As expected, based on the tenor of the Justices’ questions during oral argument, the U.S. Supreme Court has ruled against a trademark applicant seeking to register a mark commenting on former President Donald Trump. The decision represents something of a shift from prior decisions striking down portions of the federal Lanham Act on First Amendment grounds.

As we wrote earlier this year, the Supreme Court granted cert to review Vidal v. Elster, which involved the U.S. Patent and Trademark Office’s (USPTO) refusal of Steve Elster’s application to register the trademark TRUMP TOO SMALL for t-shirts. In refusing registration, the USPTO cited Section 2(c) of the Lanham Act, which prohibits the registration of marks that consist of or comprise a name, portrait, or signature identifying a particular living individual except when the individual has provided his or her written consent. The USPTO found that Elster’s trademark clearly identified former President Donald Trump, and so refused registration under Section 2(c), as Trump had not provided consent. Elster claimed, however, that this section of the Lanham Act is unconstitutional because it violates his First Amendment right to free speech, i.e., his right to criticize a political figure.

On appeal, the Federal Circuit found Elster’s argument persuasive, and held that the names clause violated the First Amendment. The Supreme Court, however, disagreed with Elster and the Federal Circuit, ruling 9-0 in favor of the USPTO in affirming the agency’s refusal of Elster’s trademark application. Delivering the Court’s opinion, Justice Clarence Thomas explained: “The Lanham Act’s names clause has deep roots in our legal tradition. Our courts have long recognized that trademarks containing names may be restricted. And, these name restrictions served established principles. This history and tradition is sufficient to conclude that the names clause—a content-based, but viewpoint-neutral, trademark restriction—is compatible with the First Amendment.”

Justice Thomas distinguished this case from other recent cases pitting the Lanham Act against the First Amendment. To wit, the high court previously struck down provisions of the Lanham Act prohibiting registration of marks that are disparaging (Matal v. Tam) and marks that are immoral or scandalous (Iancu v. Brunetti). In each case, the Court held that the relevant provisions of the Lanham Act were not viewpoint neutral, and therefore violated the First Amendment.  In the instant case, however, the Court found that the prohibition in question (trademarks identifying a particular individual) is viewpoint neutral: “The names clause does not facially discriminate against any viewpoint. No matter the message a registrant wants to convey, the names clause prohibits marks that use another person’s name without consent.” Accordingly, the Court held that Section 2(c) of the Lanham Act does not violate the First Amendment.

The author of the lyrics for Canada’s national anthem, “O Canada,” probably didn’t have trademarks in mind when he wrote “we stand on guard for thee.”  But a recent trademark infringement win for a Maryland-based U.S. non-profit corporation in Canada shows that Canadian courts will guard against consumer confusion and enforce trademark rights even when the trademark owner is a U.S. company and the defendant is Canadian.  The case also highlights the importance of obtaining trademark protection in jurisdictions outside of the United States–particularly in U.S.-bordering Canada. 

The U.S.-based Telegu Association of North America (which we’ll refer to as US-TANA) brought suit in a Canadian Federal Court against a Canadian corporation using the same name and identical trademarks.  US-TANA owns US and Canadian trademark registrations for the mark TANA, and also claimed rights in the marks TANA & DESIGN and TELUGU ASSOCIATION OF NORTH AMERICA.  US-TANA operates a website at tana.org and several social media accounts, provides printed materials such as a monthly newsletter, and hosts numerous events that are attended by members across North America.  The Canadian Court recognized that the US-TANA’s use of the TANA-related marks was more extensive in the United States.  Nevertheless, the Canadian Court noted that US-TANA had continuously used its marks in Canada beginning at various points since 1992.

The Canadian company (which we’ll refer to as CA-TANA) used the identical name, Telugu Association of North America, and was incorporated in 2020 – two-months after one of its founder’s membership with US-TANA was terminated.  In addition to adopting an identical name, CA-TANA registered the domain tana.live and used similar TANA and TANA & DESIGN marks. 

The Court ruled in favor of US-TANA, finding that the CA-TANA’s use of and identical company name, identical trademarks, and confusingly similar trademarks violated several sections of Canada’s Trademark Act.  Key to the Court’s ruling was the fact that US-TANA had used its marks, and accumulated goodwill among consumers, in Canada.  While the Court’s ruling is foreseeable, the case holds several reminders for U.S.-based companies operating beyond the confines of the United States. 

First, register your trademarks outside of the United States.  U.S. companies should, costs permitting, apply for trademark registrations in those jurisdictions where they do business and where they plan to do business in the future.  In this case, US-TANA’s registration of certain marks in Canada provided specific causes of action and remedies that would have otherwise been unavailable to it under Canadian law. 

Second, use your trademarks.  While not available in all jurisdictions, US-TANA’s use of the its marks in Canada provided it with causes of action for common law use of the marks.  In addition, US-TANA’s use of the marks in Canada allowed US-TANA to validate its rights in Canada without having to rely on ancillary uses from the United States. 

Third, the United States is not the sole venue to enforce trademark rights.  In fact, the U.S. court system may not even be an option depending on the circumstances.  For example, US-TANA could have perhaps sued CA-TANA in the United States, arguing that Canada borders the United States and that through CA-TANA’s tana.live website, U.S. citizens might have been confused by CA-TANA’s use of identical or similar marks.  Nevertheless, it is not clear whether any potential confusion among U.S. citizens would have been sufficient to for a U.S. court to assert jurisdiction over CA-TANA if, among other things, its conduct was not specifically directed at the United States.  In addition, U.S. courts are unlikely to award any damages originating from conduct in Canada in light of the U.S. Supreme Court’s recent decision in Abitron v. Hetronic (which you can read about here).

In any event, the decision by the Canadian Federal Court is a good reminder that U.S. trademark owners should not forget about “the True North strong and free” when they are registering, using, and protecting their trademarks.

Seyfarth Shaw’s Intellectual Property Group was out in full force at the recent International Trademark Association (INTA) annual conference in Atlanta on May 18-22, 2024.  Fifteen Seyfarth attorneys, representing nine different Seyfarth offices, attended:  Amy Abeloff (Los Angeles – Century City), Anne Yates (Atlanta), Bart Lazar (Chicago), Brian Michaelis (Boston), Ed Maluf (New York), Jay Myers (New York), J.C. Heinbockel (New York), JC Zwisler (Boston), Ken Wilton (Los Angeles – Century City), Lauren Leipold (Atlanta), Lisa Meyerhoff (Houston), Matthew Moersfelder (Seattle), Owen Wolfe (New York), Robert Terzoli (Washington, D.C.), and Stephen Lott (Charlotte).  The Seyfarth team also included IP professionals from Seyfarth’s Atlanta office: Eugene Ahn, Ganell Hume, Marissa Williamson, Melissa Hazel, and Tanya Tann.  Our practice development leads, Danielle Freeman and Joan Gwak, also attended, with assistance from Atlanta office team members Jennifer Malcho and Lisa Stewart.

On Sunday, May 19, Seyfarth hosted a reception for INTA attendees that was attended by hundreds of lawyers and IP professionals from around the world.  Seyfarth attorneys also participated in panels and committee meetings, as well as networking meetings with foreign associates, and of course reconnecting with old friends.  As always, the INTA Annual Meeting provided a reminder of how wonderful the world could be if everyone practiced trademark law!  The team looks forward to having a strong presence at INTA’s Leadership Meeting in November, as well as next year’s Annual Meeting in San Diego! In the meantime, below is a peak at our time in Atlanta!

We are still waiting for a formal ruling on the Andersen v. Stability AI defendants’ second round of motions to dismiss, but so far it’s looking like most of the case may be allowed to proceed to discovery. The judge heard oral arguments on May 8, 2024 in this case involving image-generating AI software, a day after issuing a tentative ruling seeming to give the plaintiffs a chance to try to prove up at least some of their claims.

As a brief recap, a class of  visual artists are suing Stability AI, Runway AI, Midjourney, and DeviantArt, alleging that the defendants’ image-generating AI programs or related activity infringed the plaintiffs’  original works in violation of the Copyright Act.  (For additional background, see our most recent update on the case here).  Victory originally seemed unlikely for the plaintiffs after the judge tossed out most of their case in response to the defendants’ first round of motions to dismiss.   But the plaintiffs filed an amended complaint addressing what the judge said was a failure to allege “substantial similarity” by including side-by-side comparisons of their works and the AI programs’ allegedly similar outputs.  After we highlighted some examples here, our readers weighed in on whether they thought the examples were “substantially similar” or not.

Although the judge did not address substantial similarity in his tentative ruling, he wrote that plaintiffs’ allegations that certain defendants used the plaintiffs’ works to train their AI programs “suffices for direct infringement as to Stability, Runway, and Midjourney,” and that “plaintiffs have plausibly alleged facts to suggest that compressed copies…of their works are contained in the versions of the Stable Diffusion” AI program used by the defendants.  The judge stated that “[]the facts regarding how the [AI] models operate, or are operated by defendants, should be tested” after discovery.

The judge chastised the plaintiffs a bit in his tentative ruling, stating that they should have sought permission before trying to add additional plaintiffs to the case.  But he then went on to say that he is inclined to give plaintiffs leave to file a new complaint to add the new plaintiffs with his permission. 

It wasn’t all bad news for the defendants, however.  The judge indicated that he would dismiss some of the plaintiffs’ claims, including those brought under the Digital Millenium Copyright Act (DMCA) and a claim for breach of contract or breach of the implied covenant of good faith and fair dealing against DeviantArt.

According to reports of the May 8 hearing, the defendants worked hard to convince the judge that he should dismiss all of the claims.  The reports suggest that the judge kept his views close to the vest and did not indicate which way he might ultimately rule after more than an hour of argument.

Ultimately, the judge’s tentative ruling is just that: tentative.  We’ll be watching closely to see what the judge’s final decision says.  We suspect other plaintiffs in similar suits will be watching closely as well, given that a final order reflecting the tentative ruling could be persuasive to other judges juggling similar claims.

In 2014, the Supreme Court upended U.S. patent law in the landmark ruling for Alice Corp. v. CLS Bank International.  The Alice decision established new standards for determining whether inventions, especially those related to software and business methods, are eligible for patents.

Following the Alice decision, there was a significant increase in 101 rejections, particularly for software and business method patents.  A “101 rejection” occurs when a patent application is rejected based on Section 101 of the U.S. Patent Act, which outlines patentable subject matter.

In extremely simple terms, based on the Alice decision, you cannot patent an idea simply because your idea uses a computer.  You need to show that your invention does something more than just using a computer to perform a basic, abstract task.  As we are approaching the ten-year anniversary of the Alice decision, one would think 101 issues would have been settled by now.  But, of course, nothing is ever easy in the world of patents.  It’s as if we are playing a game of 101 whack-a-mole—just when you think you have 101 figured out, a new case is decided!  Take for instance one of the most recent 101 decisions, AI Visualize, Inc. v. Nuance Communications, Inc., where the Federal Circuit once again held that software-related claims were ineligible for patent protection. 

AI Visualize

In AI Visualize, the patents in question relate to the advanced visualization of medical scans using a dedicated web portal.  These patents describe systems and methods for creating and viewing three-dimensional virtual views of medical imaging data, such as MRI scans, stored in a centralized server.  Specifically, claim 1 included the step of creating “on the fly” virtual views.  The technology was designed to facilitate the viewing of large volume visualization datasets (VVDs) over low-bandwidth internet connections, without the need for local storage of the data.  

The Federal Circuit applied the two-step Alice test to determine patent eligibility of AI Visualize’s claims under Section 101.  At the first step of the Alice analysis, the court assessed whether the claims of AI Visualize were directed to an abstract idea.  The claims involved creating, manipulating, and displaying medical imaging data using standard computer functions such as data storage, retrieval, and transmission over networks.  In past cases, inventions that created data typically overcame the first step of the Alice test.  In their reply brief, exasperated counsel for AI Visualize desperately pointed towards the creation of virtual views, stating “[t]he claim limitations expressly recite the … requirements for a system to create virtual views, which the district court and Appellees removed from the character of the invention.”  The fact, then, that the patents seemingly had all the ingredients to pass muster is what makes the court’s following decision so surprising.

The court found that such functions, when claimed broadly and generically, represent abstract ideas rather than concrete technological advancements.  Regarding virtual views, the court stated, “the claim language makes clear that virtual view ‘creation’ is achieved by the manipulation of a portion of the existing [data].”  The opinion emphasized that the claims did not focus on a specific means or method that improved technology, but rather on abstract data manipulation.  Therefore, the claims were merely directed to an abstract idea.

For the second step of the Alice test, AI Visualize argued that its method for generating virtual views “on the fly” was an inventive concept that transformed the abstract idea into a patent-eligible application.  However, the court disagreed, noting that the process of generating virtual views, as claimed, was a conventional computer function that did not move beyond the abstract idea.  The court emphasized that the creation of a virtual view was an abstract idea that was also known in the art, as conceded in the patent specifications and later at oral arguments.  Therefore, that limitation could not supply the inventive concept required to transform the claims into patent-eligible subject matter.  

Key Takeaways from the Decision

The court’s decision underlines the necessity for claims to demonstrate a tangible enhancement in computer technology.  At first glance, the court painted a bleak picture for patent eligibility by stating that the virtual views are merely “achieved by the manipulation of a portion of the existing [data].”  After all, it would be incredibly difficult to create information that is not based on existing data.  

However, other statements from the opinion and oral arguments indicate that the court focused on the lack of specificity in describing how the virtual views were created.  For example, one claim limitation included “creating the requested frames of the requested views from the volume visualization dataset.”  The court addressed this limitation by stating “this ‘creation’ of a virtual view from the existing [data], recited in general terms, is abstract data manipulation.”  Therefore, it is not sufficient for a claim to merely recite creation steps in general terms; claims should include a specific, substantial improvement over existing technology.

The case also emphasized the importance of careful patent drafting.  In its opinion, the court seemed to only work the specification against AI Visualize.  First, the court bluntly refused AI Visualize’s arguments that the specification clarifies the creation of virtual views.  The court instead insisted that the claims themselves must include a technical solution to a technical problem.  Second, the court trivialized the creation of virtual views by noting that the patent specification admitted that the technology for creating virtual views existed prior to the invention.  This contributed to the court’s conclusion that the patent did not offer patentable subject matter.  This all goes to show that courts are increasingly willing to weaponize specifications against patentees that have not carefully drafted their patents.

Conclusion

AI Visualize exemplifies the ongoing trend of raising the bar for patent eligibility, particularly in cases where software-related patent specifications and claims lack clear demonstrations of technical solutions to technical issues.  The decision in AI Visualize could also provide valuable guidance for patents involving generative AI.  Merely generating data from existing sources may not suffice to overcome 101 rejections.  Therefore, practitioners must recognize that generative AI, or any form of AI, is not a guaranteed solution for obtaining patent approval.  What’s crucial is specificity and precise articulation of the technical solution.

The U.S. Supreme Court affirmed the Eleventh Circuit’s holding in Warner Chappell Music v. Nealy that copyright plaintiffs bringing timely claims of infringement may recover damages for acts occurring outside the three-year statute of limitations. The ruling addresses a longstanding circuit split over whether monetary relief is available even where infringement occurred more than three years prior. However, it does not squarely address the larger question of whether the “discovery rule” always applies in this context, leaving open issues the Court may address in another copyright case on its docket this term.

As discussed in our previous post, the Supreme Court granted certiorari in this case over a song by the rapper Flo Rida on the narrow question of whether a copyright plaintiff can recover damages for acts that allegedly occurred outside of the three-year statute of limitations for copyright claims. The Ninth and Eleventh Circuits had been applying the discovery rule, holding that copyright plaintiffs could recover damages from any point in time, as long as the claims were filed within three years of discovery of the infringement. The Second Circuit, on the other hand, held that a copyright plaintiff cannot recover damages for acts that occurred more than three years prior to the filing of the suit under any circumstances.  As we previously reported, at oral argument, a majority of Supreme Court justices seemed skeptical of the Second Circuit’s approach.   

Justice Kagan penned the majority opinion of the Court, and Justice Gorsuch, joined by Justices Thomas and Alito, wrote a dissenting opinion. In the majority opinion, the Supreme Court did, in fact, reject the Second Circuit’s approach. The majority assumed that the discovery rule applied because Warner Chappell never challenged the Eleventh Circuit’s use of the discovery rule, and therefore the issue was not fully briefed. The majority then turned to whether damages were time-barred if they were incurred more than three years prior to the filing of the lawsuit, and held that they are not. Justice Kagan wrote: “If any time limit on damages exists, it must come from the Act’s remedial sections. But … [t]here is no time limit on monetary recovery [in those sections]. So a copyright owner possessing a timely claim for infringement is entitled to damages, no matter when the infringement occurred.” 

The Second Circuit had relied upon an earlier Supreme Court decision, Petrella v. MGM, which we discussed more here. Justice Kagan explained that in Petrella, the plaintiff had longstanding knowledge of the defendant’s infringements, so the discovery rule could not be invoked, and there was no way for the plaintiff to bring timely claims for infringements occurring more than three years prior to filing suit. As such, the Court found that the Copyright Act’s statute of limitations allowed a plaintiff “to gain retrospective relief running only three years back from” the filing of a suit. “Taken out of context,” Kagan wrote, “that line might seem to address the issue here. But that statement merely described how the limitations provision worked in Petrella. The Court did not go beyond the case’s facts to say that even if the limitations provision allows a claim for an earlier infringement, the plaintiff may not obtain monetary relief.” Here, because the plaintiff’s claims were timely under the discovery rule, the majority held that damages accruing more than three years prior were still on the table.

At oral argument, several justices questioned whether the Court should even be reviewing the Warner Chappell case, at least pending the resolution of another case pending before the Court, Hearst Newspapers, LLC v. Martinelli.  Unlike Warner Chappell, the Martinelli case directly raises the question of whether the discovery rule applies in copyright cases. In his dissenting opinion, Justice Gorsuch wrote that the question of the discovery rule’s application needed to be decided before taking on the question of whether a temporal damages bar applies in copyright infringement cases.

The Warner Chappell decision is being hailed as a victory for copyright owners.  Nevertheless, it remains to be seen how the Court will decide the Martinelli case, and whether it will hold that the discovery rule does apply to copyright actions.  Based on the majority opinion in Warner Chappell, it is likely that the Court will hold the discovery rule applies generally in these types of cases. However, we will not know for sure until the Court addresses Martinelli later this term.

Before we start the show, the casts of characters may drive you Off the Wall, so here’s a little primer to help you:

Triumph International, Inc. (“Triumph”) is a merchandise licensing company for the estate of Michael Jackson.

MJJ Productions, Inc. (“MJJ Productions”) is a California-based record label and the custodians of Michael Jackson’s estate.

MJL 12, LLC (“MJL 12”) is a third-party company behind the “MJ Live” tribute act in Las Vegas.

Now let’s begin…

On July 2, 2019, Triumph moonwalked into the USPTO with Application Serial No. 88497689, hoping to secure a registration for the mark “MJ” for use in connection with a medley of services including live and televised performances that span music, dance, drama, and comedy. This mark was filed under the “intent to use” section of the Lanham Act, signaling a future where the mark would be used in the commercial arena.

Triumph Inc. didn’t Stop ‘Til They Got Enough, obtaining five extensions to show actual use of the MJ mark in commerce. On December 19, 2022, Triumph filed a specimen of use with the USPTO: a picture of the marquee for MJ THE MUSICAL shining bright like a sequined glove at the Neil Simon Theatre in New York City, alleging use dating back to October 2010. On February 28, 2023, the USPTO granted U.S. Trademark Registration No. 6,991,973 to Triumph. Interestingly, the mark has not yet been proclaimed the King of Trademarks!

With registration in hand, MJJ Productions really Wanted to Be Starting Something. The battle lines were drawn when MJJ Productions told MJL 12 to Beat It, claiming trademark infringement and violation of the right of publicity through use of the King of Pop’s likeness. This really put MJL 12 in a Jam. Would MJL 12 accept defeat without letting any Blood on the Dancefloor, or would they tell MJJ Productions to Leave Me Alone?

MJL 12 decided to Rock With MJJ Productions to see who’s Bad. Specifically, MJL 12 retaliated with a lawsuit against MJJ Productions and Triumph in Nevada, seeking declaratory judgment in its favor regarding trademark infringement and right of publicity.  This legal Thriller challenges Triumph’s MJ trademark and seeks a declaration that MJL 12’s “MJ Live” show does not infringe the MJ mark and does not constitute a violation of the right of publicity. The drama intensifies with allegations of “MJ THE MUSICAL” and “MJ LIVE” being confusingly similar, leaving fans puzzled, whispering Who Is It?

In its pleadings, MJL 12 asks the court to Remember the Time. Since 2012, their “MJ Live” logo has been a constant feature, with over 3,800 performances under their belt. MJL 12 also tries to undermine Triumph’s registration for MJ, challenging the claimed first use date in 2010 in light of the 2022-dated specimen submitted with Triumph’s application and the start of MJ the Musical’s marketing around 2019 or 2020. This discrepancy will make anyone Scream!

Navigating the intricacies of these intellectual property law arguments is not as Black or White as it seems. This lawsuit dances on the edge of legal nuances at the intersection of trademark law and the right of publicity. With time-sensitive defense strategies at play, the case, MJL 12 LLC v. MJJ Productions LLC, spins into a legal showdown, leaving everyone in suspense, wondering who will emerge as the Man in the Mirror, claiming Victory in this battle over the legacy of the King of Pop.

A recent decision from the Trademark Trial and Appeal Board may make bars and alcoholic beverage brands think twice about their trademark selections. In In re Caymus, the Board upheld a refusal to register TABLEAU, based in large part on an inherent relatedness between wine and bar services. Exparte Appeal No. 97040804 (TTAB February 22, 2024). The Board’s opinion is silent on whether other alcoholic beverages and bar services may be considered related, but suggests that is likely the case. 

Caymus Vineyards applied to register its TABLEAU mark for “wine” in Class 33. The USPTO Examining Attorney refused registration of Caymus’ mark based on the prior-registered, identical mark TABLEAU for “restaurant and bar services; cocktail lounges” in Class 43. Caymus appealed the decision to the Board alleging, among other things, that the Examining Attorney should have required a showing of “something more.” The “something more” requirement is typically applied in the context of determining whether food and beverage products and restaurant services are related. See TMEP § 1207.01(a)(ii)(A). Caymus also argued that the Examining Attorney’s reliance on the infinitesimally small percentage of restaurants that offer wine under the same mark as their restaurant services was insufficient for relatedness. The Board did not table this case and swiftly affirmed the refusal.

The Board noted that because there was no dispute over the fact that the parties’ marks are identical, a lesser degree of similarity between the goods and services was required to support a likelihood of confusion. Further, it rejected Caymus’ argument that “something more” has only been found in three instances: (1) when the applicant and registrant specialize in the same cuisine; (2) the registered mark is very unique; and (3) a registrant’s wine is sold at an applicant’s restaurant.

The Board noted that, in addition to a lack of authority to support Caymus’ position, Caymus’ arguments failed to consider the entirety of the registration’s identification of services. The registration not only covered “restaurant services,” but “bar services” and “cocktail lounges.”  Because wine is often sold at bars and cocktail lounges, the Board found an inherent relationship between these goods and services. As such, the Board did not require a showing of “something more” to establish relatedness. Even so, the Board went on to analyze whether the “something more” standard was reached in this case and it confirmed that it was, based on the Office’s submission of 11 third-party registrations covering wine and restaurant services, examples of more than 10 third-party entities that offer wine and restaurant services, and at least 8 articles discussing entities that offer both wine and restaurant services. 

This decision appears to fall in line with other decisions in which the Board has found a likelihood of confusion where marks are identical or highly similar and the goods and services at issue are alcoholic beverages and restaurant or bar services. See e.g., In re R.S. Lipman Brewing Company, LLC, Serial No. 88209633 (TTAB May 3, 2023) (affirming a refusal of CHICKEN SCRATCH for beer in view of an identical mark for restaurant services).  But the Board is not always consistent. See e.g., In re 1729 Investments LLC, Serial No. 90694523 (TTAB April 24, 2023) (reversing refusal to register RAO’S for wine not sold in restaurants in view of the identical mark for bar and restaurant services); and In re Iron Hill Brewery, LLC, Serial No. 86682532 (TTAB July 28, 2017) (reversing refusal to register THE CANNIBAL for beer in view of CANNIBAL for restaurant services). It is important to note that this case is not precedential. Because the Board disfavors citing non-precedential opinions, the impact of this case is uncertain. However, dicta in the opinion does suggest that the Board would adopt a similar stance for other types of alcoholic beverages, indicating that wine is just one subset of the types of beverages that would be found related: “[I]t is clear that bar services and cocktail lounges, such as Registrant’s, focus on serving alcoholic beverages such as Applicant’s wine.” Therefore, purveyors of alcoholic beverages should ensure that when selecting a new trademark, they have evaluated whether any prior trademarks registered or used in connection with bar services might conflict with their chosen mark. Similarly, bar owners should consider whether their brand is too close for comfort to an alcoholic beverage brand.  If not, an unwanted tableau of consumer confusion may be created.

In the wake of several Congressional hearings over the past year on AI and intellectual property, Representative Adam Schiff (D-California) has introduced the Generative AI Copyright Disclosure Act of 2024 (H.R. 7913).  The proposed law addresses concerns over lack of transparency in the data sets used to train generative AI models by requiring submission of a notice to the United States Copyright Office regarding all copyright-protected works used to train a particular AI system prior to the release of that system.  Although the bill has almost no chance of gaining traction in Congress, it showcases concerns that current U.S. copyright law may not adequately address new generative AI technologies.

Act Particulars

The Act’s notice requirement would apply to those who create or significantly alter “a training dataset,” which the bill defines as “a collection of individual units of material (including a combination of text, images, audio, or other categories of expressive material, as well as annotations describing the material) used to train a generative AI model” and build a generative AI system.  The notice would need to include a “sufficiently detailed summary” of any and all copyrighted works—whether registered or not—used in the training dataset or to alter the training dataset.  Moreover, if the dataset is publicly-available when the notice is submitted, the notice would need to include the URL where the dataset appears.  If a required notice is not submitted pursuant to certain deadlines set forth in the Act, violators would be a subject to a civil penalty of at least $5,000.  The Act would take effect within 180 days of enactment, and would require the U.S. Copyright Office to issue implementing regulations within that same timeframe.  The Copyright Office would also need to establish a publicly-accessible online database consisting of each notice filed.

Stakeholders’ Perceptions

The interests of those creating or altering the generative AI datasets are no doubt at odds with those of copyright owners.  We predict that, generally, copyright owners’ perceptions of the Act will be very positive, while generative AI companies will have a very negative reaction.  This is not surprising given the challenges generative AI companies would face in tracking down individual rights holders, and then, once the rights holders are on notice of what is going on, likely having to obtain formal licenses and pay them royalties before incorporating their works in training datasets.  Such requirements could be viewed as chilling technological growth and promoting AI monopolies, since, for example, smaller or niche companies would likely be unable to pay the requisite costs.  The effects could also have a negative spillover effect on rightsholders because any royalties paid to them would be very low even for financially successful AI models, given that the royalties would need to be spread among millions of creators.  Moreover, AI companies may focus on compensating larger copyright holders, not only because they own the rights to a large volume of works, but also because their rights are likely more well-known or easily discovered by generative AI creators or editors.  As a result, individual creators’ rights may be overlooked, even if generative AI companies make good faith efforts to identify and pay them for use of their protected works.   

On the other hand, many rightsholders currently feel as though AI companies are unfairly profiting from their works without permission and without compensating the rightsholders, despite copyright protections being enshrined in the U.S. Constitution.  As early decisions in AI-related copyright litigation indicate (see our prior coverage here), copyright owners may have a difficult time preventing the unauthorized use of their works to train AI models under the current version of the U.S. Copyright Act.  Accordingly, many content creators and rights owners seem to be embracing the bill as a step in the right direction to avoid unfair infringement and provide protection that may be otherwise difficult to achieve.  From the perspective of copyright owners, protecting their works encourages the creation of new works—after all, if you create a work, and then someone can copy it and profit it from it, without paying you, why spend the time to create a new work?  This, they argue, is the reason that copyright protection exists in the first place. 

What Does the Future Hold?

While strong opinions have been voiced on both sides of this issue, it bears noting that the Act has no co-sponsors to date and is unlikely to receive much consideration in Congress.  Even so, the bill reflects the fact that there is significant public support, particularly among writers, artists, and other creatives, for copyright protections that address the advent of generative AI.  But it also highlights the thorny issues that come with attempting to regulate new technologies, including the possibility that regulations could inadvertently stifle new technological development.  This is a debate that will continue not only in Washington, D.C., but throughout the country.

In the meantime, it remains to be seen how courts addressing AI-related issues will decide.  Those cases may provide guideposts for Congress to further draft and propose an AI act that would better suit the involved parties.

Jay Myers, Seyfarth partner and Director of Innovation for the Intellectual Property Group, joined Arun Hill, senior consultant at The Clarivate Center for IP and Innovation Research, as a guest in Clarivate’s most recent podcast episode, “Ideas to Innovation: Navigating the AI Frontier in Intellectual Property Law.” Clarivate, a global leader in providing solutions to accelerate the lifecycle of innovation, helps customers solve some of the world’s most complex problems. The podcast episode covers the transformative influence of artificial intelligence (AI) on intellectual property (IP) and trademark law. It offers valuable insights into how organizations can leverage AI to navigate the complexities of IP law, striking a balance between innovation and the critical human element of legal expertise. Follow the links below to listen in and explore how AI is not only automating tasks, but also augmenting the capabilities of IP professionals to achieve higher-quality outcomes for their clients.

Podcast Links:

Clarivate

Amazon

Apple

Spotify