By , June 01, 2026.

The Fifth Amendment of the US Constitution prohibits the federal government from taking private property for public use without just compensation, embracing a principle at least as old as Cicero, who wrote in his 44 BCE book On Duties, “he who will administer a commonwealth must especially see to it that each person keeps what is his and that no public confiscation of private goods occurs.”1Marcus Tullius Cicero, On Duties, pg. 118 (Cornell Univ. Press 2016).

But this is not an article about the Takings Clause. It instead uses a recent lawsuit involving the Takings Clause as a springboard to consider a threshold point, which is this: a government taking of a digital copy of a copyrighted work does not convey any of the exclusive rights under 17 USC § 106 to the government. This is such a critical point that, though obvious to anyone with a working knowledge of copyright law, bears stating upfront so as to be clear about the proper scope of the takings inquiry.

Earlier this year, a federal district court in Valancourt v. Perlmutter entered final judgment in longstanding litigation brought by a small independent publisher to challenge the mandatory deposit provisions in 17 USC § 407 under the Takings Clause and Free Speech Clause.2Valancourt Books, LLC v. Perlmutter, 2026 U.S. Dist. LEXIS 63414. The judgment came following remand from a 2023 DC Circuit decision that held “that the mandatory deposit requirement, as applied by the Copyright Office against Valancourt, was an uncompensated taking in violation of the Fifth Amendment.”3Valancourt Books, LLC v. Garland, 82 F.4th 1222, 1226 (DC Cir. 2023).

The mandatory deposit provisions of the Copyright Act state that “the owner of copyright or of the exclusive right of publication in a work published in the United States shall deposit, within three months after the date of such publication . . . two complete copies of the best edition” of the work.417 USC § 407(a)(1). These copies “shall be deposited in the Copyright Office for the use or disposition of the Library of Congress.”517 USC § 407(b). To enforce the mandatory deposit requirement, the Copyright Office may make a written demand for the required deposit. A copyright owner who fails to make the “required deposit” within three months of a demand becomes liable for a “fine of not more than $250 for each work,” with an additional $2,500 fine for willful or repeated failure or refusal to comply with demands.617 USC § 407(d). (Note that copies of works submitted as part of an application for copyright registration under 17 USC § 408 satisfy the mandatory deposit requirement, but the mandatory deposit requirement applies even if a copyright owner does not register their work.)

In June 2018, the U.S. Copyright Office sent an email to Valancourt demanding the publisher deposit one physical copy of each of 341 works with the Library of Congress pursuant to the mandatory deposit provisions, subsequently revising the number of requested books to 240 in an updated demand letter sent in August 2018. In its final judgment, the district court issued a declaration that the August 2018 demand letter violates the Takings Clause and enjoined the Copyright Office from enforcing that demand.

This final decision, addressing only the demand for physical copies, leaves a number of issues open, among them, whether a demand for deposit of an electronic copy without compensation would violate the Takings Clause.7The decision also leaves open the question of whether 17 USC § 407 violates the First Amendment. The DC Circuit declined to reach the First Amendment claim after holding the demand violated the Takings Clause, as it sought the same relief. 82 F.4th at 1226. The DC Circuit explained that it wouldn’t proceed with that question because “neither party appears to ask us to reach the question, and because the presentation of the case does not require us to do so.” It deferred to the district court’s observation that the “analysis might raise unique questions,” such as “how the principles of the Takings Clause ‘developed in the context of ‘real property’ . . . would apply to a requirement that can be fulfilled by the transmission of digital copies.’”8Neither of the two courts acknowledged Supreme Court precedent applying the Takings Clause to intangible property; see, e.g., Ruckelshaus v Monsanto, 467 US 986, 1003 (1984) (“the Court has found other kinds of intangible interests to be property for purposes of the Fifth Amendment’s Taking Clause… That intangible property rights protected by state law are deserving of the protection of the Taking Clause has long been implicit in the thinking of this Court.”

But regardless of whether a government demand for deposit of an electronic copy requires compensating the copyright owner, there is a risk that a court or the government will conflate possession of the copy with transfer of the copyright. This is a bedrock principle of copyright law. Section 202 of the Copyright Act states, “Ownership of a copyright, or of any of the exclusive rights under a copyright, is distinct from ownership of any material object in which the work is embodied. Transfer of ownership of any material object, including the copy or phonorecord in which the work is first fixed, does not of itself convey any rights in the copyrighted work embodied in the object.”

If the government acquires a digital copy under the mandatory deposit provision, it cannot make additional copies, public performances, or public displays of the work, or exercise any of the other exclusive rights of the copyright owner. The statute provides only that such copies acquired through deposit are “for the use or disposition of the Library of Congress.”917 USC § 407(b). This language does not express any departure from Section 202 or the rest of the Copyright Act and so must be read to limit the Library to using the copy only in ways that do not implicate the copyright owner’s rights. To put it another way, no one would argue that when the Library of Congress acquires a physical book that it is permitted to reproduce additional copies of that book under the “use or disposition” language.

The use and disposition of digital copies is rather limited. Many uses of digital copies may in fact implicate the exclusive right of reproduction, a point the U.S. Copyright Office has long noted. For example, in its 2001 DMCA Section 104 Report, the Office wrote, “All of the familiar activities that one performs on a computer — e.g., execution of a computer program, retrieval and display of information, browsing the World-Wide Web — necessarily entail making reproductions in RAM,” and such reproductions “are generally ‘fixed’ and thus constitute ‘copies’ that are within the scope of the copyright owner’s reproduction right.”10Section 104 Report at 107-110. The Office cited to that discussion in a 2016 report on Software-Enabled Consumer Products, adding, “It appears that the RAM copy doctrine is today firmly established as a matter of case law.”11Software-Enabled Consumer Products Report at 19 n.101.

There is also an open question as to what authority the Library of Congress would have to display to members of the public digital copies in its collections of works that are still protected by copyright law. A copyright owner has the exclusive right to display a work publicly.1217 USC § 106(5). The Act defines “display” to mean “show[ing] a copy of [the work], either directly or by means of a film, slide, television image, or any other device or process,” and it defines displaying a work “publicly” to mean, in relevant part, to display it at “a place open to the public …or… to transmit or otherwise communicate a … display of the work to a place [open to the public].”1317 USC § 101. Section 109(c) creates an exception that permits the owner of a particular, lawfully made copy to display that copy publicly “either directly or by the projection of no more than one image at a time, to viewers present at the place where the copy is located.” The plain text of these provisions would appear to permit the public display of a digital copy stored locally, i.e., on or near the device on which it is being viewed. The text would not appear to permit the public display when the digital copy is stored off site or in the cloud, since viewers would not be present at the same place where the copy is located. (Additionally, Section 109(c) is only an exception to the public display right and would not apply to any reproductions made as a result of any digital transmissions.)

Any uses of the work that implicate the copyright owner’s exclusive rights would require action by Congress to be permitted. Congress knows how to create exceptions for such a purpose when it wants to. For example, rather than relying solely on copyright owners of programs transmitted to the public to deposit copies of such works in the Library of Congress, Congress has authorized the Library itself “to make a fixation of a transmission program directly from a transmission to the public, and to reproduce one copy or phonorecord from such fixation for archival purposes.”1417 USC § 407(e)(1). The existence of this exception provides further support for the points made above regarding the use and disposition of digital copies acquired by the Library.

The public undoubtedly benefits from the collections of the Library of Congress, which encompass an unparalleled collection of historically and culturally significant works that are the product of the creativity, talents, and knowledge of countless individuals. But the Fifth Amendment of the Constitution limits the government’s ability to compel private individuals to subsidize public benefits by requiring compensation for any such takings. When the subject of a governmental taking is electronic copies of copyrighted works, then the analysis must begin with the understanding that the government’s acquisition of the copy does not also convey the exclusive rights to the work embodied in the copy.

References

References
1 Marcus Tullius Cicero, On Duties, pg. 118 (Cornell Univ. Press 2016).
2 Valancourt Books, LLC v. Perlmutter, 2026 U.S. Dist. LEXIS 63414.
3 Valancourt Books, LLC v. Garland, 82 F.4th 1222, 1226 (DC Cir. 2023).
4 17 USC § 407(a)(1).
5 17 USC § 407(b).
6 17 USC § 407(d).
7 The decision also leaves open the question of whether 17 USC § 407 violates the First Amendment. The DC Circuit declined to reach the First Amendment claim after holding the demand violated the Takings Clause, as it sought the same relief. 82 F.4th at 1226.
8 Neither of the two courts acknowledged Supreme Court precedent applying the Takings Clause to intangible property; see, e.g., Ruckelshaus v Monsanto, 467 US 986, 1003 (1984) (“the Court has found other kinds of intangible interests to be property for purposes of the Fifth Amendment’s Taking Clause… That intangible property rights protected by state law are deserving of the protection of the Taking Clause has long been implicit in the thinking of this Court.”
9 17 USC § 407(b).
10 Section 104 Report at 107-110.
11 Software-Enabled Consumer Products Report at 19 n.101.
12 17 USC § 106(5).
13 17 USC § 101.
14 17 USC § 407(e)(1).
By , May 29, 2026.

CNN Sues AI Firm Perplexity, Alleging It Engaged in ‘Massive Copyright Infringement’ — “In the 54-page federal complaint, CNN points to Perplexity’s former ads claiming that it helps users skip the ‘extra steps and clicks’ they’d normally take to access news content. The network said it spoke to Perplexity about a deal last year but could not agree on terms, including restrictions on how the company’s chatbot uses CNN’s content.”

VerticalScope sues OpenAI, claims AI giant infringed copyright by scraping content to train GPT models — “Filed in the Ontario Superior Court of Justice on Tuesday, VerticalScope’s statement of claim alleges that OpenAI’s misuse of its content unjustly enriched the San Francisco-based AI giant. The Canadian company is seeking damages and a permanent injunction to stop OpenAI from scraping its content.”

China’s MiniMax loses bid to end Disney copyright lawsuit over AI system — “MiniMax argued that the California court could not hear the case because the company does all of its business in China. But Blumenfeld said there was evidence ​that the company was ​offering the Hailuo ⁠app in the United States.”

X is Cracking Down on Accounts That Rip Off the Work of Photographers and Other Creators — “In an announcement, X’s head of product Nikita Bier says his team has ‘identified a number of large accounts that have been programatically reuploading content.’ Bier says these accounts will circumvent crediting the original author.”

Pass Go, Collect Copyright: The Legal Side of Board Games — “Board games are fun, engaging and a treasure trove of IP protection. Board games are a great example of how copyright draws that line: the mechanics are free for anyone to use, but the creative choices: the artwork, the wording, the overall look and feel, are where protection sits.”

By , May 15, 2026.

C in a Circle: Why I Started this Blog Series — From Steven Tepp. “There are already plenty of blogs and public discussion about current events in copyright. My focus is a future-looking vision of what the Copyright Office and the American copyright system can and should be. That may from time to time include addressing current events, but that is not my focus. I will be writing about the big-picture questions of the Copyright Act, how individual issues and cases illustrate strengths and/or shortcomings, and the operation of the Copyright Office as the agency that administers the law and in particular the registration system.”

Authors, publishers near final approval of $1.5 billion Anthropic copyright settlement — “In addition to monetary relief for the plaintiffs, Anthropic will also be required to destroy the original files of the works they downloaded from the pirated book datasets within 30 days of the final judgment and certify that Library Genesis and Pirate Library Mirror datasets with pirated material were used in training any of the company’s commercially released large language models.”

Copyright isn’t dead in the age of AI, it’s key to growing UK creative industries — “In the 2010s, when I served as minister for culture and digital economy in David Cameron’s government, we heard the same chorus (though back then it was about illegal file-sharing): copyright is outdated and infringement is the price of modernity. It wasn’t true then and it isn’t true now. In reality, copyright is the market mechanism that underpins creativity and turns it into a precious asset: something you can buy, sell, license, finance and build a business around. If we want Britain to be both an AI superpower and a creative powerhouse, the answer is to recognise the market making power of copyright and support commercial licensing at speed and at scale.”

ECJ ruling on ancillary copyright: Meta must pay for press content in Italy — “A key point of the ruling concerns the power imbalance between platforms and publishers. The ECJ explains that publishers are in a weak negotiating position because often only the platforms have the data to quantify the economic value of a use. Therefore, it is lawful to oblige providers like Meta to disclose the information necessary for calculating remuneration. Furthermore, operators may not simply restrict the visibility of news content during ongoing negotiations to exert pressure or obscure the value of the content.”

Access to sites with pirated content blocked under copyright law revision — “A revision to the Copyright Act that was passed in January went into effect on Monday, allowing the Ministry of Culture, Sports and Tourism to order network operators to block user access to websites that repeatedly commit copyright offenses, namely piracy. Prior to revision, the consequences of copyright offenses were determined by the Korea Communications Commission’s review committee.”

By , May 08, 2026.

Publishers and Authors Sue Meta, Alleging ‘Massive’ Copyright Infringement Behind Its Llama AI Service — “‘Meta’s copying of these literary works was neither incidental nor accidental,’ the publishers’ suit claims. ‘It was a calculated decision driven by the need to appropriate the expressive value of copyrighted written works to produce an AI system capable of generating outputs that directly compete with literary works written by human authors.'”

NVIDIA’s Shadow Library Scripts ‘Have No Other Purpose’ Than Infringement, Judge Rules — “This appears to be the first AI training case to apply the new Cox standard, and the result didn’t go the way NVIDIA hoped. The scripts it offered satisfied both the new ‘inducement’ and ‘tailored to infringement’ standards required for a contributory infringement finding.”

ASTM v. UpCodes is a Troubling Misapplication of Warhol’s Transformative Use Clarifications — “The bottom line is that both ASTM and UpCodes published the works for the purpose of helping the public understand the codes and technical standards that impact and govern safety. To say that UpCodes’ purpose is transformative simply because they take the works and post them where the public can access them for free is antithetical to the foundations of copyright law. If that logic were applied widely, it would excuse almost any type of piracy so long as the infringer claimed it was in service of some kind of public interest.”

Annie Leibovitz copyright infringement lawsuit revived in case involving Star Wars movie photos — “The appeals court held that Leibovitz’s retention of certain rights does not automatically foreclose the representative from asserting status as an exclusive licensee, namely because rights can be divisible and subdivisible under Section 106 of the Copyright Act.”

CCC adds AI reuse rights to higher education copyright licence — “CCC said the expanded licence is intended to help institutions address what it describes as a ‘growing permissions gap’ by enabling the authorised internal reuse of lawfully acquired copyrighted content within AI systems for purposes including summarisation, chatbot development, prompting, and research support.”

By , May 01, 2026.

Devious New AI Tool “Clones” Software So That the Original Creator Doesn’t Hold a Copyright Over the New Version — “The advent of generative AI continues to undermine the very concept of copyright, from entire books shamelessly ripping off authors to tasteless AI slop depicting beloved characters going viral on social media. The sin is foundational: all today’s popular AI tools were built by pillaging copyrighted material without permission. Even software isn’t safe.”

Books, language and knowledge ownership: Why publishing and copyright matter in South Africa — “In this rapidly evolving digital and AI-driven era, the continuing significance of books becomes even more pronounced. While artificial intelligence accelerates the production, dissemination, and consumption of information, it often privileges speed, volume, and efficiency over depth and reflection. Books, by contrast, invite sustained engagement, critical thinking, and ethical deliberation.”

Japan ruling party seeks AI penalties over deepfakes, piracy — “Japan’s ruling Liberal Democratic Party is urging the government to consider adding penalties to the country’s artificial intelligence law, as concerns grow over deepfakes and copyright infringement tied to generative AI.”

Netflix Prevails in ‘Tiger King’ Copyright Case, a Win for ‘Fair Use’ in Documentaries — “Two years later, after additional briefing and argument, the court decided it was wrong. In so doing, the court relied on cases from the 9th and 4th circuits that have allowed documentary makers to use copyrighted clips without permission to make some broader point. ‘Tiger King’s use of the Funeral Video clips is classic documentary-style borrowing,’ Holmes concluded.”

The Best (IP) Offense is a Good (IP) Defense — “Effective IP protection requires laws that are kept up to date to address how IP crimes have evolved over time. To truly protect American sports fans, teams, and rightsholders in the era of live piracy, the U.S. Congress should create a judicially supervised website blocking tool similar to those proven to work in over 55 nations around the world, including many of our strongest allies. Judicial site blocking allows a court to order US intermediaries (most often internet providers) to block foreign piracy sites from reaching and victimizing American viewers”

By , April 24, 2026.

Thousands of authors seek share of Anthropic copyright settlement — “The authors sued Anthropic in 2024, arguing that the company, which is ‌backed by ⁠Amazon and Alphabet, used pirated versions of their books without permission or compensation to teach Claude to respond to human prompts. The case is one of dozens brought by copyright owners including authors and news outlets against tech companies over the training of their large language models, and the first major U.S. case to settle.”

The Character Without an Author: Character Copyright Protection in the Age of Generative AI — “As GenAI tools become more integrated into creative pipelines, courts and the U.S. Copyright Office will need to clarify how these technologies interact with longstanding character protection doctrines.”

Kraftwerk Loses Two-Decade-Long Copyright Dispute After Court Rules Unauthorized Sample Was a ‘Pastiche’ — “The path to the decision has been a long and complex one, with 22 years spent circumnavigating the European justice system across appeals and remandments that ping-ponged between two regional German courts, the German Federal Court of Justice, the German Federal Constitutional Court and, now, the ECJ.”

Pro Codes Critics Have It Exactly Backwards — “For more than 125 years, nonprofit standards development organizations (SDOs) have produced the rigorous safety codes and standards that protect American lives, addressing everything from building safety to wildfires to active shooters. These organizations develop standards through a transparent, consensus-based process at no cost to taxpayers. It’s one of the most successful public-private partnerships in American history.”

By , April 17, 2026.

Pirate site Anna’s Archive hit with $322M default judgment over music scraping — but will Spotify and record labels ever see the money? — “The judge’s ruling also requires internet service providers including Cloudflare to disable access to Anna’s Archive’s domain names and prevent websites that host the infringing content or facilitate the distribution of this content.”

CJEU delivers pastiche-style judgment on … pastiche — “With its second referral in the long-running dispute in Pelham, Germany’s Federal Court of Justice (BGH) asked the CJEU to tackle the notion of pastiche, this being a defence that – just a few years ago – very few people would have noticed let alone considered invoking. Things have changed: pending the second Pelham referral, some considered (and recommended considering) pastiche a potentially very broad defence, which would cover nearly every ‘creative’ re-use of copyright works and other protected subject-matter. Unsurprisingly, this is not the case.”

EU copyright law roundup – first trimester of 2026 — “Welcome to our first roundup for 2026. In the first trimester of this year, several important developments have taken place in the EU copyright law arena. As our regular audience will know, in this series we have been reporting on CJEU judgments, Advocate General (AG) Opinions (if any) and significant policy developments.”

MPA’s Rivkin: “Protecting copyright an economic necessity” — “Motion Picture Association Chairman and CEO Charles Rivkin has reaffirmed the body’s commitment to protecting copyright in the age of AI. Delivering his annual State of the Industry address at CinemaCon in Las Vegas, Rivkin noted that every time there had been a new advancement, the industry had persisted.”

Anthropic Code Crisis Creates Copyright Contradiction — “Anthropic has sought to minimize the impact of the unintended disclosure, and turned to US copyright law to do so—issuing a takedown notice under the Digital Millennium Copyright Act (DMCA) to GitHub in a bid to have its leaked code taken offline, or at least to restrict its circulation somewhat. The company’s critics, though, have been swift to spot the irony: Anthropic, like most of the broader AI industry, has been embroiled in copyright lawsuits for several years—which have seen the company’s lawyers make some expansive claims about why copyright does not prevent the training or development of AI models.”

By , April 10, 2026.

How Creators and Creative Industries Are Pushing Back Against AI Theft — “These campaigns call on policymakers, politicians, and the creative community to take action to ensure that artists’ works are licensed and that they are compensated before AI companies use them and respect the foundational copyright law frameworks that drive human-centric creativity and innovation.”

Is ChatGPT allowed under EU copyright law to generate a ‘Game of Thrones’ sequel? — “Overall, sequels to fictional novels and other works of fiction are generally considered copyright infringement under EU copyright law if they are created without the consent of the rights holders. But does the fact that a generative AI such as ChatGPT writes such a sequel change anything? Most likely, it does not.”

UMG and Believe settle lawsuit that alleged ‘industrial-scale’ copyright infringement of Universal’s music — “The complaint focused heavily on the distribution of so-called “manipulated” audio — tracks that were typically sped-up or remixed versions of copyrighted recordings, uploaded to major streaming platforms including Spotify, Apple Music, YouTube, TikTok, and Instagram without authorization.”

Judge slams key OpenAI witness in copyright infringement case for ‘hazy recollections’ — “An unimpressed Manhattan judge ordered a corporate representative for OpenAI to undergo a second deposition after finding he failed to answer ‘even the simplest questions’ the first time around about what the company has described as efforts to limit chatbots from stealing writers’ work.”

By , April 03, 2026.

Top Noteworthy Copyright Stories from March 2026 — “March 2026 was one of the busiest months in recent memory for the copyright world, with huge developments on critical copyright issues that all creators and copyright owners must know about.”

Penguin to sue OpenAI over ChatGPT version of German children’s book — “The lawsuit, which was filed on Friday with a Munich court against OpenAI’s Ireland-based European subsidiary, states Penguin Random House’s legal team had prompted ChatGPT to write a story in the vein of Penguin author and illustrator Ingo Siegner’s Coconut the Little Dragon series.”

Publishers Back Concord Music Group’s AI Suit — “Anthropic’s actions, the brief argues, fail to meet the fair use doctrine in at least two ways: it is not transformative and its unauthorized training interferes with copyright holders who are signing licensing agreements with various tech companies.”

Training Data, Market Dilution, and the Elephant in the Room: Why the Three-Step Test Matters for Generative AI — “It is a curious feature of the genAI copyright debate that an instrument of binding international copyright law, one that limits the legislative discretion of every WTO member and governs the scope of every copyright exception in the EU and the US, has so rarely been subjected to sustained analysis in this context.”

Anthropic has ‘come to copyright’ epiphany after Claude code leak — “The code quickly found its way to GitHub and the curious were suddenly trawling the more than 2,000 files containing 512,000 lines of code.So developers did something very Anthropic – gobbled it up and began sharing it with others, often in reworked versions.Anthropic’s response was a ‘come to copyright’ moment. The company began issuing takedown notices to GitHub, using the Clinton era Digital Millennium Copyright Act (DMCA).”

By , March 27, 2026.

Supreme Court Reverses $1 Billion Verdict, Rules Cox Not Contributorily Liable for Subscribers’ Copyright Infringement — “Justice Sotomayor, joined by Justice Ketanji Brown Jackson, filed an opinion concurring only in the judgment. Justice Sotomayor agreed that Cox is not liable, but argued that the majority unnecessarily limited secondary liability. She said that the Court’s precedents leave open the possibility that ‘other common-law theories, such as aiding and abetting, could apply in the copyright context.'”

We were wrong on AI copyright so we changed our minds, admits Culture Secretary Lisa Nandy — “The Wigan MP added: ‘Copyright law in this country is based on one very fundamental principle, which is that people should own their own work.’”

Copyright holders are ready to do AI deals – under existing laws — “Publishing, music and entertainment bodies say their members are tech-savvy and ready to do licence deals with artificial intelligence companies, and existing copyright laws are capable of achieving that outcome. And in the strongest indication to date that the Albanese government holds the same view, Attorney-General Michelle Rowland said the existing copyright regime had served Australia well for more than half a century.”

AG Emiliou advises CJEU to rule that OCSSPs’ authorization under Article 17 of the DSM Directive extends to acts of reproduction — “On a broader level, the Opinion offers a valuable reminder that distinct rights are indeed separate and that there is a requirement to guarantee their high level of protection. Insofar as the right of reproduction is concerned, the latter also stems from its broad formulation in both international and EU law.”