Free Riding as a Pro-Defendant Impulse

Jan 25, 2024 Graeme DinwoodieAdd a Comment

Graeme Dinwoodie

American trademark scholars have almost uniformly decried the role of free riding in calibrating the scope of trademark rights. They have argued that the language of “reaping where you have not sown”—to use the famous, but doctrinally discredited, agricultural metaphor of INS v. Associated Press in 1918—distorts trademark law and expands protection beyond that necessary to ensure consumers receive accurate information about the source of goods in the marketplace. Yet, there is a (relatively well-grounded) suspicion that, despite this almost universal scholarly condemnation, the impulse to protect mark owners against free riding remains a resilient force when courts decide trademark cases. If this suspicion is indeed sound, arguments for less expansive trademark rights that rest on the rejection of free riding as a relevant variable are destined to fail.

In Trademark Free Riders, Mike Grynberg brilliantly and persuasively outlines (with real panache) an alternative approach by which to advocate for less robust trademark rights. He assumes arguendo the reality of judicial attention to (and distaste for) free riding and articulates an approach to trademark law that uses the free riding of trademark owners to justify a less expansive scope of protection. That is to say, what happens when trademark owners reap where they have not sown? If courts are in fact sensitive to free-riding as a relevant consideration in allocating rights, might they not confine the scope of right when evidence of trademark owner free-riding is presented to them?

For example, should trademark law (or courts in trademark cases) not take account of the fact that mark owners may opportunistically exploit meanings developed in popular culture generally or, more narrowly, by their customers? The use of HOGS to describe Harley-Davison motorbikes was initially opposed by the company. Its meaning is attributable to efforts of its customers. Should the tardy and begrudging acceptance by the company of a name developed by its customers figure into the ownership and scope of any trademark rights?

Grynberg thus operates in a reality that too many scholars are not even willing to countenance. This approach is very much consistent with Grynberg’s broader approach to trademark law, having written in an important article 15 years ago about the way that consumer interests can be reconfigured (and differently advanced) to secure a more balanced trademark regime than one-sided consumer paternalism might suggest.

This piece thus highlights a fascinating conundrum for scholars and litigants in trademark cases which parallels that faced by advocates before the U.S. Supreme Court. If, as Justice Elena Kagan suggested in her 2015 Scalia Lecture, that “we are all textualists now,” would not a wise advocate or scholar seeking to persuade the court advance textualist arguments? Many litigants, including in intellectual property cases before the Supreme Court, have taken the lesson offered by Justice Kagan on board (in my view, to the detriment of intellectual property law), even as Justice Kagan has voiced doubts about her 2015 pronouncement). So should trademark advocates and scholars grasp the nettle of free riding and make arguments for why trademark owners should have narrower (or no) rights because of their free riding (even if they believe, normatively, that free-riding should have no role to play in trademark law)?

Grynberg does carefully note some reasons why free riding might actually have a resonance for judges in trademark cases. For purposes of this Article, Grynberg is non-committal about the wisdom of that choice. There is surely a case for its relevance at the margins. The legislative history of the Lanham Act explicitly talks of protecting against the exploitative conduct of “pirates and cheats”. It takes some methodological insularity to read that reference and conclude that the statute simply targets “those who increase search costs”, as advocates of law and economics would have us believe. That’s a bigger and more contentious debate, however, and one with which Grynberg does not need to engage to make his point. Grynberg simply takes the relevance of free riding to judges as a given. As a result, he offers an approach to trademark law grounded in the reality of judges who decide trademark cases. And those scholars who wish to influence the development of trademark law would be well-advised to heed his counsel.

Without rehashing all of the numerous examples that are teased out with meticulous rigor, Grynberg develops a taxonomy that would helpfully inform the arguments of litigants and scholars in trademark cases seeking to confine the scope of trademark rights. For example, Grynberg notes that trademark owners free ride on culture, using and potentially appropriating memes and other forms of communal communication (something very much at issue in current debates over the use of the “failure to function” ground for rejecting applications for trademark registration). Likewise, he recognizes that what Mark Janis and I have called “surrogate uses” (where the public adopts a source-identifying term not used, and perhaps opposed, by the trademark owner, such as the HOGS example noted above) represent annexation by the trademark owner of understanding developed without an iota of effort or intent by the mark owner.

Moreover, Grynberg suggests that certain trademark owners may even obtain free rider benefits from competitor activity. Where protection is sought for a term that straddles “source” and “category”, or a shape that is a both a configuration mark and an attractive product design, trademark law seeks to ensure that any trademark rights protect the former and not the latter. But Grynberg points out that when we protect a term or shape that possesses such duality (as we do unless the competitive effects are substantial enough to trigger genericness or functionality doctrine), the mark owner may benefit from rival producers seeking to promote a competing product in the same category or of a similar design. That is, the efforts of competitors may consolidate public understanding of the first producer’s mark. Grynberg offers a vocabulary that the courts appear to recognize instinctually – “they didn’t earn it” – which might persuade courts to discount any public understanding not closely tying the product to the first producer (as opposed to the product type).

Whether one agrees with every example (and there are good debates to be had about some), this is a conversation-changing tour de force of an article. Professor Grynberg opts pragmatically to operate in the world we observe – perhaps the most basic commitment of a social scientist – and offers an approach to trademark law that litigants and scholars seeking (thus far without much success in recent years) to limit the scope of trademark rights, would do well to bear in mind.

To Pay or Not to Pay (for Training Generative AI), That is the Question

Dec 18, 2023 Christophe GeigerAdd a Comment

Christophe Geiger

As the first empirical evidence is published on the consequences of Generative AI systems on labor markets¹, broad anxiety is felt from creator communities on the effects of this technology on their income streams. Consequently, the question of how to deal with Generative AI from a copyright law perspective is gaining a lot of attention globally. Several lawsuits have been filed in the US by creators against AI operators and the first attempts to legislate that matter have already been introduced at the national level. The EU is currently finalizing an ambitious regulation package called the “AI Act” with important implications for its copyright regime, in particular the implementation of transparency obligations concerning copyright-protected works used to train the AI algorithms. In this context, Martin Senftleben’s new article Generative AI and Author Remuneration is particularly timely and proposes a very inspiring reflection on what could be the way forward regarding copyright reforms in this field.

One of Senftleben’s main concerns is to find a workable approach not to disincentivize AI innovation while at the same time creating new revenue streams for “flesh and blood authors” to secure remunerations that will improve their working and living conditions. Indeed, the starting point of the author is that:

the increasing sophistication of AI systems will inevitably disrupt the market for human literary and artistic works. Generative AI systems provide literary and artistic outputs much faster and cheaper. It is therefore foreseeable that human authors will be exposed to substitution effects. They may lose income as they are replaced by machines in sectors ranging from journalism and writing to music and visual arts.

In the article, much attention is devoted to analyzing whether the training of the machine learning algorithm with copyright-protected works can be permitted under copyright law. This is a complex question, and it is fair to say that no jurisdiction in the world has a straightforward answer to it, as no copyright law has yet passed having generative AI technology in mind. As machine learning is based on Text and Data mining (TDM), from a copyright point of view, the question often concentrates on whether exceptions and limitations that allow TDM can cover these uses as well.

While the Big Tech AI industry claims this situation falls under the US fair use exception, the content industry considers on the contrary that these uses are covered by the exclusive right and should be licensed. In Europe, a recently introduced TDM exception offers the possibility (under certain circumstances) for right holders to “opt-out” of the exception for text and data mining and to retain full control of their work (article 4 of the Directive for Copyright in the Digital Single Market). Some large collective management organizations have already announced that they will opt out their entire repertoire from TDM activities for machine learning, which will significantly reduce the available training material for AI systems.

As Martin Senftleben rightly underlines, applying this “opt-out”-mechanism to generative AI is not a satisfying solution because it would inhibit the development of this technology and thus make the European Union unattractive for AI developers. For the same reason, he rejects the idea of submitting these uses to the exclusive right: “The need to obtain individual authorizations and manage remuneration payments for AI training constitutes an additional cost factor in the form of transaction costs and licensing fees. If the costs involved are too high, it will negatively impact the ability of the EU’s AI sector to compete on the world market”.

The core of Senftleben’s proposal lies in the argument that copyright law, in order to compensate human authors for the reduction in their market share and income through Generative AI, should introduce an AI levy system in the form of a statutory remuneration and ensure the payment of equitable remuneration to creators. In his model, the statutory remuneration would however not be related to the TDM use of protected works for AI machine learning purposes, but it is “the literary and artistic output of generative AI systems” that serves “as a reference point for a legal obligation to pay remuneration”.

According to the author, focusing on an “output-oriented AI levy system can be applied uniformly to all providers of generative AI systems in the EU. In contrast to a remuneration obligation focusing on the input dimension and AI training activities, the output-oriented levy approach avoids the risk of disadvantages for EU high-tech industries. All providers of generative AI systems are equally exposed to the levy payment obligation the moment they offer their products and services in the EU.” This lump-sum remuneration would have to be paid by AI developers when their systems produce AI-generated output that have the potential to serve as substitute for works made by human authors. To counter legal/doctrinal concerns and to give theoretical support to the proposal, Senftleben refers to the theory of the “domain public payant” (“paying public domain”), according to which the exploitation of the public domain should at least partly serve the living generation of authors.

Admittedly, the proposal put forward by Martin Senftleben is very European in its spirit as its income redistribution rationale might not be an easy fit for all copyright traditions, in particular the US one. From a European point of view, however, the proposal is certainly compatible with a tradition of remunerated exceptions, as there is an established practice and case law about the distribution rules in favor of creators of this kind of remuneration via collective management organizations. In this respect, maybe submitting TDM for machine learning in the context of generative AI to a remunerated exception could be a workable alternative?² Indeed, at the policy level, it will be more difficult to achieve consensus on a proposal based on a paid public domain, as advocates of a robust public domain might be favorable to ameliorate the remuneration situation of creators, but less sympathetic to the idea of a domaine public payant.

In any case, there is no doubt that this important article provides further arguments to consider the position of creators in forthcoming copyright reforms in the field of AI and more generally helps to reflect on how to finance creative ecosystems in a fast-moving technological environment.

1. Xiang Hui, Oren Reshef, & Luofeng Zhou, The Short-Term Effects of Generative Artificial Intelligence on Employment: Evidence from an Online Labor Market, available at SSRN (Aug. 1, 2023).
2. See in this sense for example Christophe Geiger & Vincenzo Iaia, The Forgotten Creator: Towards a Statutory Remuneration Right for Machine Learning of Generative AI, 52 Comp. L. Sec. Rev. (forthcoming, 2024), available at Elsevier (Nov. 24, 2023); Christophe Geiger, Elaborating a Human Rights friendly Copyright Framework for Generative AI, available at SSRN: (Nov. 16, 2023).

All About That Base Rate

Nov 15, 2023 David FagundesAdd a Comment

David Fagundes

Some years ago I attended a presentation by a musicologist who specialized in giving testimony in copyright litigation. Here’s how he tried to grab the audience: First, he would play a clip from a well-known track by a popular musician or band. Then he would play a selection from an earlier, lesser-known track by an obscure musician or band that sounded similar to the first clip, all while giving the audience a wide-eyed stare. The impression this created was intentional and unmistakable. Clearly the well-known artist had copied from the lesser-known one!

The audience, mostly laypeople, certainly bought it, based on the gasps that accompanied the presenter’s schtick. I did not, and left frustrated that the musicologist-turned-expert-witness had tricked the audience into thinking that he had exposed several instances of egregious copyright infringement. I knew something was wrong but had difficulty putting my finger on just what was the problem with the presenter’s move.

Thanks to Christopher Buccafusco and Rebecca Tushnet’s sparkling essay, Of Bass Notes and Base Rates: Avoiding Mistaken Inferences About Copying, I finally have a clear picture of the error that afflicted that presentation and so much copyright litigation. As the authors explain, the application of copyright’s substantial similarity problem suffers from base rate neglect, which causes courts and litigants to significantly overstate the likelihood that a defendant copied from a plaintiff.

Let’s unpack this a bit. Plaintiffs in copyright infringement litigation must make a threshold showing that the defendant copied their work. This usually entails proof that the defendant had access to the plaintiff’s work and that the works share similarities that are probative of copying. Answering the latter question in the plaintiff’s favor depends on their amassing enough evidence to support an inference of copying. At this stage, plaintiffs often retain expert witnesses—usually musicologists,³ like the speaker I mentioned above—who testify that the degree of similarity between the two works is so great and distinctive that it could not be explained by, for example, independent creation or copying from a public domain source.

Here, the authors argue, is where base rate error creeps in. When a musicologist testifies that the quantum of similarity between two works enables an inference of copying, that ignores the crucial issue of how likely such similarity would be absent copying (the base rate). After all, there are only so many appealing chord progressions out there, and the plaintiff presumably created the work herself, so unless we know how likely it is that the similarities at issue would occur in the normal course, it is impossible to say whether the similarities in a given dispute are truly probative of copying.

The authors state that their argument is “modest.” I dissent. Not only did their central insight help me understand what was wrong with the musicologist’s presentation I resented so much (thanks!) but it has the potential to change the way we think and conduct copyright litigation. The past couple of decades have seen an uptick in infringement lawsuits by lesser-known artists alleging that hit tracks by popular artists (e.g., Katy Perry, Led Zeppelin, Marvin Gaye, Ed Sheeran, Taylor Swift, Lana Del Rey, Dua Lipa, and many many more) infringe their earlier, lesser-known works. A central feature of these plaintiffs’ cases is the testimony of musicologists that the degree of similarity between the works in suit is so great that it is explicable only by the defendant’s copying.

But Buccafusco and Tushnet have shown the expert testimony in these cases shares a common, fatal flaw: base rate neglect. Not one of the experts in any of these cases actually knew how likely it was that the given similarity would occur absent copying. This means that experts’ central evidentiary contribution to the litigation (quantum of similarity implies copying) is not a reliable, informed opinion but just highly articulate hand-waving. Many critics decry the purported irrelevance of legal scholarship to law, but this essay represents a conspicuous counterpoint.

The authors limit their discussion of the implications of their insight to the admissibility of expert testimony. Yet it may have other important payoffs. For example, some courts have (controversially) held that where the degree of similarity between two works is extremely high, that enables an inference of copying, regardless of evidence of access. Judges applying this “striking similarity” doctrine often commit base rate errors, assuming that a high degree of similarity itself warrants a conclusion that the defendant copied the plaintiff’s work absent any sense of how likely the similarity would otherwise be (i.e., the base rate).

Another tantalizing question provoked by this essay is whether and how the base rate problem varies within and across genres. Music is especially ripe for this fallacy because the number of notes in the scale is low and the number of appealing combinations of them is lower still, all of which suggests a higher base rate of similarity. But some genres of music are notoriously self-similar, such as ska or reggae, which are defined by a syncopated guitar rhythm. Musical works in these categories are especially likely to have a high base rate of similarity. The same may be true in some literary (formulaic romance or fantasy novels) and artistic (traditional portraiture) contexts that are defined by certain core features. By contrast, the alphabet yields many more available combinations of appealing words and phrases than the musical scale does appealing progressions of notes, so the base rate of similarity in most literary works is likely lower. None of these assertions about relative base rates can be reduced to specific numbers (the authors point out that we cannot actually know the base rate of similarity in any genre) but they each illustrate how the idea of base rates serves as a useful heuristic when thinking about copyright infringement generally.

The authors conclude that because we cannot deduce the base rate of similarity in any instance of claimed infringement, at present the wisest move is to simply bar experts from rendering conclusions about whether similarity supports an inference of copying. And while this does reflect the current state of play, it raises an intriguing possibility that may be realized sooner than one might think. If a large language model can learn from many billions of data points to produce astonishingly plausible simulacra of human responses to prompts, why couldn’t a similar artificial intelligence learn from the millions of available tracks to find the likelihood that certain similarities between musical works occur naturally? Big data might hold the solution to copyright’s base rate neglect problem.

By importing a known but underappreciated idea from quantitative analysis, Christopher Buccafusco and Rebecca Tushnet have generated a simple but significant insight that has the potential to change the way lawyers, judges, and academics think about both the doctrine of copyright litigation and how it unfolds in litigation. And they do it all in under 9000 words. This article is based and I rate it very highly.

1. Judges’ reliance on musicologists to compare musical works in suit to preexisting compositions has grown so prevalent that some authors have compared it to the patent law practice of assessing inventions in light of prior art. See Joseph Fishman & Kristelia Garcia, Authoring Prior Art, 75 Vand. L. Rev. 1159 (2022).

Copyright Fiduciaries: Problems and Solutions

Oct 18, 2023 Jessica SilbeyAdd a Comment

Jessica Silbey

In this forthcoming article, Andrew Gilden and Eva Subotnik begin an important conversation about an underexplored area of copyright law. Their focus is copyright law’s inconsistent treatment of mental capacity. Under copyright law, copyright authors can produce valuable copyrighted work but those same authors may lack the legal capacity to make decisions about if, when, or how to exploit that work. For example, children and people with mental illness or disability can be copyright authors, but they cannot license that work (or refuse to license it) without a legally competent surrogate. The authors explain that this inconsistency leads to injustices for which they offer reforms.

The article starts with the engaging example of the Britney Spears’ 13-year conservatorship, controlled by her father, which from the age of 26 prevented her from making decisions about her life and career. All the while, Spears wrote and performed her songs, building a multimillion dollar portfolio over which she had no control. She was the author of her music, but she had no control over it because she lacked the legal capacity to form binding contracts, or so said a court. She resisted the conservatorship without success for over a decade. The article is full of many other such examples, including of teenage authors, elderly creators, and authors with mental illnesses.

The article argues that this “capacity gap” between authorship and control is a problem from within copyright law as a matter of doctrinal consistency and from a fairness perspective of avoiding exploitation. It further argues there is something unique about copyright law’s capacity gap because, unlike a usual trust situation when assets are transferred to a competent person upon incapacity, in a copyright situation the incapacitated person can produce new wealth while being subject to the control of the trustee or conservator. As the authors say “this dynamic creates unique opportunities and incentives for abuse” and undermines copyright law’s solicitude for authors’ wellbeing.

Both Gilden and Subotnik teach trust and estates, and so they are a great pair to explore the intersection of fiduciary law with intellectual property. Both have focused recent writing on problems that arise for copyright authors after death, in, for example, what Gilden calls the “social media afterlife” and what Subotnik has recently called “dead-hand guidance” in a “preferable testamentary approach for artists.” This new article is about authors still creating, some with cognitive disabilities and others who are simply young. Fiduciaries encourage valuable artistic productivity, for the fiduciaries’ benefit and audiences’, no doubt. But it is at best debatable, according to all the examples the article provides, whether the rate and nature of artistic production is in the best interest of the authors.

The article explains that this conflict between author and fiduciary undercuts the utilitarian justification for U.S. copyright law, which focuses on increasing productivity and the financial incentive of copyright’s exclusive rights. As explained, copyright is available to those lacking legal capacity to be incentivized, which challenges the law’s carrot-stick mechanism for creative production; moreover, the copyright incentive is possibly working on what the article calls the “wrong” people who do not have the authors’ or the audiences’ best interest in mind.  The capacity gap also quite clearly undercuts the personhood theory of copyright law, in which the author’s dignitary interests are paramount and protected through authorial control over the work. By situating the copyright’s capacity gap within several major theoretical justifications for copyright law, the article adds to the rich scholarship on copyright’s evolution from its origins to its contemporary manifestations.

This article has many virtues. It teaches those of us in the intellectual property field a lot about critical features of trust and estate law, which frequently intersect with copyright law, in particular, and with which I’d surmise most of us are fairly unfamiliar. It weaves the two fields in an elegant and clarifying manner and is rich with contemporary examples in which the capacity gap problem arises. These examples would be wonderful classroom discussion topics for those of us teaching in either legal field. The article also engages recent copyright law scholarship concerning alternative IP theories and IP’s failure to protect marginalized authors, advancing these burgeoning fields within IP law. And it provides concrete guidance and reform suggestions to limit the risks of exploitation and abuse that can emerge from copyright’s capacity gap.

The last part of the paper titled “Minding the Gap” will be most interesting to practicing lawyers, administrators, and authors. It harnesses some existing mechanisms within the Copyright Act (such as termination of transfers) and of the Copyright Office (registration and recordation) offering some clear-eyed and reasonable first steps toward addressing problems arising from copyright law’s capacity gap. I would not be surprised to see this article cited and used for reason of its last section alone. Overall, the article is insightful and its topic important. It is a commendable contribution to copyright literature.

Name-Dropping Government Agencies in Advertising

Sep 20, 2023 Alexandra RobertsAdd a Comment

Alexandra Roberts

When shoppers see “Now FDA approved!” on a bottle of Excedrin, does it make them more likely to select that option over a competing product? Will consumers choose a brand of dietary supplement marketed as a “patented blend” over one that doesn’t make patent claims?

In Conjuring the Flag: The Problem of Implied Government Endorsements, Michael Mattioli argues that advertisers use claims about intellectual property and regulatory approvals to mislead consumers about their products’ quality, safety, efficacy, or legitimacy. By reference to or use of the US Patent and Trademark Office or the Food & Drug Administration, advertisers borrow those agencies’ halos to imply that branded products from bugspray to hairspray to nasal spray are superior to competitors’ versions. Mattioli cites survey evidence finding most consumers interpret government stamps of approval like “FDA-approved” and “patented” as endorsements of quality. In Conjuring the Flag, he amasses data on the different ways advertisers reference agency or IP approval to appeal to consumers, highlights why that strategy is misleading, and proposes ways FTC could curb it.

What does it mean when an advertiser claims a product like a can of bugspray is “patented”? Patent lawyers know it only means some aspect of the product or packaging was deemed useful, novel, and non-obvious or some element of its design qualified as original and ornamental. For the bugspray, that might mean an inventor or designer acquired a utility patent that covers the spray mechanism on the dispenser or a design patent for the pattern adorning the can, not that the spray itself is particularly safe, effective, innovative, or non-toxic. Mattioli’s study analyzed hundreds of ads that reference patents, but none of the ads in his data set provided patent numbers or identified what the patent actually covered. Patent-related advertising claims appear most often in ads for products intended to be ingested or applied to the body, such as supplements, toothpastes, and skincare products; consumers may crave reassurance that these types of products have been tested and found safe, so “a significant number of companies are using their patents to cultivate an aura of legitimacy and safety in industries that lack rigorous regulatory oversight.”

Mattioli also reviewed eight hundred ads that tout FDA approval, clearance, or registration—these claims appear most frequently in connection with hair care, cancer treatments, and food and drink containers. But FDA approval means only that the FDA has determined a drug or device’s benefits outweigh its known risks for the intended use, not that the product is safe or effective for other uses or that it lacks side effects or complications. Here, too, advertisers conjure the flag in misleading ways to signal safety or superiority to consumers and sell more products, capitalizing on the likelihood that consumers either don’t know or aren’t thinking deeply about what the claim conveys.

When it comes to trademarks, the misleading advertising mechanism is different. Advertisers sometimes use trademark registration to conjure the flag, as with marks like U.S. HEALTH CLUB for vitamins. But more often advertisers obtain trademark registrations that make their patent- and FDA-related claims look even more official: they register marks like JEANS WITH PATENTED FIT for jeans; PATENTED INGREDIENTS for cosmetics; FDA APPROVED MEDICAL SUPPLIES for devices; and FDA REGISTERED for dietary supplements. More aggressive application of failure to function and deceptiveness doctrines could and probably should be used to screen out marks like these at the USPTO, and the agency could update its manual of examining procedure to include that guidance.

While advertising claims and trademark registrations that reference patents and FDA approvals are sometimes literally false, they are more often merely misleading: “Half-truths and implications are, of course, what conjuring the flag is all about.” To address the problem and better protect consumers, Mattioli proposes FTC launch an initiative focused on monitoring and investigating such ads along with a comprehensive public awareness campaign to teach consumers how to identify and report ad claims that imply government endorsement. Because consumer surveys or expert testimony are typically needed to establish a claim is misleading, he further proposes modifying the legal standard to create a rebuttable presumption that any ad referring to a patent, FDA approval, or other government action in a way that implies approval of a product is potentially misleading, shifting the burden of proof to the advertiser to demonstrate the ad is not misleading.

While materiality is a fact-specific inquiry, it is well within FTC’s power to conduct a broader survey or study to gauge whether ads that proclaim products or services “FDA approved” or “patented” mislead consumers in ways that impact their purchasing decisions. In fact, FTC’s Advertising Division has performed surveys and other research on advertising claims like “organic,” “recycled content,” “results not typical,” and “made in the USA” to undergird its guidance and policy statements on those categories of representations. This intervention is crucial because while it’s possible these representations are misleading and constitute unfair competition, it’s also possible consumers skate right past them or regard them as mere puffery. Mattioli’s intuition that these types of claims are misleading is compelling and likely correct, but further study would give FTC a solid basis to pursue claims that conjure the flag more aggressively and systematically.

Gender Equity in the Market for Collegiate Name, Image, and Likeness Rights

Aug 8, 2023 Michael W. Carroll1 comment

Michael W. Carroll

Within the field of intellectual property law, there are not too many legal or economic developments that would qualify for an event study. But, on July 1, 2021, such an event occurred when a new rule issued by the National Collegiate Athletic Association (NCAA) took effect. Prior to that date, intercollegiate athletes were prohibited from exercising their right of publicity or any other rights in their name, image, or likeness (NIL) to endorse products, services, or businesses in a commercial manner. Under the rule change, these athletes, numbering nearly 500,000 at the time, suddenly became free to license or otherwise use their NIL rights commercially, and a new market was suddenly born.

In The NIL Glass Ceiling, Professor Boston explains how the market for intercollegiate NIL rights has quickly evolved in a way that provides these athletes with long-denied revenue but with disparate outcomes for athletes who identify as men or women.⁴ She argues that these disparities are problematic both because female athletes should be entitled to a greater share of the revenue in this market and because these disparities send an unwelcome message to female athletes about the state of gender equity in intercollegiate athletics and in the workplace. She argues that more gender-equal outcomes could arise if schools were subject to scrutiny under Title IX, applicable Department of Education regulations under Title IX, and NCAA rules that govern certain third-party support for intercollegiate athletic programs in the case of disparities in NIL revenues paid by certain third parties directly to athletes.

This article makes three main contributions to the literature. First, it explains that the NIL market is comprised of three main forms of agreement or licensing. Athletes are paid either from deals they self-arrange, from deals they make with their schools, or from those made with third-party Collectives that pool assets to finance NIL licensing. The second contribution is to focus on the outsize role of the Collectives in this market and to argue that the relationships between these and the schools whose athletes they support are sufficiently close to subject payments made by these Collectives to scrutiny under the school’s legal duties, under applicable law, to not discriminate on the basis of gender. The third contribution is to argue that the important role that this new NIL market is playing in intercollegiate athletics requires legal reform to account for this development. Professor Boston discusses a range of options for how such reform might be achieved.

One feature of Professor Boston’s description of this market I found interesting is that even though the right of publicity is not recognized in every state, as she notes, and only some athletes have made uses in commerce of their name, image, or likeness in a way that would qualify for trademark protection, the market for NIL licenses appears to disregard these differences and treat all athletes as being able to supply consideration by granting NIL permissions in exchange for payment. I have no doubt that a court would treat such permission as sufficient consideration under contract law, but it would not be surprising if this new market were to lead to judicial or legislative recognition of a right of publicity where none exists today.

Professor Boston explains that in this market, some athletes represent themselves, or work with an agent, to negotiate NIL endorsement deals. Some sponsors, however, seek NIL arrangements with an entire team. In these situations, the sponsor may seek an alternative to individual negotiations. One alternative is for schools to facilitate NIL transactions on behalf of their athletes. Some states permit this practice, while others prohibit it. Even when a school is permitted to provide this facilitation, the NCAA’s rules prohibit such facilitation for prospective student-athletes.

Most schools do not provide such facilitation, which leaves space for third parties to play a role, which they quickly have moved to do. Professor Boston divides these into third parties that provide technology platforms for NIL licensing, talent agents, and other advisors who provide support for NIL transactions, and traditional Boosters, supplemented by a new entity, the Collective.

There are more than 100 of these Collectives, which use different mechanisms, such as crowdfunding or membership tiers, to create an asset pool that they use to pay athletes for their NIL endorsements. To date, these Collectives limit their support to individual schools. These Collectives focus their resources on schools’ football and men’s basketball teams, which helps fuel the gender disparities in NIL revenues.

Professor Boston explains that advocates for athletes’ ability to monetize their NIL rights thought that such opportunities would play a potentially equalizing role between men’s and women’s sports. That is not how it has turned out so far. Overall, male athletes receive three times as much as female athletes in all three of NCAA’s divisions. She shows that because Collectives provide a significant share of NIL revenues, the gender disparities in their practices substantially contribute to overall disparities in NIL revenues. That raises the question this article addresses: should schools be held legally responsible under Title IX for the gender disparities attributable to Collectives’ NIL practices?

Professor Boston provides a very nice, succinct explanation of applicable law. Under applicable regulations interpreting Title IX, schools’ compliance obligations fall within three categories: (1) athletic scholarships, (2) benefits and services, and (3) effective accommodation of students’ interests and abilities. Within the “benefits and services” category, equal treatment obligations extend to a school’s recruiting and publicity activities.

Professor Boston argues persuasively that even though schools are prohibited from explicitly using potential NIL revenue as a recruiting tool, they do, and they do so in a gender-disparate manner. Similarly, publicity enhances athletes’ NIL licensing opportunities, and women’s sports on average receive less publicity than men’s. Professor Boston shows, for example, how when all women’s NCAA basketball tournaments games became televised, players’ NIL opportunities increased measurably.

She further demonstrates how schools have been, and can be, held responsible if athletes receive gender-disparate benefits and services from third-party sources. She argues that the relationship between Collectives and schools and the gender disparities arising from Collectives’ focus on men’s NIL licensing are actionable under Title IX.

Professor Boston offers regulators and Congress a menu of options to update Title IX from the easiest, most feasible options, to the most difficult but most desirable outcome, which would be legislative amendments to Title IX.

I learned a lot from this article. Like Professor Boston, I recognize some of the challenges involved in holding schools responsible for NIL licensing done between third parties and athletes. But, I agree with her that schools are not in a fully arms-length posture with these Collectives. I found most persuasive her argument about how potential NIL revenue is playing a significant role in recruiting in at least some sports. Since recruiting is an equal treatment factor, and gender-disparate potential NIL revenues fueled by Collectives’ practices have gender-disparate impacts on recruiting practices, that seems like a Title IX problem in need of a solution.

1. Recognizing that not all athletes so identify, for purposes of discussing NIL revenues from intercollegiate sports, Professor Boston’s article and this review are constrained by existing NCAA rules, under which athletes must select from its binary sports classifications, and must meet its requirements for those classifications.

Generative AI Meets Copyright

Jul 11, 2023 Pamela SamuelsonAdd a Comment

Pamela Samuelson

ChatGPT, Midjourney, and Copilot are among the numerous generative AI systems launched in the last year or so. They have attracted a huge number of users as well as several lawsuits. Among the lawsuits’ claims are that the makers of these systems are direct and indirect infringers of copyright because of their use of millions of in-copyright works as training data and because outputs of these generative AI programs are infringing derivative works.

At the core of these AI systems are foundation models on which the authors focus in their fascinating new article. They define this term as “large pre-trained machine learning models that are used as a starting point for various computational tasks,” including generative AI systems that may produce text, images, and/or software code in response to user prompts. The article identifies various actors who contribute to elements of these AI systems, including data creators, data curators, model creators, model deployers, and model users.

Those of us who are intent on understanding the legal implications of generative AI systems must, of necessity, be prepared to learn about the technology underlying these systems. Fortunately, these six Stanford researchers—some in computer science and some in law (our own redoubtable Mark Lemley among them)—have provided an essential guide for intellectual property and technology law scholars to the development and deployment of these systems. The article explores the extent to which developers and deployers of generative AI systems may rely on fair use to justify their use of in-copyright works as training data and how developers may limit their potential liability for infringements at the output stage.

For many copyright scholars, the article’s discussion of the fair use cases will be familiar, but the application of these precedents in the context of generative AI will be particularly useful. Yes, of course, the Authors Guild v. Google and iParadigms decisions suggest that computational uses of in-copyright materials can be fair use, but other decisions such as Associated Press v. Meltwater and Fox v. TVEyes suggest that much will depend on the particular uses that generative AI systems make of the in-copyright materials.

Foundation Models is not an advocacy article asserting that all uses of in-copyright works (or at least all that can be found on the open internet) as training data is fair use. Nor does the article argue that all outputs should be non-infringing so long as the outputs are not verbatim copies of the contents of specific training data. It offers a much more nuanced perspective about the challenges for system developers in understanding how to model computationally the degree of “transformativeness” that may be achieved by a second comer’s use of copyrighted works, as well as how to distinguish facts and expressions within those works.

The most novel section of Foundation Models is its discussion of technical strategies that AI system developers can employ to reduce the risk of copyright infringement when generative AI produces outputs in response to user prompts. These include data and output filters to detect similarities between the input data and outputs generated by the systems. Some technical mitigation strategies the authors describe must be done at the training data stage, while others, including data and output filters, can be done at the deployment stage.

The article discusses the Field v. Google decision for its recognition that Field had not used the “robots.txt” exclusion standard as a technique to stop Google from webcrawling his site. This consideration weighed against Field’s copyright claim that the search engine infringed by copying his content on that site. Foundation Models suggests that generative AI system developers and deployers would be well-advised to adopt one or more technical mitigation strategies to bolster their fair use claims.

While this article is well worth reading on the merits, it is also a noteworthy contribution to an emerging literature in which computer scientists and lawyers collaborate to explore technology law and policy issues. While not written in perhaps the most scintillating prose, this article is an outstanding example of a successful collaboration to explore ways in which technologists and lawyers can work together to co-evolve practical ways to achieve socially desirable outcomes.

Can Neuroscience Fix Trademark Surveys?

Jun 7, 2023 Christopher J. BuccafuscoAdd a Comment

Christopher J. Buccafusco

In trademark litigation, consumer surveys can determine a number of important doctrinal questions, including the most important one: whether consumers are likely to be confused into thinking that the defendant’s product was made or licensed by the plaintiff. Recently, scholars have questioned the validity and reliability of standard trademark surveys, suggesting that they are easy to manipulate and biased in favor of one party or the other. Wouldn’t it be great, then, if there was a reliable way to determine whether a survey was biased or not? Using neuroscientific imaging, an interdisciplinary group of researchers (including law professor, Mark Bartholomew) has proposed just such a possibility in a new paper, From Scanner to Court: A Neuroscientifically Informed “Reasonable Person” Test of Trademark Infringement.

Trademark surveys can suffer from a number of flaws. They may be explicitly biased in favor of one party or another, for example, by describing the defendant as a “copycat” or the plaintiff as a “trademark bully.” They may exhibit more subtle biases in how they frame questions about similarity and confusion. And, finally, survey participants always know the nature of the survey they are taking, so participants may exhibit “demand effects,” providing what they anticipate are the surveyor’s desired answers rather than their true responses.

The researchers began by demonstrating the manipulability of survey instruments. With two different plaintiff products, they manipulated the survey language to create both “pro-plaintiff” and “pro-defendant” surveys, as well as a putatively “neutral” survey. For each plaintiff product, they included a list of other products that varied in their degree of apparent similarity. For example, Reese’s Peanut Butter Cups served as the plaintiff product in one group, and the other products included Toffee Crisp (an actual defendant in litigation brought by Reese’s) as well as less similar products like Snickers, Justin’s, and Ghirardelli. When they tested a group of participants recruited from Amazon Mechanical Turk, the biases had the expected effect. Participants thought the defendant’s product (Toffee Crisp) was more similar to the plaintiff’s (Reese’s) in the “pro-plaintiff” survey than in the “pro-defendant” survey, and the “neutral” survey produced intermediate results.

To generate a more objective measure of product similarity that does not rely on participant reports, the researchers exploited an intriguing feature of human perception and cognition known as “repetition suppression.” The idea is simple: when we are presented with a stimulus that is very similar to one that we have just seen, our perceptual response to it diminishes. This is effectively a visual heuristic. Having seen something once, our brains devote less cognitive capacity to seeing it the second time. The empirical strategy, then, compares the degree of neural diminution across various stimuli to objectively measure stimuli similarity. The more similar two stimuli are, the more participants’ neural responses to the second stimulus will be diminished.

The study entailed functional magnetic resonance imaging (fMRI) scans of 26 participants who viewed the various products described above. fMRI scanning measures relative changes in brain blood levels as a proxy for neural activity. By focusing on regions of the brain known to process visual stimuli, the researchers could measure the degree of diminished neural activity associated with the Reese’s-Toffee Crisp pair compared to the Reese’s-Snickers pair.

When the researchers compared the neural similarity measures detected by fMRI to the self-reported similarity measures from the prior surveys, they found a strong correlation between the fMRI data and the “neutral” survey but no significant correlations between the fMRI data and either the “pro-plaintiff” or “pro-defendant” surveys. This suggests that the “neutral” survey is, in fact, a good proxy for participants’ actual experiences of visual similarity.

To be clear, the researchers’ methodological contribution isn’t to suggest that all trademark cases should require incredibly expensive neuroscientific studies. Rather, by using techniques like this one, scholars can develop a set of “best practices” or “gold standards” for trademark survey research. The goal is to use neuroscience to validate much cheaper behavioral surveys.

Of course, this is just the beginning. There is much that the current study doesn’t tell us. It can tell us that participants think the overall visual impression of certain trade dress is more or less similar to other trade dress. But it cannot tell us, for example, whether the participants were paying attention only to the protectible aspects of the trade dress or not. Nor do we know if some degree of similarity is consistent with consumers being confused as to source. The Mercedes logo and a “peace” sign look very similar, but people may not be confused by them. But this study seems a step in the right direction, and I’m excited to see where this kind of research will go next.

The Complex Effects of Piracy on the Movie Industry

May 9, 2023 Christopher J. SprigmanAdd a Comment

Christopher J. Sprigman

Does piracy of creative goods such as movies, books, or songs reduce paid demand for those goods? This seemingly straightforward question has proven surprisingly difficult to answer in the real world.

Piracy may draw away customers who might otherwise have paid. But it’s also possible that consumers of pirated copies are, by and large, not people who would have paid to consume if they couldn’t get access for free. Piracy may also help spread the word about a good movie, book, or song. This sort of informal advertising might drive up paid consumption, even if some people who would have paid are lost to piracy. It’s also possible that some combination of all these things might happen, with uncertain net results.

In a new empirical paper, titled Avengers Assemble! When Digital Piracy Increases Box Office Demand, Klaus Ackermann, Wendy A. Bradley, and Jack Francis Cameron offer a nuanced and interesting study of the effects of piracy on the movie industry. The effects of piracy, as it turns out, have a lot to do with what kind of movies we’re talking about.

The authors built a novel dataset that identified the existence and the timing of the earliest upload of a high-quality pirated copy for every U.S. movie release. The authors did this with data on the appearance of movie piracy “torrents” between January 2004 and January 2020 from online piracy site The Pirate Bay (TPB). The authors then matched this data with movie release information during the same period from the well-known IMDb database.

Merging these two streams of data allowed the authors to match up release dates and “piracy dates.” They measured changes in box-office revenue for pirated movies in the first 48 days after their releases in theaters, relative to the preceding period following releases before the movies were pirated. They then used a formula to adjust for the general fall-off in movie box-office revenues over time. If piracy was substituting for paid demand, the authors could pick up that effect by comparing (time-adjusted) pre-piracy vs. post-piracy box office revenues across many films.

The authors hypothesized that piracy has different effects on different types of movies. Specifically, they theorized that “spectacle” movies—the kind of movies that people want to see in the movie theater—may be less affected by piracy than “story” movies that people are more content to watch on their computers. In other words, spectacle movies may benefit more from word-of-mouth advertising that piracy may provide while losing fewer customers to demand substitution, compared to story movies.

To aid this assessment, the authors constructed two measures of movie “spectacleness.” One used a movie’s release in 3D or IMAX formats as a proxy for that quality (because “spectacle” movies are the kinds of movies that people want to see in these especially immersive formats). A second categorized movies into genres associated with spectacleness and story-focus by measuring the number of movies in various genres nominated for the “best visual effects” Oscar (associated with spectacleness) as opposed to the “best original screenplay” Oscar (associated with story-focus).

Based on data for more than 400 movies, about half of which have been pirated within the first 48 days of release, the authors concluded that piracy had the mixed effects they predicted. For films for which in-theater viewing adds value (“spectacle” films), there is a 13% increase in average daily box office revenue after the appearance of a high-quality pirated version of the film online. For story-focused films, on the other hand, there is as much as a 30% decline in average daily box-office returns after the appearance of a high-quality pirated version of the film online. This is consistent with the idea that piracy acts as a substitute to films focused on story, where the full value of the film can be consumed at home.

The authors’ findings shouldn’t be too surprising. Think for a moment about the music industry. Recorded music is more vulnerable to piracy than live music because a big part of the appeal of live music is the immediacy and communal experience of the concert. Such experiences cannot be replicated in a pirated recording.

So we might expect that during the post-Napster but pre-Spotify/Apple Music era when online piracy was driving down revenues for recorded music, there would be an industry shift toward more focus on live music. There was indeed a very rapid growth in that period of big live music firms such as Ticketmaster and Live Nation. Moreover, during that period the rise of live music revenue very closely mirrored the decline of recorded music.

In 2000 (just after Napster’s debut), recorded music represented 53% of the global music industry. By 2017 (when paid streaming started to restore lost recorded music revenues), recorded music’s share of total music industry revenues had dropped to 38%, while live music went from 33% to 43% of the industry.

Something analogous is happening in the motion picture industry, although the effect is probably not as pronounced. That is, the industry’s product mix may have shifted toward “spectacle” films because these sorts of film tend to be more resistant to piracy. Indeed, the authors gesture in this direction, stating that their findings suggest because the value of a film is linked to its “spectacleness,” the industry would be wise to adjust its creative output on the margin—i.e., to produce more spectacle films—to blunt piracy’s effect rather than investing in the law enforcement efforts that would be required to reduce piracy by any substantial amount.

Alternatively, movie studios may seek to insulate story films against piracy by, for example, releasing them to streaming channels simultaneously with theatrical release. Or, maybe movie studios could invest directly in upgrading theaters for these story-focused films to enhance the in-theater viewing experience in other ways, such as by making the theater a place for fun and social interaction. (Theaters such as the Alamo Drafthouse are already offering this kind of experience).

If so, then the principal effect of movie piracy may not be to lower the overall demand for movies or the number of movies produced. It may be to shift the kind of movies produced, or, more subtly, to shift the way that movies are presented to the public. Unlike the relatively simple framework in which piracy leads to fewer movies, the real effect of piracy may be more subtle, and the case for investing significant resources (especially public resources) in anti-piracy efforts less clear.

Policy Experimentation to Address Inequality Among Innovators

Apr 4, 2023 Lisa Larrimore OuelletteAdd a Comment

Lisa Larrimore Ouellette

Inequality among innovators is a substantial social problem in terms of both equity and economic growth. For instance, Raj Chetty’s Opportunity Insights group has documented that if women, racial minorities, and low-income Americans invented at the same rate as high-income white men, then the rate of U.S. patenting would quadruple. They also note the glacial progress toward closing these gaps, such as the 118 years it will take to reach gender parity at the current rate.

These inequalities affect not only the rate of innovation, but also what kind of innovations are created—for example, all-female inventor teams are more likely to focus on women’s health. Unfortunately, the evidence base for policy interventions to reduce these innovation gaps remains depressingly shallow. Most policies are tested without a rigorous evaluation strategy or control group, making it difficult to determine whether they had any effect.

A new paper from the U.S. Patent and Trademark Office (USPTO), Closing the Gender Gap in Patenting: Evidence from a Randomized Control Trial at the USPTO, is a remarkable addition to this literature. For the first time ever, the USPTO has tested a policy intervention as a randomized experiment, allowing a credible evaluation of its effectiveness. Changes in patent policy have rarely been tested with any element of randomization and have never been tested previously by the USPTO itself. Even if this experiment had yielded null results, the effort would still have been laudable as a model for how agencies can assess the impact of a new policy and publicly disclose the results. But the experiment also documents that the intervention—a new program to help patent applicants without legal representation—led to a sizeable decrease in the gender patenting gap.

The USPTO’s experiment began in 2014, when it created a new “Pro Se Pilot Examination Unit” to help pro se inventors (those without professional assistance) through the patent examination process. Obtaining a patent is not a user-friendly process, with most patent applications receiving a “rejection” or even a “final rejection” (which is actually more akin to a “revise and resubmit”) before eventually being allowed.

One study suggests that around half of the patent gender gap is due to women being more likely to abandon their patent applications after these discouraging replies rather than persisting in this back-and-forth process with the patent office. To address a concern that pro se inventors may be particularly disadvantaged in this process—for reasons unrelated to the merits of their inventions—patent examiners in the Pro Se Pilot received training on strategies to assist these inventors. For example, examiners would encourage applicants to call them with questions and would proactively help applicants draft better patent claims.

In the same way that promising new medicines are rigorously tested in randomized controlled trials that assign patients to either the new treatment or a control group, the USPTO decided to test this new examination unit by randomly assigning pro se applicants to either the Pro Se Pilot or to the regular examination process. By comparing outcomes across the two groups, they found that the Pro Se Pilot increased the likelihood of receiving a patent for all pro se applicants, and that it had a particularly striking effect for women. The likelihood of receiving a patent increased by 6.1 percentage points for men compared with 16.8 percentage points for women. The gender effect was even larger among first-time U.S. applicants: the likelihood of receiving a patent increased by 5.8 percentage points for men and a remarkable 23.5 percentage points for women. These results provide strong causal evidence of the new program’s value in closing the patent gender gap for pro se applicants.

Of course, this intervention is only one small step toward addressing the innovation gender gap more broadly. Future research should investigate whether similar changes in examiner training could help reduce the patent gender gap for broader groups of applicants. Less than 1% of U.S. patent applicants are pro se, but the additional guidance provided through the Pro Se Pilot might also help a larger group of inventors, such as those at small and micro entities who are currently disadvantaged by lower-quality legal representation.

In addition, the USPTO should study whether the reduced gender gap persists beyond patenting. Receiving a patent is worth little in isolation; financially benefiting from patents depends on other institutions with their own gender biases, such as corporate rent-sharing and venture capital. The Pro Se Pilot increased the likelihood that a pro se applicant would receive a patent, but it is worth examining longer-term outcomes such as assignments of these patents, new patent applications from these inventors, and non-patent outcomes gathered by survey or by linking to other datasets.

But the need for further research should not detract from the monumental nature of this study, which has simultaneously tackled two problems of bipartisan interest: inequality among innovators, and the need for better evidence to improve government effectiveness. In 2018, President Trump signed the SUCCESS Act of 2018, which tasked the USPTO with studying and recommending solutions to the problem of inequality among innovators. And the USPTO’s current Learning Agenda—developed pursuant to the Evidence Act of 2018—commits the agency to develop evidence on how to improve the effectiveness of patent examination in general, and with assessing participation in the patent system by underserved populations. The success of the first randomized controlled trial run by the USPTO on both of these fronts will hopefully inspire the use of rigorous experiments to test other policy interventions, both within and outside the patent context.