into law behind closed doors is just a lie. And, of course, the hyperventilating tone meant to scare readers is standard EFF style, which often makes the organization indistinguishable from any other conspiracy-monger.

The reason I let that statement stand about news being ahead of the text of the felony streaming bill is because it might be half true. The text of the Protecting Lawful Streaming Act (PLSA) was introduced on December 10, 2020, so it definitely appeared late in the session—but not because its provisions came out of the blue, were secretive, or were written without debate. On the contrary, one reason the bill was introduced so late is that various stakeholders were still debating the language into December. And if the EFF did not directly engage at the time, there is no way it was unaware of these negotiations because the issue itself was hardly new.

For more than a decade before introduction of the PLSA, various parties had been addressing the fact that large-scale piracy remained a misdemeanor under the statute rather than a felony as long as the pirate enterprise streamed material rather than made it available for download. Here’s Senator Klobuchar referring to the issue in 2011 by way of analogy during a hearing about cyber security:

I have been working in the area of some of the streaming issues to try to come up with a way with a number of the other Senators to acknowledge that if someone is standing on a street corner and sells DVDs that are over $2,500 that we already know is a felony, and right now if you do it, if you have a business and you are illegally selling anything—movies, books, music—and you do it maybe $1 million and you are profiting—you have to profit from it under our bill—it is still a misdemeanor. And so we are trying to fix that without, you know, hurting anyone’s rights or teenagers that are simply trying to share some information

Or you can link to just one House IP Subcommittee hearing in which felony streaming was being discussed in 2014. “We should consider today whether there is a need to clarify that felony infringement includes infringement by streaming,” stated Chairman Coble. Or you can review a summary of the legislative history at Copyright Alliance. Or, oddly enough, my satirical post A Guide to Critiquing Copyright in the Digital Age was a response in 2015 to an article published in The Hill complaining about the idea of felony infringement by means of streaming.

The desire to close what many copyright owners, the DOJ, IPEC, the FTC, et al. considered a legal loophole had been present in both official and unofficial records for quite some time before introduction of the “secret” bill in 2020. And unless you’re planning to engage in mass piracy by building an illegal streaming platform, you probably have little reason to know what PLSA is or why it exists.

The implication by EFF that PLSA or CASE or the other copyright provisions mentioned in the blog are somehow a threat to free speech is as unfounded as the allegation that these proposals were ever clandestine. Yet, Trendacosta writes, “Because so many copyright-related bills involve proposals that would harm lawful free expression, they are not the kind of controversy-free proposals that belong in such legislation.” Here, she means that PLSA and CASE should never have been attached to a must-pass omnibus spending bill at the end of the year.

Notably, copyright proponents who worked on the legislation were not exactly thrilled with the rider outcome either; but you know why it happened that way, especially with the CASE Act? Because Senator Wyden, had a solitary hold on that bill, and in the roughly nineteen months between introduction of CASE and December 2020, he neither engaged in good-faith negotiation nor publicly articulated a realistic objection to its provisions. In fact, after telling Oregon constituents in December 2019 that he saw a path to compromise, he then asked for several poisoned pill amendments, and failing that, simply ghosted with his hold intact.

So, if we’re going to talk about shenanigans on the Hill and things that should never happen, I’m of the opinion that one senator should not be allowed to single-handedly kill bi-partisan supported legislation without very clearly and very publicly articulating his passionate and well-reasoned opposition. And I believe Senator Wyden agrees because he co-sponsored legislation in 2011 based on this same general principle.

Meanwhile, aside from the fact that copyright’s alleged conflict with speech is a construct of anti-copyright ideologues—one that finds little purchase in history or law—I would love to know what EFF imagines could be more “controversy-free” than copyright legislation? It happens to be one of the very few subjects where we see Democrats and Republicans collaborating and voting in large bi-partisan numbers.

That is not only significant in the current climate, but I am told by at least one colleague who has worked on the Hill in less turbulent times that nothing controversial survives the frenzied process of attaching riders to must-pass legislation. “The leaders of both parties from both Houses have to agree, as well as the chair and ranking members of both the House and Senate Appropriations Committees,” he says. So, that seems like a resounding override of Senator Wyden’s hold and a confirmation that the hold was the only reason the copyright bills were passed in this manner.

What I think EFF really means to say is that because they fervently believe that copyright enforcement is always a threat to speech, they don’t like it when nobody on the Hill agrees with that premise. And these days, it is just so damn easy to allege that when Congress or a court or an investigative journalist or a federal agency doesn’t agree with your trumped-up narrative (it writes itself), then some dark arts must clearly be at work. So, if you’re into sowing conspiracies but can’t go full InfoWars, writing about copyright for the EFF just might be your next career move. Good luck!

Photo by: mrwed54

AWF v. Goldsmith: The Need for a Workable Standard of “Transformativeness”

The Supreme Court on October 12th heard oral arguments in Andy Warhol Foundation (AWF) v. Lynn Goldsmith, and presumably every copyright nerd (pro and con) was listening. In general, I would describe the Court as consistent—all justices focused on the narrow question presented with very little discussion outside those lines. The question, which badly needs an answer, is this:  What is “transformativeness” under factor one if the fair use test?

Petitioner AWF’s argument is that a use of a protected work to create a follow-on work that contains “new meaning or message” is sufficient to find that not only does factor one tilt toward fair use but that “new meaning and message” should be determinative of outcome in any fair use consideration The fact that AWF narrowly asks the Court to consider this one question—and insists that the art world depends on their standard—demonstrates how much weight factor one has accumulated in the years since the decision in Campbell (1992).

Whether the Court opines more expansively on fair use, I think it is safe to say that it will decide whether factor one requires a more rigorous standard—namely whether a follow-on work must contain at least a modicum of comment on the work being used.  While we wait, the copyright skeptics and critics—many who have filed briefs in this case—will say that the fair use doctrine itself, and even the speech right, are in peril unless the Court sides with the Warhol Foundation. But this is simply untrue.

AWF’s very broad interpretation of the first factor analysis is certainly what many copyright critics would like to see, but they are asking the Supreme Court to maintain confusion on the question presented—to give a nod of approval to an application of fair use circumscribed by little more than the imagination of the copier and their lawyers. But if the Court rejects this expansive view—if it clarifies the sprawling confusion as to what “transformative” means—fair use as an affirmative defense will remain a vibrant and appropriately balanced aspect of U.S. copyright law.

Comment is at the Core of Fair Use

As argued in the past, it is no coincidence that the preamble of the statute (Section 107) cites exemplary purposes for fair use that imply the presence of some discussion about the work being used. While understood not to be an exhaustive list, “criticism, comment, news reporting, teaching, scholarship, [and] research” are named because the fair use doctrine evolved as judge-made law weighing cases entailing these types of uses. And because “purpose” is not defined beyond that illustrative list in the preamble, it is reasonable to hold that factor one of the four-factor test, which identifies the “purpose” of the use at issue in the very next sentence of the statute, should not become unmoored from the spirit of that preamble. 

Campbell does not stray from this principle. Some may disagree that parody exists in the follow-on work (2 Live Crew’s “Pretty Woman”), but as long as parody was the basis for the finding, the “comment on” requirement was met—albeit in context to a troublesome term of art called “transformativeness.” Since the popularization of that term in Judge Leval’s paper in 1990, courts and defendants have articulated factor one rationales so broad as to be undefinable. And AWF’s “new meaning and message” standard is precisely that—undefinable as a legal standard.

What is Meaning and Message?

In fact, a precedent case in the same lower court (the Second Circuit) illustrates some of the difficulties with the AWF arguments in this case. In Cariou v. Prince, follow-on artist, Richard Prince, rejected any notion of a “meaning” or “message” in the works he made using photographer Patrick Cariou’s images. What does a court do with AWF’s theory when the defendant himself will not define “meaning and message”? There, the Second Circuit held that some “new expression” was sufficient to find that factor one favored fair use—that comment on the original work is not required.

But tellingly, it was the Second Circuit’s own factor one language in the Cariou decision which provided the district court with a rationale for finding “transformativeness” in the Warhol screens, and which the Second Circuit then reversed stating:

“…the district court appears to have read Cariou as having announced such a rule…that any secondary work is necessarily transformative as a matter of law ‘[i]f looking at the works side-by-side, the secondary work has a different character, a new expression…”

Many legal practitioners have commented on the serpentine reasoning applied by the circuit court in order to square the Warhol and Cariou decisions—and all because “transformativeness” has become a doctrine without clear meaning.

Considering that a defendant artist might present any form of “message”—from baroque to minimalist to Richard Prince’s silence—and that it would be famously unwise for courts to apply a legal standard that turns on the judge or jury’s opinion of artistic merit, the Court should decline to engage in these semantic entanglements.  What is definable and identifiable is whether any comment on the original is present in the new work.  This would provide the factor one analysis with an articulated legal standard that judges and juries are able to apply in a principled way.

Necessity Implies Commentary

Several justices at oral arguments focused on the subject of “necessity,” asking whether it was essential that Warhol use Goldsmith’s photograph to make the “Prince Series” silkscreens. The rational answer to this question in this case is No. It was not necessary, in a legal sense, that Warhol use that particular photograph. Aside from the fact that a middleman (Vanity Fair) obtained the photograph from Goldsmith and provided it to Warhol, even if that were not the case, the necessity question is predicated on the commentary requirement.

Absent commentary on the work used, any alleged “need” for that exact work may be technical or functional but is not defensible as a matter of fair use. For instance, Goldsmith’s photo may be conducive to Warhol’s method because it’s a headshot in front of a plain background, but this kind of “need,” which facilitates the user’s process, is not a proper consideration for courts weighing fair use. Likewise, Richard Prince’s almost haphazard cut-outs of Cariou’s photos for some of his works are more suggestive of “opportunity” than “need,” and my guess is Richard Prince would confirm this assumption.

This view of necessity should focus attention on the question as to whether the expression in the original work has in some way been transformed through commentary upon that expression. By contrast, under AWF’s theory, nearly all uses of, for instance, underlying musical compositions would be fair uses merely by adding new lyrics to famous melodies. This is anathema to fair use doctrine in general and in conflict with Campbell in particular.

AWF’s Theory Did Not Exist in Warhol’s Time

Perhaps it is worth contemplating the legal landscape when Andy Warhol made the “Prince Series.” It was 1981, and most of Warhol’s career was behind him. The current Copyright Act—the first to codify fair use—had only been in effect for about three years, and the seminal application of Judge Leval’s “transformativeness” doctrine was still more than a decade in the future. It is understood that Warhol both appropriated and licensed photographs for his iconic works, though I doubt anyone could prove that he actively contemplated fair use—let alone considered the doctrine  as it has been applied or argued since Campbell.

At present, expansive applications of “transformativeness” have resulted in holdings that treat factor one as the dispositive consideration. Several empirical studies find that a defendant who wins on the “transformative” question is almost certain to prevail on the fair use defense overall,[1] and even where factor one is reasonably the most compelling, this only emphasizes the need for clarification as to the meaning of “transformativeness.”

Far too many decisions, especially in district courts, have placed fair use in direct conflict with the derivative works right, which is little surprise when the word “transformed” appears in Section 101 defining “derivative works.” Further, district courts (e.g., Brammer v. Violent Hues) have applied untenable interpretations of “transformative” in conflict with the most fundamental licensing models under the reproduction right, possibly resulting in needless time and expense on appeals for both parties.

Rather than holding that some evidence of “new expression” almost always carries the day, the other three fair use factors should be given proper consideration by mitigating the apparently mesmerizing effect of “transformativeness.” AWF and its amici’s assumption that its test would create more certainty in the art world is a plea for the status quo in which courts will continue to misapply factor one because “meaning and message” are often undefinable to the point of capriciousness. By rejecting AWF’s over-broad standard, the Court can clarify the vagueness which, since Campbell, has caused unnecessary confusion for rightsholders, users of works, and the courts.

[1] For instance, “Is Transformative Use Eating the World?” – Asay, Clark D. et al.  https://lira.bc.edu/work/ns/5f6a0b59-6497-4457-a063-153dae3cee94