Generative Fair Use for All

Jul 28, 2026 Christopher J. BuccafuscoAdd a Comment

Christopher J. Buccafusco

Artificial intelligence platforms justify their unauthorized reproduction of copyrighted training data on the expansive creative capacities that their models allow. They argue that generative AI is a “radically transformational tool for creators of many kinds … which enables new expression and innovation to flourish.” In her new paper, Engineering Semiotic Democracy, Katrina Geddes sets out to hold AI firms’ feet to the fire.

To motivate the intervention, Geddes reports asking ChatGPT to produce an image of Captain America as a queer Black man from Brooklyn. The platform refused the request as a violation of its content policy that prohibits generating certain copyrighted characters. Because the output that her request would produce is likely protected as a transformational fair use, Geddes worries that risk-averse AI firms are blocking the creation of content in a manner that is more restrictive than copyright law requires. Doing so, she argues, undermines people’s capacities to engage and create with AI in precisely the sorts of radically generative ways the platforms tout in their litigation.

For Geddes, generative AI likely has an important role to play in expanding the diversity of creators and creations by lowering the costs of making new works and increasing the capacities of people who want to make them. While AI poses a risk of de-skilling some creators, Geddes also recognizes its potential to up-skill others. Many people who would never contemplate creating visual art if they had to draw or paint will be enabled by AI to express their ideas. This diversity of new voices and new creations represents the semiotic democracy in Geddes’s title.

Of course, some of these new works will incorporate copyrighted works in ways that content owners find objectionable. And content owners will point to them as evidence that generative AI is harming the market for their copyrighted works. To avoid these risks, AI platforms have begun restricting the prompts that they will accept and the outputs that they will generate. It is now much harder to generate copyrighted characters and other protected content, even if the resulting output isn’t unlawful.

Geddes’s intervention is important. One of the fundamentally transformative features of AI is its capacity to encourage broad swaths of cultural production. But if AI users can’t generate the equivalent of 2 Live Crew’s “Pretty Woman” or the Air Pirates version of Mickey Mouse, the value of AI as a creative catalyst is meaningfully less than we otherwise might have expected. Calls to restrict model memorization risk burdening otherwise protected—and important—speech.

Ben Sobel warned us about these issues in his recent Copyright Accelerationism paper (that I JOT’d here). AI firms will claim the benefits that fair use provides, allowing them to use others’ content for free, without extending the same benefits to their users. Read together, these papers should give copyright scholars pause about the realities of generative fair use for all.

Geddes’s concerns are ones we’ve seen before, as she readily acknowledges. Tech firms have long blocked user-generated content that is likely protected by fair use. Studies on copyright take-down requests have found that platforms like YouTube excessively block legal content. Geddes also acknowledges that some platform limitations on generation will arise, not from copyright risk, but from the platforms’ own prohibitions on certain kinds of outputs, including sexually explicit ones. If ChatGPT won’t create a queer Captain America, not because of the copyright but because it blocks prompts relating to sexual material, then copyright isn’t the problem.

In a forthcoming article in the Duke Law Journal, Matt Sag argues that the emerging arrangement of licensing deals between AI platforms and content providers may be the best option open to society. Sure, some stuff will be blocked, but the alternative of massive copyright liability is worse. Sag may turn out to be correct, but Geddes’s article helpfully clarifies what we stand to lose from this scenario. In the coming years, courts, scholars and policymakers must grapple with the tradeoffs that content licensing and output blocking create.

Copyright’s Constitutional Crisis

Jun 26, 2026 Pamela SamuelsonAdd a Comment

Pamela Samuelson

In May 2025 the Trump Administration summarily fired Librarian of Congress Carla Hayden and Register of Copyrights Shira Perlmutter and announced that Deputy Attorney General Todd Blanche would become the Acting Librarian and Associate Deputy Attorney General Paul Perkins the Acting Register.

The Library and the Copyright Office did not contest Trump’s ability to fire the Librarian, but they successfully blocked Blanche and Perkins from assuming the acting roles. Perlmutter has sought a declaratory judgment that the Administration lacks authority to fire her as the Register and asserts that she is still the Register. The D.C. Circuit Court of Appeals ruled in Perlmutter’s favor, but the Supreme Court will review that ruling. (Blake Reid explores the issues raised in Blanche v. Perlmutter in his Separation of Copyright Powers article, which is also slated for publication in the Seattle U.L. Rev. Online).

Street and Runyon are concerned primarily with preserving the independence of the Library of Congress from executive branch interference. They explain very well the important role the Library has played not only as a substantial resource for members of Congress, but also as the country’s national library, with more than 178 million items in its collection. The Library adds more than 10,000 items to its collection every working day, largely by virtue of the mandatory deposit copies of works of authorship it obtains from the Copyright Office when copyright owners provide deposit copies when they register their claims of copyright.

The Library and the Office have long been regarded as units of the Congressional branch of the U.S. government (the Office is a subunit of the Library), even though both the Librarian and the Register engage in some executive functions. Under the unitary executive theory, Trump claims that he is entitled to fire any government official who performs executive functions. Trump also claims that he has authority to appoint Blanche and Perkins under the Federal Vacancies Reform Act, which allows the President to appoint acting officials to executive agency positions for a temporary period. Because the Register performs some executive functions and the Librarian supervises the Register (an executive function), the Trump Administration argues the Library and Office are executive agencies, and hence, the acting appointments of Blanche and Perkins are sound.

What, you might ask, does this have to do with copyright? Two issues stand out: First and foremost, Street and Runyon argue that the Copyright Office must be severed from the Library so that the Library’s status within the Congressional branch, independent of the executive branch, is made more secure. Second, they argue that the mandatory deposit rules must be retained and strengthened, so that the Library’s broad collection of literature and cultural heritage works can continue to grow apace, although it is unclear how the deposits now made to the Office will get to the Library’s collection.

Where, though, is the Copyright Office supposed to go? Street and Runyon suggest that the Office could become a subunit of the Patent and Trademark Office (PTO), which has been proposed in the past, or an independent agency. Although Street and Runyon do not seem to care about the substantive copyright issues raised by their proposal, many of us in the copyright field may have views on this subject. I, for one, would just as soon that the Copyright Office remain as a subunit of the Library consistent with copyright’s constitutional purpose of promoting the progress of knowledge and culture. But is this realistically an option in the current era? (Reid’s article also discusses the placement of the Office if removed from the Library).

Street and Runyon also discuss at length the history and significance of the mandatory deposit rules in 17 U.S.C. §§407 and 408. Mandatory deposit rules have been part of U.S. copyright law since 1790. Section 407 requires that copyright owners deposit two copies of their works with the Library within three months of publication. Yet, when the Copyright Office tried to enforce that requirement as to books whose copyright owner made them available only in digital form, the publisher challenged the Office’s insistence on physical copy deposits as a taking of private property in violation of the Fifth Amendment to the Constitution and won in Valancourt Books, LLC v. Garland, 82 F.4th 1222 (D.C. Cir. 2023). While the D.C. Circuit’s decision was limited to the facts of Valancourt, Street and Runyon worry that publishers may be emboldened to challenge mandatory deposit rules more generally.

The Valancourt decision is also of concern because in the digital age, a great many copyrighted works are now available only in digital formats. So, unless the mandatory deposit rules are modified so that the Library and Office can accommodate digital copies, the Library’s collection will cease to include the broadest array of knowledge. Street and Runyon propose ways to shore up the mandatory deposit rules (with details I will not share here).

Library at a Crossroads is an excellent and passionately argued article that richly deserves to be read. It considers some important issues for the future of knowledge that most of us copyright professionals have long ignored.

What Should IP Law Do (if Anything) About Cultural Appropriation?

May 27, 2026 Christopher J. Sprigman1 comment

Christopher J. Sprigman

For several decades now, a debate about whether or how to regulate cultural appropriation, especially of indigenous creations, has been brewing at the edges of American IP scholarship. This topic has, however, never really broken through—that is, it has never surfaced as an issue that captures the attention of the field in the U.S. as a whole. It is heartening to read recent contributions to the literature from scholars including Margo Bagley, Sonia Katyal and Angela Riley, Jessica Kisser, Ruth Okediji, Michael Goodyear, Trevor Reed, and Aman Gebru who recommend some form (often narrow) of protection against cultural appropriation and/or the unauthorized use of traditional knowledge. Another smaller recent legal literature raises questions about such protections (including noting weaknesses in the normative case for legal rules to restrict cultural appropriation).

Of these many worthy contributions, I especially want to praise Aman Gebru’s article Remediating Cultural Appropriation. Gebru provides three invaluable services to the literature on cultural appropriation. The first is a careful articulation of the various possible harms and benefits of cultural appropriation. This analysis is fair-minded and inclusive, reviewing and critically assessing the literature that fills out both sides of the ledger. Second, Gebru proposes a taxonomy to assess varieties of cultural appropriation claims with particular focus on two factors: the cultural symbol’s level of “diffusion” (i.e., is the element shared by other cultures, or is it strongly identified with a single originating culture?) and the extent to which the use is commercial in nature.

Of the four categories of cultural appropriation that Gebru proposes, he advocates legal intervention to limit appropriation as to only one, namely, commercial uses of cultural elements that are distinctive of a single culture. For that category, Gebru favors an approach based on trademark concepts although he proposes limits analogous to laws such as the U.S. Indian Arts and Crafts Act, which is, essentially, a truth-in-labeling requirement. For the other categories, the cost of even narrow cultural appropriation limits, Gebru argues, are too high in terms of free speech harms and the violence that would have to be done to numerous legal doctrines, such as standing rules, to permit judicial enforcement. In a brief but valuable part of the article, Gebru details a number of lawsuits that are, at their core, about cultural appropriation, but which avoid making cultural appropriation claims directly because they simply do not fit under existing law.

Third and finally, Gebru supports his recommendation for focused legal intervention with a theory of “cultural passing off.” He proposes a framework for enabling legal claims against commercial use of a distinct symbol that falsely implies affiliation with an identifiable community. Gebru draws on the trademark passing off cause of action and adapts it to the cultural appropriation context, laying out elements of collective goodwill, commercial appropriation of distinct cultural symbols, and deprivation of material advantage as the claim’s core elements. This framework, Gebru argues, addresses harmful cultural appropriation while preserving the beneficial dynamics of cultural diffusion.

In a final part of the article, Gebru illustrates the application of his framework with some examples. One that stands out is Jeep’s use of the term “Cherokee” for one of its SUVs. Gebru argues that all elements of the cultural passing off claim are satisfied in this case. The “Cherokee” name enjoys “collective goodwill”—i.e., its value inheres in large part from its association with a distinct (and geographically and culturally cohesive) social group. The use also, Gebru argues, involves misrepresentation, in that there is no consent, affiliation or other connection with the group. And finally, there is deprivation of material economic advantage, arising from the lost licensing opportunity.

Gebru proposes that a cultural group pressing a “cultural passing off” cause of action must show, among other elements, that the appropriation has caused it to suffer a material “deprivation of economic advantage.” But establishing that element will often, if not always, involve some circularity—the assertion often involves a foregone licensing opportunity, but the licensing opportunity exists only if there is a right to license, and that is the very thing at issue in the claim. This is an important element of my skepticism that the framework is either workable or normatively justifiable.

A different critique can be aimed at the misrepresentation requirement. Gebru frames it as an absence of consent, affiliation, or other connection with the group whose cultural identifier is used. But that does not track with passing off law, which traditionally has thought of misrepresentation as creation of the impression that there is such a connection.

Despite my reservations about this proposal, I admire Gebru’s article. It is the most helpful summation of a pro-intervention position that I’ve seen, even if the intervention it recommends is narrow. And importantly, Gebru is careful to deal with opposing arguments, and to be generous to them. And this, at bottom, is perhaps the most important of Gebru’s contributions. He has given us a model for engagement on a topic that, over the years, has failed to gel into the fascinating and productive debate that we could be enjoying within the community of IP scholars.

Clinical Trial Transparency and Patent Prior Art

Apr 30, 2026 Lisa Larrimore OuelletteAdd a Comment

Lisa Larrimore Ouellette

The push for greater clinical trial transparency is typically framed as a conflict between companies’ interest in protecting trade secrets and the public’s interest in accessing health-relevant information. But a new article by economists Dennis Byrski and Lucy Xiaolu Wang analyzes an underappreciated dimension of this debate: clinical trial disclosure also affects patenting. When trial data becomes public, it creates prior art that can block subsequent patents related to the drug. The authors document that European marketing authorization—which requires substantial data disclosure—leads to significant declines in secondary patenting on drug modifications like new dosages and formulations. The result is a regulatory check on pharmaceutical patent “evergreening”: once authorization-related disclosures enter the public domain, many incremental follow-on patents become harder to obtain.

Timing is critical in patent law. To obtain a patent in the United States, the European Patent Office, or other jurisdictions, an applicant must show that their invention is new and nonobvious (or in Europe, has an “inventive step”) compared with the prior art, which includes a host of publications and activities that predate the patent filing (including sale of the drug itself for post-marketing patents). At the same time, applicants must disclose sufficient information about the invention; for pharmaceutical patents, human clinical trial data generally are not required, but applicants must provide a credible reason to expect the drug to work, such as in vitro or in vivo testing. These requirements place temporal limits on patenting: applications must be filed after gathering sufficient evidence to satisfy the disclosure requirements, but before prior art destroys the invention’s novelty.

An important challenge for studying European pharmaceutical patenting is linking patents and drugs. Europe does not have an equivalent of the FDA’s Orange Book linking approved drugs with their patents, so Byrski and Wang construct a novel dataset. They link drugs authorized by the European Medicines Agency (EMA) to the core patent for which a patent term extension is sought through a Supplementary Protection Certificate. They use additional databases to gather information on the subsequent European patents that cite these core patents, treating these forward citations as a window into follow-on research activity. Their empirical strategy then exploits variation in approval lags—the time between the primary patent filing and the drug’s first EU marketing authorization. Because development time is uncertain at the time of patent filing, drugs with different lags can be compared to study the effect of authorization.

Their main result is that follow-on patenting declines after marketing authorization, especially for the kinds of patenting they view as most associated with “strategic” behavior. The decline is concentrated in secondary patents on subsidiary drug features such as dosages and formulations, and especially in secondary patents that stay within the same therapeutic space as the original drug. In contrast, they do not observe the same drop in citations associated with product patents covering new molecular entities, or in follow-on patenting aimed at different disease areas. The authors interpret this pattern as reassuring for meaningful follow-on innovation: firms continue to pursue new compounds and explore new therapeutic indications, but the disclosures accompanying marketing authorization make it harder to secure patents on marginal modifications within the same therapeutic space.

Byrski and Wang present various additional analyses supporting their interpretation that the decline is driven by prior art generated during marketing authorization. They find no similar decline at the end of phase II trials, a milestone that similarly reduces uncertainty about drug quality but involves less comprehensive data disclosure and thus does not create the same prior art obstacles. They also show that the post-authorization drop is concentrated in citations that signal legal vulnerability—examiner-generated citations and “novelty-threatening” citations (categories X or Y in the European Patent Office search report)—suggesting that the forgone follow-on patents are disproportionately ones that are very similar to the core patent. The paper supplements these results with robustness checks, including by examining a subset of drugs with constant market exclusivity despite varying approval lags, and by conducting an instrumental variable analysis using the time from patent filing to the start of phase I trials.

These findings add a new dimension to ongoing debates about clinical trial transparency. Even before disclosure to regulators, clinical trials can block later patents, though trial sponsors can prevent the trial from becoming prior art with confidentiality agreements. But regulators are changing this. In the United States, most drug trials (excluding phase I studies) must be registered at ClinicalTrials.gov within 21 days of enrolling the first participant, including details such as the disease and intervention being studied, although enforcement has been inconsistent. These registrations have been used to invalidate later-filed patents. New transparency requirements will likely amplify these effects: the EU’s 2022 Clinical Trials Regulation requires more comprehensive disclosure (including mandatory registration of all clinical trials, not just those tied to authorization), and there is push for the FDA to more vigorously enforce the U.S. requirements.

To be sure, distinguishing strategic patenting from meaningful follow-on innovation is difficult, and Byrski and Wang have to rely on proxies with potentially unaddressed confounders. But the data they present is suggestive of an optimistic take, that mandatory clinical trial disclosures could cut down on low-value patents on trivial modifications—which may delay generic competition without delivering meaningful therapeutic benefits—without obviously discouraging the kinds of follow-on efforts that generate new therapeutic value. This pattern is also consistent with recent work suggesting that high-value follow-on innovation is most responsive to forms of protection that meaningfully delay generic entry, rather than to incremental secondary patents that raise the costs of generic entry but do little to extend effective exclusivity.

Even so, the net welfare effect of these changes to patenting is nonobvious. For example, by incentivizing all patents to be filed earlier, greater clinical trial transparency will also reduce the effective exclusivity period so that some drugs—or some new indications for existing drugs—on the margins might never be developed. At the same time, because patent law does not generally require proof of clinical benefit, this dynamic may also induce a perverse substitution away from evidence-generating clinical trials and toward secondary patenting strategies that avoid disclosure-creating trials altogether, even where additional clinical evidence would be socially valuable. And even if these changes are accompanied by efforts to require disclosure of more evidence before firms may obtain a patent, the combination of earlier prior-art creation and heightened disclosure requirements could greatly narrow—and in some cases eliminate—the window between when companies possess patentable information and when it becomes prior art. Byrski and Wang’s data cannot resolve how these effects should be balanced, but both their conceptual framing and empirical contribution will be important to ongoing policy discussions of regulatory data transparency.

Dynamic Inventorship

Apr 2, 2026 Laura Pedraza-FariñaAdd a Comment

Laura Pedraza-Fariña

Inventors and their inventions are the building blocks of patent law: There simply is no patent law without inventors who seek to patent their inventions. One might think that a concept as foundational as inventorship would be both well-settled and extensively argued in legal opinions. But that is not the case. In fact, understanding who counts as an inventor and what constitutes an invention has surprisingly flown below the radar both in patent law scholarship and judicial opinions. In her article, The Inventorship Fallacy, Amy Motomura helps us with a deeply-researched and carefully-parsed analysis of the myriad and often contradictory ways in which courts have defined the “inventor” and the “invention.”

Motomura’s article makes a number of important contributions to the literature. First, it sheds light on the inventorship doctrine’s hidden role in policing the boundaries of related patent filings (what she calls the “indirect” role of inventorship). It then provides a taxonomy for how to think about who counts as an inventor, disaggregating the concept into three strands: manner, timing and substance. Ultimately, her inquiry into how courts define the “substance” of inventorship—that is, what a participant must contribute to become a true inventor—leads her to a second key descriptive contribution: defining who counts as an inventor requires defining what counts as an invention for inventorship purposes. This task, in turn, has no clear doctrinal answer. Rather, the Federal Circuit has implicitly articulated three mutually exclusive understandings of the invention as reflected in (1) the full scope of the claims; (2) only the novel and nonobvious elements of the claims; or (3) only the “not-well-known” or “not publicly accessible” elements of the claims.

Normatively, there are multiple ways to understand “the invention,” ways that go beyond what is perhaps the most traditional approach of seeing the invention as coterminous with the claims. The article closes by considering how these multiple conceptions of “the invention” address inventorship’s direct and indirect functions. They are an uneasy fit. No single definition of “the invention” can successfully fulfill both the direct and indirect functions of the inventorship doctrine, in large part because inventorship turns out to be a dynamic, “temporally unstable,” (P. 2453) concept. Inventors must be named early in patent prosecution, but all existing theories of inventorship subject inventorship to change, either through new information about the content of the prior art or through changes in claim construction.

Given Motomura’s diagnosis, the pressing question becomes: what next? Professor Motomura leaves a normative solution for another article—emphasizing that the doctrinal ambiguities and failures she has unearthed make the case for reform but provide no answers for how that reform should proceed.

In the next few paragraphs, I consider three avenues for future research that the Article opens up: (1) A deeper analysis of the relationship between inventorship’s indirect functions and the structure of patent law; (2) How do actual inventive communities decide whom to name as an inventor? How do these naming conventions overlap with conventions around scientific authorship? (Understanding the actual, on-the-ground naming practices of innovator communities might give us a grounded way to reconstruct what inventorship should normatively look like.) And, finally, (3) Given inventorship’s temporal instability, should the PTO adopt the “broadest reasonable inventive unit”– in a manner akin to its approach to claim construction during prosecution?

Inventorship’s Indirect Functions and the Privileging of Stable Inventive Entities
Professor Motomura highlights how inventorship structures the relationship among related patents. First, inventors matter for continuation practice because to claim priority to an earlier-filed application, the later application must have at least one inventor in common with the earlier one. Second, if two patent applications are filed by the same inventive entity (that is, if they share identical inventorship), the earlier-filed application will not count as prior art vis-à-vis the second one so long as it is filed within a year of the earlier-filed application becoming publicly available. Third, if two patents are sufficiently related and have identical ownership, obviousness-type double patenting will allow both to issue so long as their patent terms expire at the same time (through the filing of a terminal disclaimer).

Taken together, these three rules (and in particular the last two) give significant preference to stable inventive teams. In other words, an inventive team is allowed to obtain what in effect amounts to broader patent rights both by avoiding having to show later-filed patents are both novel and nonobvious over earlier ones, and through obviousness-type double patenting. In an earlier piece, Professor Motomura engaged with the justifications for these “own-prior-art” exceptions.¹ She reasoned that own-prior-art exceptions exist because innovation does not occur in discrete, fully formed jumps; rather, inventive entities make continuous, often incremental innovation.²

Nevertheless, by tying own-prior-art exceptions to the stability of an inventive entity, patent law appears to have created a prospect-like right that enables a single inventive entity to build broad patent estates incrementally. If this perspective is correct, the indirect functions of inventorship not only structure the relationship among patents but reveal a latent “prospect” feature of the patent system that concentrates patent ownership and management into the inventive entity’s hands.

On-the-ground Inventorship Norms
Professor Motomura recognizes the dynamic nature of inventorship and closes her article by asking:

“Are applicants naming inventors based on their subjective view of what is patentable? Their view of what is not well known? The full language of the claims? Given the doctrinal ambiguity . . . applicants are surely inconsistent in their approaches to naming inventors.” (P. 2472.)

These questions are ripe for on-the-ground investigation. A rich literature in scientific authorship norms provides a promising starting point. For example, in The Collective Author, Peter Galison details the complex set of authorship norms that emerged in the community of physicists at the Stanford Linear Accelerator Center project, and which aimed to balance both unity in the creative team and individual attribution.³ And Mario Biagioli has written about how, in scientific communities, authorship seeks to both assign credit and responsibility for a particular set of results.⁴ Interestingly, one scientific journal used the word “substantial contribution” to police who should be added as an author on a scientific paper, a standard that is somewhat reminiscent of the Federal Circuit’s requirement that an inventor’s contribution be “not insignificant in quality.” (P. 2444.) Conversely, critics of this approach have called for listing as a “contributor” to a scientific work each person who has “added usefully to the work”—a more inclusive approach.⁵ Perhaps current debates about scientific authorship mirror latent conceptions of inventorship in patent opinions. And these debates, as well as the on-the-ground practices of scientific communities, may help inform who ultimately should count as an inventor.

The Broadest Reasonable Inventive Unit?
Finally, the dynamic nature of inventorship may counsel a precautionary approach of sorts. The PTO could take the position that, during patent litigation, when important extrinsic facts about inventorship are not yet known, the relevant unit of analysis will be a capacious “broadest reasonable inventive unit”—mimicking the PTO’s approach to claim construction. This approach would have the virtue of sidestepping disagreements among different theories of inventorship. On the other hand, it may prove too inclusive: allowing tenuous claims to inventorship from individuals who did not contribute to the invention’s point of novelty, and undermining patent law’s incentive and attribution functions.

1. Amy Motomura, Innovation and Own Prior Art, 72 Hastings L. J. 565, 570 (2021) (“The Patent Act has been amended over time to include provisions giving limited preferential treatment to original patent applicants [that] allows the original applicant to patent otherwise unpatentable innovation.”).
2. Id.
3. Peter Galison, The Collective Author, in Scientific Authorship: Credit and Intellectual Property in Science 325-53 (Mario Biagioli & Peter Galison, eds., 2003).
4. Mario Biagioli, Rights or Rewards? Changing Frameworks of Scientific Authorship, in Scientific Authorship: Credit and Intellectual Property in Science 253, 262 (Mario Biagioli & Peter Galison, eds., 2003).
5. Id. at 265.

Inspiration versus Authorship

Mar 3, 2026 David FagundesAdd a Comment

David Fagundes

Abe Lincoln was fond of relating a story about a lawyer who tried to argue that a calf had five legs by calling its tail a leg. The folksy Lincolnian upshot was that this lawyer not only lost, but also looked foolish in the process, because simply calling a tail a leg does not make it so.

Mark Lemley’s Authoring While Dead spins a copyright version of Abe’s old yarn. Songwriters have recently begun listing as co-authors of their musical works artists who cannot be regarded as “authors” under any remotely plausible reading of the Copyright Act. Lemley’s vivid article explores the origins and rapid ascension of this industry practice. He explains with welcome drollery why this is the copyright equivalent of calling a tail a leg. He further offers a caution why what may seem like mere formalism, in fact, poses serious problems for copyright law.

The first thing I liked (lots) about Authoring While Dead was that I discovered something that was entirely new to me: In the past couple of decades, songwriters have responded to allegations of copyright infringement by offering to settle such disputes, not only with cash or an ownership interest, but even co-authorship credit to the accuser.

What makes Lemley’s article particularly engaging is that it’s chock full of colorful examples of so-called “interpolation credit.” I’ll give one of the most extreme examples here: The lyrics of The Killers’ Run for Cover includes the phrase “redemption song.” Run for Cover is not a reggae song, and it bears no musical or lyrical resemblance to the classic reggae anthem Redemption Song. Yet, when the estate of Bob Marley—who once authored Redemption Song—made (legally unavailing) noises about copyright infringement, The Killers agreed to add Marley as a co-author of Run for Cover. By doing so, The Killers apparently conceded that their 2017 track was a collaboration with an artist who died in 1981, when the band’s members were all about five years old.

Authoring While Dead explains with a welcome dose of wit that tossing around authorship credit to any artist who may have influenced you is not at all what the Copyright Act means by the term. “Author” is not defined in the statute and so remains enigmatic in some respects, but it is well understood that it means that the individual intentionally contributed some protected expression to the work of authorship. Joint authorship, in turn, means that *all* contributors to a work agree that they are consciously aware of the act of creation and buy in to the collaboration giving rise to the work. A decades-old anthem might inspire or influence a more recent composition, but that doesn’t make the author of the earlier work a conscious contributor to the creativity embodied by the new one.

The legal analysis in Authoring While Dead is straightforward, thanks in large part to Lemley’s lucid takedown of the false elision of influence and authorship. Somewhat harder, though, is sorting out the objection to this trend. If The Killers want to say Bob Marley is a co-author of Run for Cover, and the Marley estate is good with it, why should anyone object?

Authoring While Dead does an outstanding job of laying out why objecting to this practice is more than just a formalistic quibble. Adding “authors” who aren’t statutorily valid authors to a work upends the length of copyright terms, which are measured by authors’ lives. It muddies the termination of transfer right by introducing new claimants who may seek (or resist) cancellation of licenses of the work. And it makes a hash of licensing, royalties, and other rights allocations, especially because newly added authors often claim an outsized share of copyright ownership.

The deeper problem that Lemley identifies with “interpolation credits” is that as pervades the music industry, artists and labels and music publishers may begin to treat this fractured, expanded notion of authorship as valid. Norms can, and sometimes do, become industry standards even when they are not compelled by law. The film industry, for example, does life story rights deals with subjects of docudramas, even though no body of law creates an exclusive right in the facts that comprise one’s life. Here, too, as more and more artists and entities seek “interpolation credit” and demand authorship as compensation, industry practice threatens to become more unmoored from copyright law.

Authoring While Dead proposes a plausible solution, one that begins with Lemley asking what explains the surge in desire for “interpolation credit.” He locates the answer in a notion foreign (literally and figuratively) to U.S. copyright law: moral rights. The desire to insist on attribution is not rooted in something practical like seeking more royalties. It’s an artist’s understandable desire to be recognized as an influence on later work. To that end, Lemley suggests that the solution to the problem should be giving owners of earlier works “inspiration credit.” This would be a right to get credited without the later artist having to cede authorship or ownership—a right that looks a lot like the kind of attribution mandatory in continental legal systems.

This solution has promise, but my view is that Lemley is being too generous to the parties demanding interpolation credit. Sure, many or even most of them just want to be recognized. But from reading about the dozens of disputes involving interpolation credit catalogued in Authoring While Dead, it sure seems like a lot of them are also motivated by the desire to take advantage of what seems like a juicy new way to squeeze money out of a copyright catalogue.

Still, Lemley’s proposal seems promising precisely because it would force the right kind of choice on those seeking attribution. If what you want is just due recognition, then seek inspiration credit. If you deserve it, the law will back you up. No legal process is free, but the entities and older, wealthier artists who so often seek this form of credit can likely afford it. And if artists aren’t willing to pursue this form of credit, that tells you that recognition wasn’t that valuable to them in the first place. Of course, this leaves seekers of inspiration credit with (possible) recognition but no money. But it’s not like there’s no option there: If what authors really want is money, then assert a successful copyright infringement claim. But if your only basis for asserting infringement is some diffuse notion of “influence,” you’re going to lose. In suggesting a formal option for securing credit then, Lemley’s proposal would channel parties motivated by a desire for recognition versus a desire for profit into the place each belongs.

IP Savvy at the U.S. Border

Jan 30, 2026 Margo BagleyAdd a Comment

Margo Bagley

Some years ago, I reviewed a book about patent office administration, a topic even the book’s author considered “excruciatingly dull”.⁶ As such, I had low expectations for the book but was delightfully surprised by how engaging it was. I have experienced déjà vu with Jorge L. Contreras’ short article, The Role of U.S. Customs and Border Protection in Enforcing Intellectual Property Rights. An article on a topic that I would have expected to be boring in fact was quite enlightening in important ways and a pleasure to read.

I may not be alone among Intellectual Property (IP) law scholars in acknowledging a lacuna in my knowledge of IP enforcement at the U.S. border. For many of us, the topic may surface in course discussions about, inter alia, the benefits of federal trademark registration on the principal register or litigation at the International Trade Commission (ITC), but border enforcement is seldom explored in any depth.

Professor Contreras’ concise yet engrossing piece largely fills this gap. His discussion of the interplay between the treatment of different forms of IP, the different sources of jurisdiction and enforcement authority, and different actors (IP owners, importers, the ITC, district courts and US attorneys) is fascinating, and far more complicated than I previously realized.

Professor Contreras begins by situating the historical importance of the Customs and Border Protection agency (CBP) — or Customs Service as it was called when established in 1789 — as one of the first (and for a time the largest) federal agencies. It once accounted for 95% of US government revenue in the mid-1800s. He deftly explains the distinctions between CBP and a different and more controversial Department of Homeland Security (DHS) agency, Immigration and Customs Enforcement (ICE), and the complex network of CBP offices and functions. Some of CBP staff report through DHS, while others report to the Department of the Treasury.

Professor Contreras also provides an interesting history of the evolution and expansion of the ITC and its role in IP rights adjudication and enforcement. The article is chock full of insights that paint a surprising (and encouraging) picture of CBP’s IP actions and expertise. Preventing the importation of articles that infringe U.S. IP rights is a core function of the CBP under its trade enforcement mission, but this agency also has counterterrorism and border security responsibilities. In 2024 it seized more than 30 million infringing handbags, clothing, counterfeit pharmaceuticals, and other articles.

Yet, as Professor Contreras notes, the CBP has been little studied and is often dismissed as a policing body that lacks sophisticated IP expertise. The article’s discussion of the actual IP activities of CBP is quite extensive. Consider the following:

• Before seizing goods at the border, CBP must generally assess whether the goods infringe a duly registered copyright or trademark, or violate an exclusion order (issued by the ITC) or court injunction relating to patent or other IP infringement. It often needs to respond to objections to seizures and or/defend its seizure decisions. This can entail further investigations of the alleged infringement.
• CBP often meets with parties to an ITC action to discuss the scope and execution of an exclusion order and to discuss with respondents design-around possibilities that may avoid infringement. In addition, the ITC may grant CBP the authority to unilaterally permit products deemed to be successful design-arounds to be imported.
• CBP headquarters employs patent attorneys to assist with infringement determinations, both literal and under the doctrine of equivalents.
• CBP conducts Rule 177 inter partes proceedings in which a respondent may seek and obtain a pre-import ruling, on an expedited basis, that goods do not infringe. These rulings are available to the public and, of course, require the exercise of IP infringement expertise.
• The CBP must also address substantive infringement issues in formal protests filed by importers, and challenges can be heard and decided on an inter partes basis at the discretion of the CBP.
• CBP also has laboratories and conducts chemical, physical, electrical and other tests to assess infringement. CBP also often invites patent holders to provide infringement training and instruction to its officers.
• Moreover, the Federal Circuit has rejected the notion that CBP lacks the expertise to adequately enforce exclusion orders (which involves assessing infringement).

These activities suggest that CBP is capable of engaging in sophisticated infringement assessments, discussions, negotiations, and adjudications across a range of IP areas and products. They may not always get it right, but that is true of the ITC and courts as well. However, shedding light on the range of CBP tools, resources, and activities as Professor Contreras has done certainly creates the opportunity for a different narrative regarding the agency’s expertise.

After reading this article and further exploring some of its contentions, I have gained not only a greater knowledge of the IP enforcement work of the CBP, but also an enhanced appreciation for the ability of the CBP to engage in savvy technical and legal assessments of IP infringement at the border. We should all be grateful to Professor Contreras for writing this accessible and illuminating article on the IP activities and acumen of the CBP and I highly recommend it to anyone who has ever had questions about the functions of this critical agency.

1. Peter Drahos, The Global Governance of Knowledge: Patent Offices and Their Clients at xiv (2010).

Copyright for Dummies Humans

Dec 17, 2025 Kristelia GarcíaAdd a Comment

Kristelia García

From the use of copyright to combat revenge porn, to protecting slavish reproductions of public domain artworks, copyright law is sometimes invoked in circumstances where it either doesn’t, or shouldn’t, apply.⁷ Smith’s latest piece adds another (wonderful!) example to the growing list: the utilization of copyright to protect works that faithfully copy naturally occurring objects—from animals and flowers, to rocks and wood grain. In doing so, she sifts through a thicket of discordant case law to tease out a point of agreement: copyright protects only human contributions. This declaration should sound profoundly familiar to those following the ongoing battle between copyright and generative AI, in which the Copyright Office has explicitly limited registration to works—or the portion(s) of works—attributable to a human. Smith’s fascinating dissection of the case law surrounding copyright for natural objects demonstrates that there’s really nothing new about the copyrightability questions presented by generative AI, while her prescription for a “work of nature” doctrine offers a prospective blueprint for a “work of humans” doctrine with application far beyond the nature context.

In the 2003 case of Satava v. Lowry, the Ninth Circuit considered the copyrightability of glass-in-glass jellyfish sculptures. Ultimately, it issued an opinion that governs the copyrightability of creative works concerning natural objects today; namely, that “no copyright protection may be afforded to the idea of producing a glass-in-glass jellyfish sculpture or to elements of expression that naturally follow from the idea of such a sculpture.” The first part of this holding—that no copyright protection extends to the idea of a glass-in-glass jellyfish sculpture—merely restates copyright’s long-standing idea-expression dichotomy, which says that copyright protects only expressions, and not mere ideas. This explains, among other things, the curiously extensive collection of young adult vampire romance novels.⁸

The second part of the holding—that copyright protection should additionally be withheld from any part of an expression that naturally follows from the idea—lends substantially less clarity. In its relatively short opinion, the Ninth Circuit makes several (ultimately unsatisfying, in my opinion) attempts to unpack this: They assert that copyright won’t protect any aspects of the sculptures resulting from jellyfish physiology, for example, nor their vertical depiction, as jellyfish swim vertically in nature. The court concludes that “[n]ature gives us ideas of animals in their natural surroundings…[t]hese ideas, first expressed by nature, are the common heritage of humankind, and no artist may use copyright law to prevent others from depicting them.”

Smith’s thorough, quirky, and captivating deep-dive into the jurisprudence of artists’ attempts to copyright their works created by, duplicating, or inspired by nature, demonstrates that this holding has been applied with frustrating inconsistency, to the detriment of both artists and the public. Through analysis of cases ranging from museum-grade T. rex skeletons to bejeweled bee pendants to imitation wood flooring to stuffed animals, Smith identifies three broad categories of works involving nature: (1) works created by nature, (2) works duplicating nature, and (3) works inspired by nature. The first category can be dealt with summarily: no copyright protection for, say, a piece of driftwood shaped by a river’s currents, because copyright requires a work to be created by a human being. The second category, which Smith defines as works technically authored by humans but which are, by design, exact duplicates of natural objects—such as a mold of an actual stone—presents only slightly more friction: slavish copies are typically devoid of even the modicum of creativity required for copyright protection. In other words, exact replicas of naturally-occurring phenomena don’t involve a sufficient amount of human intervention, and so they too don’t merit copyright protection.

The third category is where things get interesting, and where Smith offers an eloquent and timely intervention that urges consideration of copyright validity, actual copying, and substantial similarity. Beginning with validity, Smith reminds us that a copyrightable work must involve human creativity and something beyond what is readily observable in nature, and suggests (convincingly) that the merger doctrine can help to delineate creative choices made by a human from those dictated by nature. When it comes to claims for copyright infringement, Smith first emphasizes that copyright’s “actual copying” requirement excludes expression independently created by a defendant observing the same natural phenomena. Then and only then, should we filter out the elements owing to nature and focus on substantial similarity between the original (and additional) human-added expression(s). In so doing, she concludes, we can encourage both artistic and scientific progress while preserving the public domain of nature that belongs to all of us. Smith’s article is a fun and informative read, highly recommended for those seeking to better understand copyright law and its limitations.

1. For a plethora of examples, see, e.g., Blake E. Reid, What Copyright Can’t Do, 52 Pepp. L. Rev. 519 (2025).
2. By one count, at least 159 distinct titles. Goodreads, Young Adult Vampire Romance.

Readers are Authors, Especially in the AI Age

Nov 20, 2025 Jessica SilbeyAdd a Comment

Jessica Silbey

In this essay, Abraham Drassinower updates the argument in his 2015 book What’s Wrong With Copying (Harvard University Press) with an elaboration of the nature of copyright’s public domain: it is a domain of unauthorized yet lawful copying. As in his book, Drassinower explains that the public domain is not properly understood as a problem of balancing between copyright’s public benefits and private rewards. Instead, he understands the public domain as part of authorship and thus inseparable from copyright’s construction.

The elaboration in this essay is his introduction of readership to the understanding of copyright’s public domain. As he says in its opening pages, “I will argue that the work of authorship in copyright law and doctrine is radically inseparable from what we may call the work of readership. Setting forth the immanence of readership in authorship will help construe the public domain as a necessary presupposition of, rather than a constraint on, authorship.” (P. 3.)

Before outlining Drassinower’s argument, I want to explain why I love this essay:

First, Drassinower writes like a logician, building an intricate argument from very basic premises and brings the reader along step by step.

Second, despite its sophistication, his writing is conversational and intimate, as if he’s sitting and speaking with his reader.

Third, he does not import any methodology external to law to make his argument – e.g., economics, history, literature, sociology – deriving his argument from within copyright doctrine itself.

Fourth, although his book was published in 2015, this essay reminded me of the book’s timelessness, which is remarkable in a copyright field evolving rapidly with technological evolution.

Fifth and finally, this essay helped me think about the on-going AI and copyright debates in a new way – E.g, Are the inputs to gen-AI models infringing uses? — which I appreciate because I think we need some new ways of thinking about these problems.

Drassinower’s main contribution in this essay is to think about copyright subject matter as always already involving the public – the author’s audience – and thus centering readership as essential to copyright’s purpose. He says that although copyright is about authors, it is centrally about “the relation between authors and publics, creators and users” (P. 5) because an original work of expression (copyright’s subject matter) is addressed to an audience. He quotes Jorge Luis Borges to make the point clear that authors are not creating objects or things when producing an original work of expression, but initiating a process between people. “Picking up a book and opening it holds the possibility of an aesthetic happening. What are the words resting in a book? What are those dead symbols? Absolutely nothing. What is a book if we do not open it? It is simply a cube of paper and leather … but if we read it something strange happens, I believe it changes every time.” (P. 25.) This view, for Drassinower, follows from copyright’s foundational idea/expression dichotomy, which allows all authors to repeat and build on the ideas of others, facilitating and proliferating authorship through response.

Drassinower starts with what he admits is the obvious copyright proposition that “lawful copying is part and parcel of copyright” (P. 9) because not all copying is infringement. From there, however, he explains that because the kind of copying that is lawful includes criticism, parody, research, and the free availability of ideas, the immanence of readers and responses to authorship follows directly. Reading and authorial response is built into copyright law, as a bilateral construct to authorship and original works of expression. Readership in this way is part of copyright’s public “derived as an unfolding of the fundamentals of authorship itself. As a mode of public address, authorship necessarily implicates the audience it addresses. It contemplates responses.” (P. 12.)

What does this have to do with the copyright debates concerning generative AI? First, the AI debates are often (and most recently) about the copyrighted work as a thing which when copied to feed a large language model (LLM) is contested as an unlawful use. Second, the AI debates are often (and again most recently) about the harm that LLMs might cause authors through their capacity for market substitution. These are usually questions of the idea/expression dichotomy and fair use, and to my mind not hard questions at that. But new court decisions (linked above) have muddied these doctrinal waters, presumably because generative AI feels like an existential threat to many authors (and publishers). Drassinower’s essay clarifies these issues in novel ways.

His essay on the “work of readership” reminds us that these copyright and AI debates misunderstand copyright subject matter as an object rather than a relation, a product rather than an intersubjective process. A work of authorship creates an address – an engagement – as a shared activity. And it is that activity that must be “protected” by law – the thinking activity that demands a response – not the thing copied (to be or not to be ingested by a machine).

By this logic, technical uses (as tools) are not the kind of use copyright cares about because they have nothing to do with the author as speaker (or reader as responder). And personal uses or private copies likewise have nothing to do with copyright because “copying for personal use has nothing to do with addressing another.” (P. 33.) This means copies made and not distributed are also part of copyright’s public domain. Taken seriously, these two propositions would render lawful the copying done to make LLM models produced even through unauthorized downloads of millions of books and even if (perhaps especially if!) those LLM models can produce works that readers prefer to read.

I found Drassinower’s essay provocative in the way controversial and novel frameworks often are. Much like his book on which this essay builds, I predict there is more to mine in this essay, and I hope people read it as it was meant to provoke a response!

Have Tribes Been Robbed by Trademark Law?

Oct 21, 2025 Alexandra Roberts1 comment

Alexandra Roberts

What do CHEROKEE cars, the Atlanta Braves baseball team, and Urban Outfitters’ “Navajo” panties have in common? All are examples of brands and teams adopting Native American tribal names, terms, and stereotypes as their trademarks. What does that practice reflect about the complex relationship between trademark law and native peoples?

In Tradition Is a Trap, Professor Jessica Kiser argues that US trademark law has been, and remains, ineffective and biased when it comes to acknowledging, let alone protecting, the interests of Native Americans as individuals and as a people. Kiser’s article examines how trademark law has treated Native American tribal names and culture, including through application of the Lanham Act’s former bar on registration of disparaging marks, its prohibition on the registration of trademarks that falsely suggest a connection with institutions, and its standards around rights acquisition. It builds on and contributes to other important work in Critical Race Theory, Decolonial Theory, and Race/IP fields.

Through representative stories and empirical data, Kiser considers various doctrines as applied to both marks owned by Native American tribes and marks owned or asserted by others that incorporate tribal names, terms, mascots, and trade dress. She reports that Native American tribes have not been able to access trademark law and avail themselves of its protections and exclusive rights in the ways that corporations and non-minoritized individuals routinely can and do.

This finding holds true whether the tribes and tribal members are applicants, registrants, or litigants. By setting the trademark stories she tells against the backdrop of history and situating them in specific cultural moments—from the Trail of Tears to Boy Scouts “playing Indian” to racist advertising—Kiser demonstrates how tribal names and imagery became viewed as part of the public domain and Native American culture “transformed into something so unmoored from real Native Americans that it may be hard for the white majority to disentangle it from American cultural generally.”

The systematic relocation, starvation, and assimilation of Native people and their concurrent erasure from awareness left many tribal names unknown to most Americans, and thus in some cases ineligible for trademark law’s protections. As a result, corporations like Jeep and Chevrolet have been able not only to choose, use, and receive protection for tribe names as trademarks, but also to make their products the primary associations consumers have with those names.

By drawing attention to and revealing the complex relationship between Native Americans and American trademark law, Tradition is a Trap makes an important contribution to our understanding of trademark law’s development, application, evolution, and harms. Highlighting “the ways in which trademark law’s focus on the past and on ‘imagined traditions’ harms Native American communities by holding them in a historically-frozen, disadvantaged, colonial context,” Kiser takes readers below the surface to better understand the ways that trademark law can injure and erase communities.

The Lanham Act is often regarded as neutral by courts and scholars, especially after the Supreme Court’s decisions in Tam and Brunetti struck down its prohibitions on registration of disparaging, scandalous, and immoral marks. Trademark is a Trap reveals that it is precisely those facially neutral tenets of trademark law that render it ill-equipped to remedy the inequality and racial hierarchies it replicates and perpetuates. “Trademark common law does not ask who uses a mark most efficiently, who uses the mark better, or whose use of the mark most benefits society. It asks who used the mark first,” Kiser notes, even when members of marginalized groups might have lacked the legal or economic ability to make such use.

Ultimately, the article demonstrates why when it comes to trademark law, the master’s tools cannot dismantle the master’s house. Statutory or jurisprudential changes to the Lanham Act can never adequately redress the grievances Kiser articulates. Native Americans will need to look outside of the formal trademark legal system to protect their interests, employing strategies like shaming corporations, educating the public, shifting consumer perception, building coalitions with other marginalized groups, and broadening the tent to enlist those who claim Native American ancestry without tribal affiliation.

Professor Kiser’s work lays bare not only how trademark law has failed to serve Native American tribe members and trademark owners, but why it is that trademark law supports colonizer interests at the expense of those who lack power—a question we ought to ask more often and in more contexts when considering trademark law and policy.