Liability, Authorship, & Symmetrial Causation in AI-Generated Outputs

By: Jacob Alhadeff

Copyright has insufficiently analyzed causation for both authorship and liability because, until now, causation was relatively obvious. If someone creates a painting, then they caused the work and receive authorial rights. If it turned out that the painting was of Mickey Mouse, then that painter may be liable for an infringing reproduction. However, recent technological advances have challenged the element of causation in both authorship and infringement. In response, recent law and scholarship have begun to address these issues. However, because they have addressed causation in isolation, current analysis has provided logically or ethically insufficient answers. In other words, authorial causation has ignored potential implications for an entity’s infringement liability, and vice-versa. Regardless of how the law responds, generative AI will require copyright to explore and enumerate the previously assumed causation analyses for both infringement and authorship. This blog explores how generative AI exposes the logical inconsistencies that result from analyzing authorial causation without analyzing causation for infringing reproductions.

Generative AI largely requires the following process: (1) an original artist creates works, (2) a developer trains an AI model on these works, and (3) an end-user prompts the AI to generate an output, such as “a mouse in the style of Walt Disney.” This generative AI process presents a novel challenge for copyright in determining who or what caused the output because generative AI challenges conventional notions of creation.

Causing Infringement

Andersen et al. recently filed a complaint against Stability AI, one of the most popular text-to-art foundation models. This class action alleges that Stability AI is directly liable for infringing that result from end-user prompted generations. However, in a recent decision more closely analyzing causation and volition in infringement, the Ninth Circuit found that “direct liability must be premised on conduct that can reasonably be described as the direct cause of infringement.” Stability AI should not be found directly liable for infringing these artists’ copyright, in part because Stability AI cannot reasonably be said to be the direct cause of infringement. Such a finding would be similar to holding Google liable for reproducing images of Mickey Mouse on people’s computer screens when they search for “Mickey Mouse.”  

This lawsuit is particularly relevant since end-users have prompted thousands of generations that include the phrase “Mickey Mouse” and many appear substantially similar to Disney’s Mickey. If thousands of end-users have intentionally prompted the AI to generate Mickey Mouse, then what volitional conduct can most reasonably be described as the direct cause of infringement? It is clearly the end-user. However, what if the end-user simply prompted “a cartoon mouse” and the AI generated an infringing image of Mickey? Here, the end-user may not have intended to generate Mickey and reasonable notions of fairness may not find the end-user as the most direct cause of infringement. However, copyright is a strict liability tort, meaning that liability attaches regardless of a reproducer’s intent. Therefore, unless copyright applies an intentional or a negligence theory for direct liability, which it should not, then whomever or whatever is liable for infringing outputs shall be liable for both of the infringing outputs— “Mickey Mouse” and “a cartoon mouse.” Such an outcome not only feels deeply unfair, but it is unreasonable to say that the end-user is the direct cause of infringement when prompting “a cartoon mouse,” and vice versa. 

Cases called to answer similar questions have recently grappled with these same issues of volition and causation. Generally, courts have been hesitant to find companies liable for actions that are not reasonably deemed volitional conduct causing infringement. The court in Cartoon Network, for example, found that “volition is an important element of direct liability.” In the Loopnet case, the court found that “the Copyright Act… requires conduct by a person who causes in some meaningful way an infringement.” In this way, the law has so far mirrored our prior intuitions of fairness. Legal scholarship has noted that when copyright law has grappled with novel technology, it has found that causation in infringement requires volition that “can never be satisfied by machines.” This reasoning, as applied  to generative AI, may mean that an AI company should not normally be directly liable for outputs that infringe the reproduction right. 

Causing Authorship

This causation analysis has also begun for authorship rights. One copyright scholar compellingly argues that copyright law should explicitly enumerate a causal analysis for granting authorship rights. Such an analysis would follow tort law’s two step causation analysis including: (1) creation in fact and (2) legal creation. Aviv Gaon surveys authorial options in The Future of Copyright in the Age of AI, writing that there are those that favor assigning authorship to the end-user prompter, the AI developer, finding outputs joint works, or even attributing authorship to the AI itself. The simplest legal option would be to treat AI like a tool and grant authorship to the end-user. This is exactly how the law responded when photography challenged conventional notions of creativity and authorship. Opponents of finding photographers as authors argued that photography was “merely mechanical, with no place for… originality.” The Supreme Court in Burrow Giles instead found that the photographer “gives effect to the idea” and is the work’s “mastermind” deserving of copyright. 

However, treating AI like a conventional tool is an inconsistent oversimplification in the current context. Not only is it often less analogous to say that an end-user prompter is the ‘mastermind’ of the output, but AI presents a more attenuated causation analysis that should not result in  a copyright for all AI-generations. As an extreme example, recent AIs are employing other AIs as replicable agents. In these circumstances, a single prompt could catalyze one AI to automatically employ other AI agents to generate numerous potentially creative or infringing outputs. Here, the most closely linked human input would be a prompt that could not be said to have masterminded or caused the many resultant expressive outputs. Under Balganesh’s framework, no human could reasonably be found as the factual or legal cause of the output. Such use-cases will further challenge the law’s notions of foreseeability as reasonable causation becomes increasingly attenuated.

Importantly, in the face of this ongoing debate and scholarship, the Copyright Office recently made their determination on authorship for AI-generated works. In February 2023, the US Copyright Office amended its decision regarding Kristina Kashtanova’s comic book, Zarya of the Dawn, stating that the exclusively AI-generated content is not copyrightable.  Ms. Kashtanova created her comic book using Midjourney, a text-to-art AI, to generate much of the visual art involved. The copyright office stated that her “selection, coordination, and arrangement” of AI-generated images are copyrightable, but not the images themselves. The Office’s decision means that all exclusively AI-generated content, like natural phenomena, is not the type of content copyright protects and is freely accessible to all. The Office’s decision was based on their interpretation that “it was Midjourney—not Kashtanova—that originated the ‘traditional elements of authorship.’” The Office’s decision is appropriate policy, but when analyzed in conjunction with the current law on causation in infringement, it is inconsistent and may result in an asymmetrical allocation of the rights and duties that attend creation. Relevantly, how can a machine that is incapable of volition originate art? This is one of many ontological paradoxes that AI will present to law. 

Symmetrically Analyzing Causation

Two things are apparent. First, there is a beautiful symmetry in AI-generations being uncopyrightable, and the machines originating such works symmetrically do not have sufficient volition to infringe. If such a system persists, then copyright law may not play a major role in generative AI, though this is doubtful. Second, such inconsistencies inevitably result from causation analyses for mechanically analogous actions that only analyze one of infringement or authorship. Instead, I propose that copyright law symmetrically analyze mechanically analogous causation for both authorship and infringement of the reproduction right. Since copyright law has only recently begun analyzing causation, it is reasonable, and potentially desirable, that the law does not require this symmetrical causation. After all, the elements of authorship and infringement are usefully different. However, what has been consistent throughout copyright is that when an author creates, they risk both an infringing reproduction and the benefits of authorship rights. In other words, by painting, a painter may create a valuable copyrightable work, but they also may paint an infringing reproduction of Mickey Mouse. Asymmetrical causation for AI art could be analogized to the painter receiving authorship rights while the company that made the paintbrush being liable for the painter’s infringing reproductions. Such a result would not incentivize a painter to avoid infringement, and thereby improperly balance the risks and benefits of creation. Ultimately, if the law decides either the end-user or the AI company is the author, then the other entity should not be asymmetrically liable for infringing reproductions. Otherwise, the result will be ethically and logically inconsistent. After all, as Antony Honore wrote in Responsibility and Fault, in our outcome-based society and legal system, we receive potential benefit from and are responsible for the harms reasonably connected to our actions.

Legend of Zelda Mod Drives Nintendo IP Lawyers Wild

By: Nick Neathamer

Has video game fandom gone too far? Despite developing some of the biggest games on the market, Nintendo seems to think it has (at least in a legal sense). The company has recently claimed copyright infringement on multiple YouTube videos that show the use of fan-made modifications (“mods”) for the game Legend of Zelda: Breath of the Wild. 

Breath of the Wild is one of the most popular open-world video games in recent memory. Created by Nintendo, the game was deemed Game of the Year in 2017 at The Game Awards. However, one notable element the game is lacking is any multiplayer capability. YouTuber Eric Morino, better known by his channel name PointCrow, aspired to change that. In November 2021, he tweeted out a request for anyone to create a multiplayer mod for the game, offering up $10,000 to whoever could send a functional version. Two members of the modding community were able to create a mod that runs on a Wii U emulator (software which enables Wii U console games to be played on a PC), allowing multiple players to travel throughout the game’s fantastical setting of Hyrule together. On April 4, 2023, PointCrow released the mod to the public through his Discord (however, it has since been removed). 

After the release, Nintendo claimed copyright infringement on PointCrow’s videos that feature any use of the mod, prompting YouTube to take down those videos. Due to Nintendo’s reputation for being a highly litigious company, the copyright claims against PointCrow’s videos are not a huge surprise. However, PointCrow has argued and appealed the copyright strikes, saying that he has “significantly transform[ed]” Nintendo’s work and that his videos constitute fair use. 

Copyright ownership grants the holder several exclusive rights in regard to their copyrighted work, as laid out in §106 of the Copyright Act of 1976. One of these rights is the right to create subsequent works derived from the original copyrighted work. If someone other than the copyright owner creates such a derivative work, they would infringe the copyright in the original work. Unfortunately for the Breath of the Wild modders, present-day mods have been considered derivative works since the 9th Circuit Court of Appeals’ ruling in Micro Star v. Formgen. While many game developers seldom pursue legal recourse against the majority of modders, and some have even started to embrace the modding community, this derivative work status bars modders from having any copyright of their own in the mods they create. Additionally, if Nintendo does choose to sue for copyright infringement in relation to the multiplayer mod itself, PointCrow and the other creators are likely to be held liable. 

Next comes the question of whether PointCrow’s videos about the mod qualify as fair use. Fair use analysis involves considering four factors in a balancing test. Set out in §107 of the Copyright Act, these factors are (1) the purpose and character of the use, including whether the use is commercial; (2) the nature of the copyrighted work; (3) the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and (4) the effect of the use upon the potential market for or value of the copyrighted work. While courts must consider all four factors, the first and fourth factors are typically considered the most important in deciding whether an allegedly infringing work is a fair use. The first factor is more likely to weigh against fair use when the allegedly infringing work is commercial. However, commerciality may be overcome and the first factor may weigh towards fair use when the work in question has transformed the original, providing it with a new expression, purpose, or meaning. Here, PointCrow intends to monetize his videos on YouTube, making his use commercial. PointCrow’s claim that his videos have “significantly transform[ed]” Breath of the Wild indicate his belief that the videos are sufficiently transformative to warrant the first factor weighing in favor of fair use, despite their commerciality. One could certainly argue that by providing commentary and reactions to the gameplay, PointCrow has transformed Breath of the Wild by granting it a new expression. However, the entertaining purposes of both PointCrow’s videos and the game itself are very similar, despite the difference in watching a game versus playing it. For these reasons, it is difficult to predict whether a court would find this factor to weigh for or against fair use. 

The second factor most likely weighs against fair use. A use is less likely to be fair use when the original work is unpublished, because authors of unpublished works are expected to be able to decide how their work is originally used, or whether it may be released to the public at all. On the other hand, copying of a published work is more likely to be considered fair use. Even more relevant to the nature of the work is if the original work is creative, which tends to weigh against fair use in contrast to when the original work is primarily factual. Here, the second factor most likely weighs against fair use because the original game is a creative work, despite the game’s published status. Meanwhile, the third factor likely weighs in favor of fair use. PointCrow’s videos include actual gameplay, and therefore show large portions of the original game. However, displaying this large amount of the game is necessary to accomplish PointCrow’s intended purposes. Disregarding the legality of the mod itself, PointCrow needs to show gameplay in order to demonstrate differences between the original game and the modded version, as well as to show his unique experiences with Breath of the Wild that viewers want to see. Because of the need to use this large amount of gameplay for his intended purpose, a court is likely to find that the third factor weighs in favor of fair use. 

The fourth factor, effect of the use upon a potential market of the copyrighted work, weighs against fair use when an allegedly infringing work provides a substitute for the original. With this in mind, it is not entirely clear what role PointCrow’s videos play in the video game entertainment market. PointCrow would likely argue that his videos are essentially free advertising for Breath of the Wild and Nintendo, while Nintendo may argue that watching someone play the game essentially provides a substitute for playing the game itself and therefore has a negative effect on the market for the game. A court may also be persuaded by the argument that by promoting the multiplayer mod, which runs on an emulator instead of an actual Nintendo console, PointCrow’s videos are indirectly causing a substitution loss to Nintendo in console sales. This makes it more likely, although not certain, that the fourth factor would weigh against fair use. 

Despite their best intentions and love for the game, it appears that PointCrow and other fans of Legend of Zelda: Breath of the Wild are infringing Nintendo’s copyright by creating a multiplayer mod. Less clear is whether videos that promote the mod are infringing. A lack of existing litigation surrounding gaming videos only exacerbates this uncertainty. With the upcoming release of Legend of Zelda: Tears of the Kingdom, a direct sequel to Breath of the Wild, content creators are likely unsure how to make gameplay videos while complying with copyright laws. That said, Nintendo’s history of litigation has not stopped fans from making their passion projects thus far, and it certainly seems like fans will continue to create mods and videos going forward. But perhaps the takedown of PointCrow’s videos will finally send the message that despite Nintendo’s success at making games, the company is not playing around when it comes to their intellectual property. 

The State of Sampling: The Landscape of Sampling and Copyright Law in 2023

By: Cooper Cuene

De La Soul’s 1989 album 3 Feet High and Rising is a classic and hugely influential record that Rolling Stone recently described as “a landmark of the genre” when ranking it as the 33rd best rap album of all time. Until recently, however, a listener eager to give the album a listen would have quickly realized that it is difficult to find: Despite its classic status, until just last month, 3 Feet High and Rising was not available on any streaming platforms. This is because the album is densely packed with chopped up samples of other artists’ music that until recently either had not been or could not be cleared. Alas, the mere fact that sampling has been around since the dawn of rap music has not meant that the legal structure of the practice has improved for artists today. In a recent high-profile case, the late rapper Juice WRLD was forced to pay Sting 85% of the royalties for his song Lucid Dreams which sampled Sting’s Shape of My Heart. While it remains difficult (and expensive) for artists to use samples of existing music in new tracks, the state of sampling in 2023 is ripe for change. Multiple academics are eager to propose new regimes to regulate the sampling of other artist’s tracks, especially in a day and age where digital tools make the use of samples easier than ever.

The origins of the severe restrictions on the ability of producers to sample music can be traced back to Grand Upright Music v. Warner Bros, Inc., a 1991 decision by the Southern District of New York. This case set early precedent in its full-throated prohibition on sampling in music. The decision invoked the ten commandments in reminding the defendants “thou shalt not steal,” without contemplating that sampling could be a legitimate use of a copyrighted work. At issue was a sample on Biz Markie’s album I Need a Haircut, and the ruling immediately stunted the use of sampling throughout the music industry. As Pitchfork noted in their retrospective review of Public Enemy’s Fear of a Black Planet, following Grand Upright Music it became “forbiddingly difficult and expensive to incorporate even a handful of samples” into a new work. Unfortunately, later developments in the case law would be no kinder to the practice of sampling.

Later significant decisions concerning sampling were handed down in the 2000s and continued to be  unambiguous in their prohibition of the practice. Bridgeport Music, Inc. v. Dimension Films is a prime example of the way that courts have approached sampling over the last few decades. In that 2005 case, the defendant released a movie with a soundtrack containing a track that sampled a short portion of a song called Get Off Your Ass and Jam. The film contained only a four second section of a guitar riff from the original song that had been slowed and stretched to extend to a sixteen-bar loop. Despite the defendant’s argument that the small amount of the original track used was de minimis, the Sixth Circuit still found for the plaintiffs. Their opinion was resounding, commanding artists to simply “[g]et a license or do not sample.” Despite the inflexibility of this standard, it remains valid law today.

Bridgeport’s standard and its rigidity has unsurprisingly sparked calls for reform from musicians and academics alike. A common thread among calls for reform is that a reformed legal regime governing licensing should aspire to more actively promote the Constitution’s grant of power to Congress to “promote the progress of . . . useful arts.” One leading alternative regime is the idea that works that make use of samples include a clear attribution to the original song in their title, much like songs do with featured artists already. John Ehrett is a prominent supporter of this alternative, arguing in his 2011 paper “Fair Use and an Attribution-Oriented Approach to Music Sampling” that the music industry should take up citation standards for samples akin to the specialized citation styles present in other industries. Under Ehrett’s proposal, songs that include samples would include an indication in their titles that they do so, such as “Song A (samp. Song B).” This would ensure that the original work gets the necessary recognition while also providing a smoother sampling process for the artist behind the new work. Others have proposed a sliding licensing scale that requires artists to pay less for a license the more they transform the work. In either case, it has become clear to many artists and academics that the current regulatory regime governing samples is untenable and should be reformed to better enable new artists to create with existing works.

Duped or Duplicated? The Difference Between A Counterfeit And An Accessible Homage

By: HR Fitzmorris

Even non-legally versed consumers know that counterfeit or fraudulent products are illegal. In fact, some may have even experienced the euphoria of getting what they thought was a steal on a new designer product only to find out that it was a different type of ‘steal’ altogether. 

But what about “dupes”?

Unlike counterfeit products, which are copies of trademarked consumer goods meant to be passed off as the real thing, dupes (short for duplicate or duplication) are products that mimic other companies’ popular products without seeking to trick the consumer into thinking it’s the real deal. Dupes usually mimic high-end, in-demand goods and are sold at a much lower price-point—essentially the Gen Z version of a “knock-off.” Dupes have become extremely popular with the rise of social media advertising aimed at younger demographics. Teens that may not be able to afford a wildly trendy Cartier ring ($2,995) certainly may be able to scrounge up the change for the Amazon dupe ($12.99, with free next day shipping!).

The “Real” Fakes

To the everyday consumer the distinction between a counterfeit and a dupe may seem dubious, but in legal terms it’s significant. Counterfeiting is a concept used to “indicate an infringement of intellectual property rights, namely acts (use, manufacturing, or sale, for example) carried out without the consent of the intellectual property right holder.” “Counterfeiting” is the “act of making or selling fake products with the intent to deceive consumers. In the United States, it is illegal to produce, distribute, or sell counterfeit goods.” 

There are more issues with counterfeit goods outside of intellectual property infringement and their morally dubious nature. There are also possible health and safety issues with fake products that flout FDA or consumer protection standards.  There’s of course, the economic harm to legitimate businesses that lose money when their customers are lured away. The ever-present environmental harms associated with the flood of mass-produced, easily discarded items lurk behind the scenes. There’s even concern that counterfeit goods play a role in funding broader criminal enterprises.

Dupes: Duplication or Duplicity?

One of the important elements of counterfeiting is the “intent to deceive,” and this element is a significant piece of what separates dupes from counterfeits. Dupes do not claim to be the real-deal. In fact, part of their allure is that purchasers are getting the same or similar quality and functionality of the original without the original’s branding (and the associated price mark-up). As Claire Kane put it in her article for online publication MIC: 

While “fake” is a dirty word in fashion and “counterfeit” sounds unethical, the more neutral-sounding “dupe” suggests making savvy purchases and “somehow cheat[ing] the system” to get the look for less.

Companies hoping to crack down on dupes and knockoffs face an uphill battle in court. Without distinctive, trademarked branding that makes counterfeits fall within the reach of traditional trademark infringement, brands find little sympathy in the law. The current state of U.S. copyright law as it pertains to clothing and accessories has significant gaps. U.S. copyright law does not fully protect items defined as useful articles, which are “objects having an intrinsic utilitarian function” and “clothing” is the very first example of what counts. So, without the direct, obvious infringement on the branding, companies are unlikely to prevail.

Can You  Smell the Difference?

An especially interesting sector of dupes gaining popularity are designer fragrance dupes. Most dupes, like a certain handbag or shoe dupe, the knockoff brand doesn’t need (or want) to explicitly refer to the original product—it relies on the consumer to ‘get’ the reference. Not so with replica fragrance brands such as Oakcha, Dossier, or ALT Fragrances, which directly rely on references to their designer counterparts in their marketing. In fact, they often list the fragrance they were “inspired by” right on the bottle, or in the product description. So, what makes fragrance such a fertile ground for direct and blatant knockoffs without running afoul of trademark or copyright law? 

The answer is a combination of technology and law. First, new technological developments have made it extremely easy to reverse engineer specific fragrance formulations. Also, while the branding or packaging of a perfume may be eligible for copyright protection, a perfume’s scent is not because the scent serves as the “functional purpose of the product.” This makes it, under trademark law, ineligible for registration with the USPTO (functionality is a bar to registration). Perfumers can look to other avenues of protection such as obtaining a patent over the perfume formula, or trade secret protection, but these protections are difficult and costly to obtain and have drawbacks like disclosure. 

The legal landscape concerning dupes is unique and developing. It is inconsistent across products and complicated across industries. Whether you think dupes are simply knockoffs with a moral makeover or a legitimate industry that provides consumers with accessible choices, the law is unlikely to be the force that stems the tide.

AI Art: Infringement is Not the Answer

By: Jacob Alhadeff

In the early 2000s, courts determined that the emerging technology of peer-to-peer “file-sharing” was massively infringing and categorically abolished its use. Here, the Ninth Circuit and Supreme Court found that Napster, Aimster, and Grokster were secondarily liable for the reproductions of their users. Each of these companies facilitated or instructed their users on how to share verbatim copies of media files with millions of other people online. In this nascent internet, users were able to download each other’s music and movies virtually for free. In response, the courts held these companies liable for the infringements of their users. In so doing, they functionally destroyed that form of peer-to-peer “file-sharing.” File-sharing and AI are in not analogous, but multiple recent lawsuits present a similarly existential question for AI art companies. Courts should not find AI art companies massively infringing and risk fundamentally undermining these text-to-art AIs.

A picture containing text, person, person, suit

Description automatically generated

Text-to-art AI, aka generative art or AI art, allows users to type in a simple phrase, such as “a happy lawyer,” and the AI will generate a nightmarish representation of this law student’s desired future. 

Currently, this AI art functions only because (1) billions of original human authors throughout history have created art that has been posted online, (2) companies such as Stability AI (“Stable Diffusion”) or Open AI (“Dall-E”) download/copy these images to train their AI, and (3) end-users prompt the AI, which then generates an image that corresponds to the input text. Due to the large data requirements, all three of these steps are necessary for the technology, and finding either the second or third steps generally infringing poses and existential threat to AI Art. 

In a recent class action filed against Stability AI, et al (“Stable Diffusion”), plaintiffs allege that Stable Diffusion directly and vicariously infringed on the artist’s copyright through both the training of the AI and the generation of derivative images, i.e., steps 2 and 3 above. Answering each of these claims requires complex legal analyses. However, functionally, a finding of infringement on any of these counts threatens to fundamentally undermine the viability of text-to-art AI technology. Therefore, regardless of the legal analysis (which likely points in the same direction anyways) courts should not find Stable Diffusion liable for infringement because doing so would contravene the constitutionally enumerated purpose of copyright—to incentivize the progress of the arts. 

In general, artists have potential copyright infringement claims against AI Art companies (1) for downloading their art to train their AI and (2) for the AI’s substantially similar generations that the end-user prompts. In the conventional text-to-art AI context, these AI art companies should not be found liable for infringement in either instance because doing so would undermine the progress of the arts. However, a finding of non-infringement leaves conventional artists with unaddressed cognizable harms. Neither of these two potential outcomes are ideal. 

How courts answer these questions will shape how AI art and artists function in this brave new world of artistry. However, copyright infringement, the primary mode of redress that copyright protection offers, does not effectively balance the interests of the primary stakeholders. Instead of relying on the courts, Congress should create an AI Copyright Act that protects conventional artistry, ensures AI Art’s viability, and curbs its greatest harms. 

Finding AI Art Infringing Would Undermine the Underlying Technology

A finding of infringement for the underlying training or the outputs undermines AI Art for many reasons: copyright’s large statutory damages, the low bar for granting someone a copyright, that works are retroactively copyrightable, the length of copyright, and the volume of images the AI generates and needs for training.

First, copyright provides statutory damages of $750 to $30,000 and up to $150,000 if the infringement is willful. Determining the statutory value of each infringement is likely moot because of the massive volume of potential infringements. Moreover, it is likely that if infringement is found, AI art companies would be enjoined from functioning, as occurred in the “file-sharing” cases of the early 2000s. 

Second, the threshold for a copyrightable work is incredibly low, so it is likely that many of the billions of images used in Stable Diffusion’s training data are copyrightable. In Feist, the Supreme Court wrote, “the requisite level of creativity is extremely low [to receive copyright]; even a slight amount will suffice. The vast majority of works make the grade quite easily.” This incredibly low bar means that each of us likely creates several copyrightable works every day. 

Third, works are retroactively copyrightable, meaning that the law does not require the plaintiff to have registered their work with the copyright office to receive their exclusive monopoly. Therefore, an author can register their copyright after they are made aware of an infringement and still have a valid claim. If these companies were found liable, then anyone with a marginally creative image in a training set would have a potentially valid claim against a generative art company.

Fourth, the copyright monopoly lasts for 70 years after the death of the author. Therefore, many of the copyrights in the training set have not lapsed. Retroactive copyright registration combined with the extensive duration of copyrightability means that few of the training images are likely in the public domain. In other words, “virtually all datasets that will be created for ML [Machine Learning] will contain copyrighted materials.”

Finally, as discussed earlier, the two bases for infringement claims against the AI art companies are (1) copying to train the AI and (2) copying in the resultant end generation. Each basis would likely result in billions or millions of potential claims, respectively. First, Stable Diffusion is trained on approximately 5.85 billion images which they downloaded from the internet. Given these four characteristics of copyright, it is likely that if infringement were found, many or all of the copyright owners of these images would then have a claim against AI art companies. Second, regarding infringement of end generations, Dall-E has suggested that their AI produces millions of generations every day. If AI art companies were found liable for infringing outputs, then any generation that was found to be substantially similar to an artist’s copyrighted original would be the basis of another claim against Dall-E. This would open them up to innumerable infringement claims every day. 

A picture containing text, fabric

Description automatically generated

At the same time, generative art is highly non-deterministic, meaning that, on its face, it is hard to know what the AI will generate before it is generated. The AI’s emergent properties, combined with the subjective and fact-specific “substantial similarity” analysis of infringement, do not lend themselves to an AI Art company ensuring that end-generations are non-infringing. More simply, from a technical perspective, it would be near-impossible for an AI art company to guarantee that their generations do not infringe on another’s work. 

Finding AI art companies liable for infringement may open them up to trillions of dollars in potential copyright lawsuits or they may simply be enjoined from functioning.

An AI Copyright Act

Instead, Congress should create an AI Copyright Act. Technology forcing a reevaluation of copyright law is not new. In 1998, Congress passed the DMCA (Digital Millennium Copyright Act) to fulfill their WIPO (World Intellectual Property Organization) treaty obligations, reduce piracy, and facilitate e-commerce. While the DMCA’s overly broad application may have stifled research and free speech, it does provide an example of Congress recognizing copyright’s limitations in addressing technological change and responding legislatively. What was true in 1998 is true today. 

Finding infringement for a necessary aspect of text-to-art AI may fundamentally undermine the technology and run counter to the constitutionally enumerated purpose of copyright—“to promote the progress of science and useful arts.” On the other hand, finding no infringement leaves these cognizably harmed artists without remedy. Therefore, Congress should enact an AI Copyright Act that balances the interests of conventional artists, technological development, and the public. This legislation should aim to curb the greatest harms posed by text-to-art AI through a safe harbor system like that in the DMCA.