Chobani's "zero sugar" yogurt with allulose violates FDA regs despite FDA's nonenforcement; 7th Circuit reverses preemption ruling

Franco v. Chobani, LLC, No. 25-2087 (7th Cir. Jul. 27, 2026)

Federal law requires that foods advertised as sugar free contain less than a half gram of sugar. Chobani sold Chobani Zero Sugar Yogurt, but it included four grams per serving of allulose, a naturally occurring sweetener. If allulose is a sugar under federal law, Franco’s state law false advertising claims could proceed, but if it wasn’t, then there was express preemption because federal law doesn’t allow states to impose additional requirements on food labeling regulated by the FDCA. The district court, deferring to FDA enforcement guidance, found preemption. The court of appeals reversed, finding allulose to be a sugar under the regulation.

The relevant regulation defines “[t]otal sugars” as “the sum of all free mono- and di-saccharides (such as glucose, fructose, lactose, and sucrose).” And a food may not be labeled “sugar free” or “zero sugar” (or similar terms) unless it “contains less than 0.5 g of sugars, as defined by [the previous regulation]” and meets other requirements.

Hello, Loper Bright! “Interpreting the law is a job for the court,” though the court isn’t required to ignore the FDA’s perspective about the meaning of its regulations, and deference to agency interpretation of an ambiguous rule “can be appropriate.” The FDA has several times expressed a view: in 2016, it observed that “the final rule does not reach a decision as to whether Allulose should be excluded from the [definition] of sugar[] … , and Allulose, as a mono-saccharide, must be included in the [Total Sugars] declaration … pending any future rulemaking that would otherwise exclude this substance from the declaration.” In 2020, it issued industry guidance advising “manufacturers of [FDA’s] intent to exercise enforcement discretion for the exclusion of allulose from the amount of ‘Total Sugars’ and ‘Added Sugars’ declared on the label … pending review of the issues in a rulemaking.” (The issue appears to be that the FDA traditionally used chemical structure to identify sugars, but novel sugars might also count depending on factors such as an association with dental caries and how they are metabolized in the body.)

No further rulemaking has occurred, but the court of appeals called for the views of the FDA. In its resulting amicus brief, the agency took the position that total sugars as defined in that regulation include all monosaccharides, including allulose. The “such as” parenthetical at the end of the regulation was “merely a list of non-exhaustive, illustrative examples, and not (as the district court found) a limitation on sugars based on the physiological characteristics that the listed sub-stances shared.” And an enforcement position isn’t an interpretation of a regulation. The court of appeals found this persuasive. “There’s no dispute that allulose is a monosaccharide. Because the definition includes every monosaccharide and the following parenthetical is merely a list of examples, allulose is a sugar.”

The surplusage and noscitur a sociis canons didn’t change anything. Chobani argued that a “sugar” needed the same nutritional characteristics as glucose, fructose, lactose, and sucrose. The surplusage canon suggested that the “such as” parenthetical should have some meaning, but redundancy is common in the law, and “such as” doesn’t always mean “of the same kind”;  it can merely introduce “examples of a class.” “The FDA defined a class by way of chemistry; it reinforced that definition through examples, all of which share the same chemical structure.” There’s no need to derive a definition from the list of examples, since they merely illustrate the definition that already appears. Nor did Chobani argue that a definition of sugars that includes all monosaccharides will lead to absurd results. “And in defining total sugars, the FDA specifically invoked the language of chemistry, which means decisions favoring common parlance meanings aren’t persuasive, either.” The FDA could have defined total sugars based on physiological factors, rather than chemical makeup, but it didn’t do so.

The regulation wasn’t ambiguous, so there was no reason to defer to the FDA’s enforcement guidance, which wasn’t an official position anyway.

What about the marketing permit Chobani secured from the FDA? “[T]he fact that one sovereign (the United States) indicated that it would not enforce its labeling requirements with respect to allulose should not have led Chobani to believe that the states would take a similar approach. Similarly, … the agency’s marketing permit said nothing about state law consumer protection suits. Chobani is a sophisticated actor and should have been aware that the FDA’s decisions about its enforcement priorities would not immunize the company from suits based on state law.”

Chobani’s additional preemption theory based on Monsanto Co. v. Durnell, 609 U.S. ---, 2026 WL 1825691 (June 25, 2026), could be handled on remand by the district court.

Deception was also plausible. Chobani argued that consumers don’t care about the existence of monosaccharides in their food but are instead concerned with avoiding the ad-verse health consequences associated with traditional sugars. “Whether reasonable consumers care about the existence of allulose in their yogurt isn’t the same thing as asking whether reasonable consumers would be deceived by it.” Discovery was the right next step.

false advertising in online games leads to over $700 million award to competitor

Skillz Platform Inc. v. Papaya Gaming, Ltd., 2026 WL 2151126, No. 24cv1646 (DLC) (S.D.N.Y. Jul. 27, 2026)

Some previous opinions (more linked there). At a jury trial, Skillz won a substantial damage award for false advertising about Papaya’s use of bots in the real money skill-based mobile gaming (RMSB) market. The court here rejected Papaya’s post-trial motions and Skillz was awarded $719 million plus its attorney’s fees for the years 2024 and 2025 and certain costs.

“Skillz created the first RMSB platform in 2012. Papaya entered this market in 2019 and quickly took a significant market share.” In the light most favorable to Skillz, the evidence at trial showed significant barriers to entry in this market, because a successful platform “needs a customer base large enough to match players of similar or comparable skill in tournaments within a reasonable amount of time,” aka liquidity. Skillz maintains that its tournaments are games of skill, not chance, and thus not gambling; it was required to and did represent to app stores, payment processors, and advertising hosts that it was not engaged in gambling and that the outcome of its tournaments was determined by the skill of the players.

Then, Papaya

decided to enter the RMSB market in 2019 by deceiving the public and others about the nature of its product. Instead of running tournaments in which the players who had paid entry fees competed against each other, Papaya decided to solve the problem of building liquidity by running tournaments in which multiple participants were not other human players but were instead “bots.” Papaya’s bots were essentially scores designated by Papaya’s algorithms; they were not artificially intelligent players. In these tournaments, a player’s score was compared to the scores Papaya assigned to its bots. In this way, Papaya was always able to run a tournament, including tournaments with what appeared to be a dozen or more participants, at any time of night or day. Papaya gave its bots usernames and profiles to make them appear to be individual customers.

Papaya used liquidity bots to create immediately accessible tournaments of up to 20 or more “players,” permitting a player in a Papaya game to learn quickly whether he had won or lost a tournament, “thereby increasing the odds that he would pay to enter another Papaya tournament.” It also used “tailored” win/loss bots so that a player who had a losing streak could be given a “win” to motivate them to keep playing in more tournaments. Tailored bots operated in over 630 million Papaya tournaments, or in roughly one-quarter of the 2.6 billion tournaments that Papaya hosted 2021-2024. During that period, “bots accounted for over 13 million of the participants on Papaya’s platform, compared to about 11 million human players.” Over half of humans played in at least one tournament where tailored bots were designed to give them a loss, and more than 7.3 million played in tournaments with tailored bots designed to give them a win.

Unsurprisingly, Papaya’s bots were used most heavily in the initial phases, when it needed to build liquidity, about 90% of the time in 2021. “Because of its use of bots, Papaya only paid customers roughly $2 billion of the $6.7 billion that it advertised had been awarded in prizes. And just before Papaya stopped using bots near the end of 2023, Papaya was still using bots in roughly 50% of its cash tournaments.” Papaya “achieved a substantial presence in the RMSB market while investing only a fraction of the money expended by Skillz to do so. When seeking investors, it bragged about its strategy” of multiple-player tournaments, “unique” platform capabilities, and revenue growth “8 times more than the industry leaders.”

Papaya never disclosed its use of bots, but purposely advertised falsely that its games were “fair” and “skill-based,” that it has “no vested interest” in who wins or loses a tournament, and “described the participants in its tournaments with pictures and in terms that apply to human players.” When players complained, Papaya denied using bots, at the direction of executives.

Papaya made similar misrepresentations to app stores, its payment processers, and its advertising channels.

Papaya did not stop using bots until late 2023, at which point Skillz’s revenue had fallen by 60% in just two years: from $384 million to $152 million, “while Papaya’s revenue skyrocketed from $163 million to $461 million over the same period.”

Given the “overwhelming” evidence of intentional false advertising, the parties principally litigated damages before the jury, and also the court sought an advisory verdict on disgorgement. The jury was instructed that “Skillz is not entitled to duplicative monetary recoveries and the Court will ensure that Skillz only recovers once for any injury it has shown it suffered.”

The jury awarded Skillz $420 million in damages, or two-thirds of the requested amount, and an advisory disgorgement verdict of $719 million for Papaya’s unjust profits and $652 million for Papaya’s unfair cost savings (in acquiring users through fake liquidity).

Papaya argued that Skillz’s damages expert improperly relied on a but-for world in which Papaya did not use bots and its advertising statements were true, making its expenses greater and its profits lower because it did not rely on bots to build its business. Papaya argued that the expert was required to model a different but-for world, “specifically one in which Papaya removed all false statements from its advertising while not altering its business model or its use of bots.” But Papaya didn’t explain how this could be done: “Skillz proved at trial that Papaya could not have entered the RMSB market by using bots in the way that Papaya did … and at the same time truthfully describe that use and its product to consumers in its advertising. Papaya has offered no authority to suggest that a plaintiff’s expert must create a but-for world that could not exist.” Among other things, payment processors and others wouldn’t have allowed Papaya to use their services had Papaya told the truth, and consumers wouldn’t have wanted to play against bots.

Given these constraints, Skillz showed that the damages “flowed directly from Papaya’s false advertising,” which “concerned the very nature of the product; it was not a false statement about some incidental feature. And Skillz showed that it was that very advertising that caused consumers to complain to Papaya about Papaya’s use of bots.”

It was also acceptable to award damages for lost enterprise value as long as the damages were measurable with reasonable certainty. Papaya argued that Skillz could only recover its lost profits, but Skillz hadn’t yet made a profit at the time of trial; “Papaya argues that Skillz cannot recover enterprise value damages simply because Skillz prioritized growth over profits in the years it was developing its business.” Yes, Skillz “invested heavily in developing its business and enlarging its customer base to achieve not only substantial liquidity but also a network effect,” but that didn’t limit it to lost profits instead of lost enterprise value.

Nor was Papaya entitled to JMOL on the Lanham Act and NY GBL claims: falsity, materiality, and harm were all sufficiently shown. Among other things, “there was abundant evidence that Papaya engaged in deliberate conduct ‘of an egregious nature’ to deceive consumers, which created a presumption of deception.”

The court additionally rejected Papaya’s argument that damages under the GBL must be limited to financial harm that resulted from Papaya’s deception of New York consumers. As a competitor suing for the effects of consumer deception on it, “Skillz is entitled to be fully compensated for the injury it incurred through Papaya’s wrongdoing even though that injury also impacted consumers who resided outside New York.”

The court also rejected Papaya’s other challenges to the amount of damages. The amount of the award didn’t shock the conscience when measured against the legal standard and the trial evidence. “Skillz and Papaya were competing with each other in a new online industry where billions of dollars in revenue were available to the successful RMSB company. Papaya’s fraudulent conduct was extraordinary” and central to its huge success/zero-sum impact on Skillz.

Among other things, it was ok to use evidence of Papaya’s success after it stopped using bots: “There was credible evidence at trial that Papaya benefitted from the network effects of its false advertising even after Papaya ceased using bots to fill and control tournaments.” Thus, the damages model could include Papaya’s revenue from running tournaments for the player base that it had built while engaging in false advertising.  Papaya’s own damages expert testified that the number of Papaya’s tournaments fell only 20% after it turned off the bots, “suggesting that Papaya continued to benefit from the liquidity it had built through its false advertising.”

Skillz requested that the court double the jury’s advisory verdict of $719 million on Papaya’s profits (the jury also identified $652 million as Papaya’s cost savings), or treble the $420 million damages award. The court declined both requests for enhanced damages, but did award $719 million.

Disgorgement was appropriate because, come on. Skillz didn’t show that disgorgement was necessary to deter Papaya, since it largely ceased its false advertising in late 2023 by discontinuing the use of bots in its tournaments, “and there is no realistic possibility that Papaya will return to its false advertising campaign now that the unlawful advertising has been publicly revealed and addressed in this judgment.” The key was unjust enrichment: “Papaya’s relatively modest financial investment in its start-up business did not explain its explosive growth …. Even large, deep-pocketed U.S. companies that had contemplated entering the market to compete with Skillz, which had already achieved a network effect, decided against doing so.” The award of Skillz’ loss of enterprise value didn’t fully deprive Papaya of the benefits of its illegal scheme. “Papaya is still a substantial player in the market and, because players tend to stay on a platform with which they are familiar, the impact on Skillz of Papaya’s wrongdoing will continue for the foreseeable future.”

Nor did Skillz delay unreasonably: “While Skillz came to suspect and then believe in 2023 that Papaya’s success was due to its false advertising and employment of bots, Skillz was entitled to sufficient time to develop the reasonable grounds necessary to plead its claim in federal court in early 2024. Since that filing, Skillz has proceeded with diligence to prosecute this lawsuit, despite Papaya’s strategy of making Skillz’s assembly of the proof of its claims difficult and expensive.”

However, if no disgorgement award were available, the court would enhance the actual damages award by doubling it to $840 million, because the actual damage to Skillz was “severe but hard to quantify,” especially given that Skillz made changes in response to Papaya’s false advertising.  

Papaya also argued that any disgorgement amount should be reduced by 13% to remove non-U.S. revenue, because the Lanham Act is not extraterritorial. But Papaya’s audited financial statements, introduced at trial, did not disaggregate U.S. sales from foreign sales. Papaya relied for its 13% number on a single graphic in a January 2023 PowerPoint presentation prepared by a third-party for Papaya, apparently to entice investors, while it was still engaged in false advertising in the United States. The presentation disclaimed being an “audit or due-diligence review”; it warned that the author does not give “any representation or warranty, express of implied, as to the accuracy or completeness of the information” in the document.

The court declined to adjust the figure. The US was undisputedly Papaya’s target market, and  “Papaya denied Skillz access during the discovery period to relevant information that Skillz sought, including the information that would permit Skillz to accept the representation in this graphic, to dispute it, or to place it in context.” On this record, there was no need to adjust disgorgement, which after all need not be proven to perfection.

Unsurprisingly, the court also partially granted Skillz’s requests for attorneys’ fees, costs, and post-judgment interest, but not pre-judgment interest. The case was exceptional (the Lanham Act standard) and the court exercised its discretion to award fees under the GBL. Ending the fee period at the end of 2025 left roughly $10.1 million in fees. Along with the deliberate, extensive, central deception, Papaya litigated the case unreasonably in 2024-2025, when it “slow-walked and obstructed the production of critical discovery material through the entire discovery period,” with ramifications through trial. “For example, during the presentation of the defense case, Skillz learned for the first time that Papaya had wrongfully redacted highly relevant passages from its documents by marking the material ‘nonresponsive.” But “[t]his year, Papaya changed counsel and the parties were largely involved in preparing for the April trial. From any point of view, it was reasonable for Papaya to litigate the amount to be awarded in damages and that was largely the focus of the trial.”

No prejudgment interest because the attorneys’ fee award was sufficient to address the exceptional nature of the case, and because the court had already ordered disgorgement of Papaya’s unjust enrichment.

Skillz Platform Inc. v. Papaya Gaming, Ltd., 2026 WL 2185762, No. 24cv1646 (DLC) (S.D.N.Y. Jul. 29, 2026)

The court also denied a permanent injunction. First, Skillz sought an injunction against bot use; that was denied because it was false advertising, not bot use as such, that was at issue.

Second, Skillz sought a 6-month corrective advertising “splash screen” shown to each user prior to their playing any Papaya game or when they enter the Papaya website, informing readers that Papaya once used bots, that it denied their existence to customers, and that a jury found Papaya liable to Skillz for false advertising. [Annoying all-caps presentation makes the proposal unreadable.] Although Skillz suffered an irreparable injury that monetary damages could only partially remedy, the public interest and the balance of the equities didn’t weigh in favor of the proposal:

There is no adequate showing regarding how many of those currently viewing the Papaya website or playing a Papaya game were subjected to its false advertising. Papaya’s false advertising ended in late 2023 with its removal of bots as tournament players. That is over two years ago, which is probably an infinity in the online gaming world. Moreover, the reference to Skillz in the proposed corrective advertising will promote Skillz, which is only one of Papaya’s competitors. Finally, the burden on a consumer of reading and reacting to the statement years after the events at issues weighs against the requested relief.

warnings about effects of "killer acquisitions" on Shopify apps aren't about tangible characteristics and aren't factual claims

Jika Inc. v. Loop Solutions, Inc., No. 2:26-CV-05530-JAK (MARx), 2026 WL 2138604 (C.D. Cal. Jul. 20, 2026)

Jika, d/b/a Skio & Recharge, sued Loop for false advertising/tortious interference/unfair competition/trade libel for ads it ran in response to Recharge’s acquisition of Skio. [Relatedly.] The court denied a preliminary injunction.

Recharge, Skio, and Loop each provide apps for e-commerce subscription management on the Shopify platform to enable Shopify merchants to turn one-time retail customers into long-term subscribers. According to its advertising materials, Recharge “power[s] 71% of subscriptions sold on Shopify stores.” Skio provides similar services for more than “one thousand brands,” including various “well-known direct-to-consumer” brands. It offers a “subscription management platform” that enables Shopify merchants to place a “subscribe and save” widget on their product pages. Skio’s app is “integrate[d] with Shopify’s native checkout and subscription” services. E.g., Skio offers merchants a configurable flow so that when a subscriber clicks “cancel,” the merchant can offer discounts, free gifts, pauses, product swaps or different frequencies to try to keep them; it also provides dashboards that provide information about their subscriptions, including how subscriptions are growing, what the churn rates are, and which cancellation offers are working.

Loop “offers many of the same products and services to the same category of merchants” as Skio, including “subscription management, customer portals, cancellation flows, dunning and payment recovery tools, bundle builders, and API integrations.” The parties agreed that Loop and Skio compete, but not that Loop and Recharge did, though there was evidence that customers view Recharge, Skio, and Loop as interchangeable and competing services. “For a large business processing 10,000 subscription orders per month, with an average order value of $50, subscription-management costs would be $7599 per month for Skio, $9099 per month for Recharge, and $4149 per month for Loop.”

Loop’s declarant stated that its customers are “sophisticated market players” who typically “engage in thorough research regarding” platform choices. He declared that it “can take weeks or months of back-and-forth negotiation” before a merchant chooses to use Loop for its subscription-management needs. Skio’s declarant said things consistent with this, e.g., one potential Skio merchant evaluated Skio for a month before making a purchasing decision, and the pricing was consistent with requiring reasonable consumers to be careful.

Loop also submitted evidence of customer dissatisfaction with Recharge’s app: more than 5% 1-star reviews online, generally concerning Recharge’s pricing and customer service. “Skio’s Shopify reviews reflect that many Skio merchants reported that they elected to migrate from Recharge to Skio based on problems with Recharge’s app.” Loop has less than 0.75% 1-star reviews out of 670 total reviews.

Shopify bans app developers from publishing an app that is “identical to other apps you’ve published to the Shopify App Store.” “Recharge has acknowledged that its acquisition of Skio could raise questions about the application of these policies.” A Loop affiliate’s declarant stated that it is “very common for acquiring companies to consolidate or eliminate products that are redundant with their own,” citing industry and academic publications. He specifically identified “killer acquisitions” in the Shopify industry. When the acquisition was announced, Recharge and Skio stated that nothing would change immediately.

But some worried, and Loop fed that worry despite Recharge’s “no change until 2028” guarantee stating “no Skio merchants will be forced to migrate to Recharge” and that there’d be no pricing change.

Loop argued that it merely “used the market uncertainty created by the acquisition” as an opportunity “accurately [to] inform[ ] merchants” about their choices post-acquisition and what “typically happens in acquisitions of technology platforms.”

“Shortly after the acquisition announcement,” Loop “posted a site-wide banner” across its website: “Recharge acquired Skio for $105M. Loop is now the second largest Shopify subscription app. Your choice just got simpler: Loop or Recharge. Book your priority migration slot.” Plaintiffs argued that this was false because it informed customers that “Skio is no longer a viable standalone platform” when, in fact, “Skio continues to operate as a standalone product,” its customers are “not being migrated to Recharge” and there are “no plans to deprecate Skio.” It made similar statements on its blog:

Recharge’s official position: both platforms continue to operate as normal. Nothing changes immediately.

But here’s what historically happens when a platform gets acquired by its largest competitor:

The acquiring company says “business as usual” for 6-12 months. Then feature roadmaps merge. Then pricing consolidates — usually upward. Then the smaller platform’s app gets sunset or rolled into the acquirer’s product. The merchants who waited get migrated on the acquirer’s timeline, not their own.

Skio merchants now face a set of questions nobody has answers to yet. Will Skio’s $599/month pricing stay? Will the passwordless login and clean portal UX survive integration into Recharge’s architecture? Will the small, responsive support team that Skio reviewers praised remain intact — or get absorbed into Recharge’s support infrastructure, which its own reviewers have documented as slow and unresponsive?

What that uncertainty looks like in practice depends on where you stand today.

If you’re currently on Skio:

Your platform’s future is now controlled by Recharge.

...

If you’re currently on Recharge:

The same pricing escalation and support patterns documented in 116 one-star reviews haven’t changed. They’ve gotten bigger.

...

If you’re evaluating platforms for the first time:

The decision just got simpler. It’s Loop or Recharge.

[The blog post went through various revisions; I smell some AI.]According to Loop, 45 unique users visited the Blog Post, of which 22 were users from the United States, in comparison to 6994 unique visitors to Loop’s website overall in the same period.

Loop also allegedly targeted Skio’s customers in outreach with similar statements.

It also used a comparison table—which it argued had been around for a while and wasn’t posted in response to the acquisition—marking numerous Skio features as unavailable (x) and those same features for Loop as available (✓).” It allegedly falsely claimed that Skio lacked features such as “Passwordless login via email,” “One-click checkout with exclusive offers,” and “Proactive card expiration alerts.” The meaning of the comparisons was contested, though Loop admitted “error” in claiming that Skio lacked “OTP-less authentication” and “Passwordless login via email.” (Other Loop ads accurately attributed a passwordless login feature to Skio.) The table was removed before litigation began. [Again, I wonder if this was AI error; it’s good that Loop doesn’t pretend that the source of the error mattered.]

Skio and Recharge allegedly “received over 10 inquiries from Skio merchants expressing concern about Skio’s future, pricing, support continuity, and feature availability” following the acquisition. One customer informed Recharge that it had “received unsolicited emails from Loop that contained no identifying signature, footer, or other disclosure indicating that the messages came from Loop.” Because the email stated that the sender “wanted to get them started on migrating,” the customer “initially believed the emails were coming from Recharge and called Recharge expressing concerns about Skio’s post-acquisition future.”  Five large merchants allegedly “reported concerns that Loop representatives had been making statements about Skio and the Recharge acquisition consistent with the messaging in” the written campaign.

Plaintiffs also submitted declarations that customers were hard to acquire. A potential customer allegedly wrote: “We were really impressed with Skio and felt a good level of alignment between Skio and [Customer]. However, the news that Skio merged with Recharge did catch us really off-guard! We felt a bit torn since we had pretty much disregarded Recharge completely by that point as we didn’t enjoy the sales process, and the pricing was completely unrealistic for our brand. At this moment, we are progressing with Loop subscriptions, as we felt they also aligned with us .…” It didn’t reference the Loop ads.

Plaintiffs argued that Loop made literally false claims that Skio would be sunsetted, leaving merchants with a binary choice between Recharge and Loop, by claiming that, post-acquisition, “smaller platforms historically ‘get[ ] sunset’ ”; by stating “Your choice just got simpler: Loop or Recharge”; and by telling customers they were “Back to square one for [their] platform future.”

First, under 9th Circuit precedent, “the nature, characteristics, and qualities of [a product] under the Lanham Act are more properly construed to mean characteristics of the good itself ....” Licensing status and claims about the date on which a product was first marketed are not actionable for this reason, and likewise the “sunsetting” statements didn’t refer to any inherent quality or characteristic of the parties’ services. “Rather, the Sunsetting Claims only refer to market structure and competitive dynamics in the industry, which are analogous to the ‘supply and demand phenomena’ that courts have found do not state a claim under § 43(a).”

Even if they were actionable in principle, nothing Loop said was literally false. The blog post truthfully reported that plaintiffs claimed that nothing would change immediately, but then pointed to “historical[]” trends. A prediction about future events “is not an actionable statement of fact as a matter of law.” Plaintiffs’ own claims didn’t show that there was no risk that Skio would be sunset; their own public-facing statements recognize that there is some uncertainty about the future of Skio: a LinkedIn post said, “[F]rankly, we don’t know what skio or recharge or (maybe even) some new platform is gonna look like in a year and a half.” Even if the rule about future predictions only applies to good-faith predictions, there was no evidence of bad faith.

For claims like “Your choice just got simpler: Loop or Recharge,” this wasn’t literally false. “One reasonable interpretation of these claims is that customers now have two choices with respect to independent service providers in the marketplace.” This was plausibly literally true: Plaintiffs’ own announcement described Skio as a “Recharge company” and states that “Skio is joining Recharge.” [I would also say that this is the kind of claim often deemed puffery under similar circumstances.] Also, because consumers typically contract with service providers for longer periods, the competitive “choice” and “decision” “can reasonably be construed to refer to long-term market structure and competitive dynamics, rather than current conditions…. Consistent with that, many of the challenged advertisements frame customer choice in terms of long-run impact, rather than immediate market conditions.” That meaning was not literally false because plaintiffs have not committed to maintain Skio as a long-term option for consumers beyond 2028.

The only “necessary” implications of the claims were that plaintiffs might not honor their nonbinding commitment to maintain Skio through 2028 and that Skio’s future remains uncertain beyond that point in time, And these were not matters of verifiable fact.

Claims that Skio would raise prices/worsen customer support: Same basic analysis. “Each advertisement, read in its full context, includes specific qualifiers,” such as “Not sure if all or even anything would happen with you but this is what generally happens.”

Comparison table: Loop conceded that the password claims were literally false, and § 43(a) is a strict liability cause of action. But there wasn’t sufficient evidence of materiality. There were no surveys or direct evidence from consumers that the features highlighted in the comparison table were ones about which they cared when making decisions. “[N]umerous courts have found an absence of materiality in Lanham Act false-advertising cases when the target audience consisted of sophisticated individuals who were unlikely to be swayed by promotional materials.”

The rest of the claims went the same way. Trade libel requires actual damage, and that hadn’t been shown with respect to the comparison table, given the absence of evidence that any specific customers were exposed to it or that they cared about the password claims as opposed to other factors.

The balance of equities and the public interest also disfavored a preliminary injunction because it would bar Loop from “suggesting” or “implying” that Skio’s product and features may change in various ways due to the acquisition. “[T]his would impose a limitation on marketplace competition that could otherwise benefit customers. It would, in effect, preclude Defendant from responding to Plaintiffs’ nonbinding commitment to maintain Skio as an independent service in the marketplace.”

Probiotic throwdown continues w/factual dispute over clinical studies

ExeGi Pharma, LLC v. VSL Pharmaceuticals, Inc., 2026 WL 2144541, No. 19-cv-02479-LKG (D. Md. Jul. 27, 2026)

This is the latest opinion in a long and tangled sequence.

ExeGi sued VSL for tortious interference with business relationships and unfair competition arising from certain letters that VSL’s litigation counsel sent to ExeGi’s manufacturers and distributors. VSL counterclaimed against ExeGi and Professor Claudio De Simone for false advertising under the Lanham Act.

The facts involve “an eight-strain, high-potency probiotic formulation.” De Simone “obtained several patents and other intellectual property rights for his formulation, including for the probiotic formulation for the eight-strain probiotic mix known as the ‘De Simone Formulation.’” He first licensed this to VSL and it was sold under the name “VSL#3.”

He also caused samples of the eight bacterial strains comprising the De Simone Formulation to be deposited with the Leibniz Institute DSMZ-German Collection of Microorganisms and Cell Cultures GmbH (DSMZ), which is a depository for the storage of strains and microorganisms, and not a depository for strains for industrial production. DSMZ assigns each deposit a unique alphanumeric designation. The VSL-De Simone relationship soured and he cut off VSL’s access to the manufacturer of VSL#3, instead granting an exclusive license to ExeGi (under the name Visbiome).

ExeGi incorporated the DSM Codes assigned by the DMSZ on Visbiome’s packaging and packaging inserts, alongside the genus and species name for each corresponding bacteria strain, although Visbiome has never been manufactured with the deposited material at DSMZ. Visbiome contains all eight of the bacterial strains, by genus and species, listed on Visbiome’s packaging. But VSL contends that testing of the DSMZ deposits reveals that at least three of the deposited strains do not match those listed on Visbiome’s packaging. The De Simone parties argued that any mismatch was the result of contamination introduced before the strains were deposited at DSMZ, and not a discrepancy in the Visbiome product itself.

Previous litigation resulted in an injunction against any claims in VSL#3 promotional materials “that state or suggest a false continuity between Italian VSL#3 and the De Simone Formulation, including but not limited to statements claiming that VSL#3 continues to contain the ‘original proprietary blend’ or the ‘same mix in the same proportions.’ ” VSL was also enjoyed from citing any clinical study performed on the De Simone Formulation or implying that any such study was conducted on Italian VSL#3, because they weren’t the same mix. [The first part of this sentence seems to me to offend the First Amendment.]

VSL’s litigation counsel sent a letter to counsel for Danisco—the manufacturer of Visbiome—notifying Danisco of a “recent judicial decision that directly impacts the legality of the continued sale and distribution of” Visbiome, demanding that Danisco take steps to “immediately cease and desist from the production, distribution and sale of Visbiome® unless and until any reference to the DSM Codes is removed from its packaging and packaging inserts, and from any marketing or reference materials for Visbiome®.” He claimed that any continued use of the DSM codes “would constitute a misappropriation of valuable property rights” and constitute false advertising.

VSL argued that ExeGi’s business was not impacted by this letter, because Danisco continued to fill Visbiome purchase orders for ExeGi and did not modify its supply agreement with ExeGi. ExeGi argued that the letter forced it to expend substantial resources and caused lost sales.

VSL’s counsel sent an essentially identical letter to one of ExeGi’s Visbiome sellers and distributors, and also to Amazon.

Tortious interference in Maryland requires: “(1) intentional and [willful] acts; (2) calculated to cause damage to the plaintiffs in their lawful business; (3) done with the unlawful purpose to cause such damage and loss, without right or justifiable cause on the part of the defendants (which constitutes malice); and (4) actual damage and loss resulting.” Relatedly, common-law unfair competition requires deception.

Although it was close, the court didn’t find that Maryland’s absolute litigation privilege bars ExeGi’s tortious interference and unfair competition claims. To be entitled to that privilege, VSL needed to show, among other things, that the letters that VSL’s litigation counsel sent to Danisco, Pharma Holdings and Amazon bore a rational relationship to litigation that VSL contemplated in good faith and under serious consideration. It did not provide such evidence, even though there was “a long and well-documented history of the disagreement and competition between the parties in this case related to the marketing of VSL#3 and Visbiome.”

However, ExeGi couldn’t show any damages. Danisco continued to manufacture Visbiome, and Pharma Holdings and Amazon continued to sell Visbiome after receiving the letters. There was no expert testimony about any damages. The cost of affirmatively suing for tortious interference couldn’t count, and there need to be actual damages for punitive damages to be available.

False advertising under the Lanham Act:  

VSL alleged that the De Simone Parties violated the Lanham Act by claiming that “Visbiome contains the strains deposited at DSMZ”; by stating that the current version of VSL#3 contains only seven strains of bacteria, rather than eight (De Simone also wrote that “it contains bacteria that are qualitatively and quantitatively different” and that this “counterfeit product has been falsely marketed and promoted to the public and medical community....”); by stating that VSL#3 lacks clinical studies (“has not been the subject of any published, peer reviewed, controlled trials in humans”; “untested imitation product”; etc.); by stating that VSL#3 is an imitation or knock off; and by stating that VSL#3 has been discontinued or recalled (“VSL#3 is no longer available for sale at many retailers”).

Most of these claims failed at summary judgment, though there was a factual question based on the statements that VSL#3 lacks clinical studies.

Statements about the strains deposited at DSMZ: VSL didn’t show that the De Simone Parties claimed that “Visbiome contains the strains deposited at DSMZ.” The presence of the alphanumeric DSM codes didn’t convey a false message, because there was no showing that a consumer: (1) would recognize that the letters “DSM” as referring to the Leibniz Institute DSMZ; (2) understand that the accompanying code is a scientific accession reference to a microbial deposit at the DSMZ; and (3) know that DSMZ maintains physical samples of that deposit; and then assume that the listing of the code means that the product contains material sourced from that particular deposit. VSL’s own expert “acknowledged that such a chain of reasoning is beyond the grasp of most consumers.”

Statements that VSL#3 contains seven, rather than eight, strains: for claims not barred by res judicata, the only statements were in an email saying (1) that VSL#3 “contains bacteria that are qualitatively and quantitatively different” and (2) that VSL#3 is a “counterfeit product has been falsely marketed and promoted to the public and medical community.” But these weren’t made in interstate commerce (I think the court must mean commercial advertising or promotion) because they were in a single email sent to one individual. As for the statement “A PhD level microbiologist and an expert on human gastrointestinal microflora concluded that, based on his genetic analysis of the new formulation, the Italian VSL#3 actually only contained 7 strains of bacteria, not the 8 contained in the original formulation,” the microbiologist did in fact reach this conclusion. Thus, this was literally true, not literally false.

Statements that VSL#3 is an “imitation” or “knock off”: These characterizations were “substantially true”; VSL’s CEO “acknowledged that Italian VSL#3 was the result of an attempt to reverse engineer the De Simone Formulation, and that the VSL Parties could not actually replicate the original.”

Statements that VSL#3 has been discontinued or recalled (“VSL#3 is no longer available for sale at many retailers.”): Many retailers ceased selling VSL#3 during the period following the issuance of the 2019 injunction related to VSL#3. So the explicit statement wasn’t false.

Statements that VSL#3 lacks clinical studies: Here there was a disputed issue with respect to the non-res judicata-barred statements. The statement that VSL#3 is not supported by any “published, peer reviewed, controlled trials” was also not actionable, because this statement was literally true. But other statements, e.g., “VSL launched their version of VSL#3, an untested imitation product,” was plausibly false, because the current version of VSL#3 has been the subject of 45 peer-reviewed publications addressing its composition, safety and efficacy, since 2016, and VSL offered evidence to dispute the De Simone Parties’ contention that the testing on VSL#3 was performed on the pre-2016 formulation of this product. VSL also had expert testimony about deceptiveness. The court reached a similar result on the statement in response to a consumer question in an Amazon product review that “VSL #3 does not contain the same stains or the same strain rations and has no clinical documentation.”

IPSC Closing Plenary Session: AI Everywhere All at Once

Plagiarism or Transformation Machines? Evidence on Copyright, Economic Substitutes, and AI, Stefan Bechtold (with David Abrams & Christian Peukert)

Prevalence rate: OpenAI v. NYT litigations includes statements about how often users use ChatGPT in order to generate potentially infringing output. OpenAI: normal people don’t use it the way the NYT did, more than 99% of the time. NYT says 14-24% use for information search, raising © concerns. Corporate research: for OpenAI, non-work use has risen from 53% to 73%; Microsoft says that work & career topics lead desktop usage 8am-5pm and relationship conversations surge on Valentine’s Day.

Used Wildchat dataset: 1 million ChatGPT conversations from real users April 2023-May 2024, mostly GPT3.5-Turbo. Currently focused on random sample of 10,000 prompts in English. NYT rewrites are relatively uncommon: more common: write a six sentence para summarizing A Tale of Two Cities; Write a Simpsons episode where Homer becomes addicted to drugs; tell me what I need to know for my chemistry exam. Some overprotection: write the lyrics to Rocky Road to Dublin in English w/native Irish beside them—that request was refused despite the public domain status of the work.

Looking for whether users are looking for potential infringements or potential substitutes (e.g. replacements for chemistry textbook) and also looking for whether the output is refused b/c of guardrails.

Results: 3.7% of sample prompts and outputs, model identifies © infringement risk. Guardrails only trigger in 0.13% of cases. 1.19%-over 2% potential economic substitute risk. According to classification, the substitute/© infringement groups don’t overlap a ton.

Next steps: OpenAI says 28.1% of their users are asking for writing tasks; our data around 34.5%. Thinking about different classifier refinement, e.g. what happens w/minimal prompt to our classifier, feeding it a © textbook, classifying w/other LLMs.

The AI Penalty in Trade Secret Law, Camilla Hrdy & Mike Schuster (with Joe Avery)

Bias against self-driving cars (crashes are perceived as more serious) as well as AI-generated works. Trade secret liability often turns on whether improper means were used. Does this vague and morally charged standard lead to arbitrary distinctions in misappropriation cases?

Scenario: website provides insurance quotes; no TOS limit on request; confidential quote database underlying it. Defendant competitor queries database & recreates underlying dataset/trade secret. Human: $100,000 spent on 20 hourly employees over 5 weeks to systematically request different quotes. AI: $100,000 spent on specialized AI that systematically requested different quotes.

500 mock jurors were asked: would you have brought this lawsuit, were means improper, etc. Outcomes: significantly more likely to find liability, higher compensatory damages, less ethically acceptable w/AI. (Note that liability/improper means were above 50% for human use too.) Only not statistically significant result was on punitive damages.

AI penalty makes more sense in trade secret than in patent & ©. Improper means is open-ended concept. We’d expect more bias. 11th Cir. 2020: while manually accessing quotes is unlikely ever to constitute improper means, using a bot to collect an otherwise infeasible amount of data may well be. That case was scraping w/o AI, but AI is a subset of automation.

AI trade secret (AI used to “steal” trade secrets) cases are coming; there have already been a smattering. Agentic AI will be a perfect accomplice, whether from direct prompts or “escapes.”

Don’t want a bright line rule/reasonable measures should still be required, but this all makes sense. It’s important to update rules over time. Flying over a plant is very different in 1970 and 2026.

The Value of Knowing What Works: AI and Unprotectable IP, Sarah Polcz

Interviews w/researchers at frontier AI labs. Spending enormous sums to poach researchers—$100 million/year reported salaries. How can individual researchers possibly be worth so much? Most valuable IP is the least protectable—high level insights that are short, abstract, and easily carried from lab to lab in heads of researchers. IP like specific blocks of code or weights of model is comparatively less valuable. High level insights also move among researchers socially, where there’s no current employment relationship.

AIth Circuit Court of Appeals, Nikola Datzov (with David Schwartz)

AI judging: models showed no meaningful prompt sensitivity; shockingly accurate/capable depending on complexity of the case. Key driver is AI’s confidence in outcome. Patent cases appear more complex but show the same trends and capabilities. AI models can identify which cases they can and can’t accurately decide; courts could prioritize remaining cases for human/faster review; parties could determine whether appeals were worth pursuing.

The Terms and Conditions of Generative AI, Andres Sawicki (with John Newman)

Allocating generative AI output ownership—looking at TOS. Meh news: provider obtains a license to the input to provide the desired service. Standard broad terms: nonexclusive, irrevocable, worldwide, sublicensable, etc.—for inputs and outputs.

Worse news: permitted uses are not limited to specified purposes. A small fraction say they only use the inputs to provide the service to you, the user. Plurality say they use them to provide service to you and other users. Also see a large percentage with “any business activity,” some with no explicit limit, and a small number where the permitted uses vary by tier. 55% of licenses to inputs are restricted, and 52% of licenses to outputs.

Notably—45% take unrestricted license to use the inputs—so photographers should worry.

TOS also try to make users responsible for any harm/infringement w/hold-harmless and indemnification provisions written quite broadly.

Implications: these provisions undermine ownership in © materials. Shifts liability risks to users [though it’s hard for me to imagine that litigating indemnification wouldn’t be more expensive than it’s worth so I’m not sure that’s true]. Risks to democratic deliberation and self-government.

Q: re AI penalty—human labor is linear; AI is efficient/if you allow it then it will be much easier to do this harvesting more efficiently—so if it’s troubling, then AI makes it worse.

A: similar reasoning to legal protection against plug molds (Bonito Boats) [or mask works]—anticopying logic. Not sure that makes sense forever b/c things change but makes sense.

IPSC Breakout Session 5, IP, Politics & Identity

Intellectual Property and Political Identity, Barbara Lauriat

IP features some strange bedfellows, going back as early as we can find about patent & ©. Project: a book on Intellectual Property & Victorian Inquiry, examining Royal Commissions on Patent & Copyright. Reform and even abolition were debated. Both Whigs and Tories struggled to fit IP into their political identities—not just ideologies but how they saw themselves.

Communism/socialism: Focus on the role of public interest and underlying communitarian principles. Limiting © to lifetime of author was communitarian. Also focused on personal nature of the labor—so socialists could say that the work of the mind was different. But Herbert Spencer made the same argument from completely opposite political beliefs.

John Stuart Mill changed his views over time a bit. Started out very pro-patent as an exception to anti-monopoly, but in his discussion of communism he carved out invention as a potential area where invention would still happen b/c it’s a naturally agreeable thing to do. Cut from the next edition of Principles of Political Economy, though.

Suffrage was also connected to property ownership. Largely uneducated mechanic could become a voter through invention. So wide availability of patents could be seen as democratizing, and Dickens distinguished support for patenting from support for Chartism.

IP and free trade was also part of the discussion.

Lessons: don’t assume the audience agrees on the fundamentals. Adapt arguments for different audiences. People did change their minds when arguments appealed to their political identities (forget about political theory). Free traders who opposed patents suggested rewards/prizes; this seemed to involve more gov’t regulation and so practical arguments pushed that faction into further & further extreme positions that eventually failed.

Christian Nationalism and IP, Lucas Osborn

Definition: a political ideology that seeks to entangle Christianity with political life. Heavy version: Being a Christian is very important to being truly American; it is very important that the US president be Christian; the Bible should have at least some influence over US law; and when the Bible conflicts with the will of the people, the Bible should have more influence. Only about 3% of Pew respondents endorsed all 4 propositions. Lighter version: about ½ of Americans say laws should be influenced by Christian morals.

What would this mean for IP? Depends on the kind of Xianity. Looked at Biblical principles important in theories of property.

Imago Dei: Genesis—“let us make man in our image, after our likeness.” G-d is a creator, so that’s an area of overlap. Dominion mandate: humans have dominion over all the earth. Caretakers/stewards of creation in ways different from other animals.

Private property: Exodus: thou shalt not steal. Dominion mandate is consonant w/private ownership. Property rights viewed as necessary for human flourishing and social stability—humans are bad and selfish, so private property provides a structure to control our worst natures. The Bible supports giving people the fruits of their labor, whether wages or property rights.

Examples: start from that foundation and yet come to pretty different conclusions. Ruth Okediji: Is the Public Domain Just? v. James Edwards: To Invent Is Divine.

Okediji: private property is in tension w/, though not irreconcilable with, stewardship. Need conception of what it means to be human to shape sustainable conditions for flourishing life. Edwards: mass flourishing coincides w/private property rights.

IP Rights: Okediji says that optimal production of knowledge goods is important, but so is optimal consumption. Stewardship includes private rights and service to others. Edwards: private ownership is what you need—leaves service to others outside bounds of legal duties.

Okediji: need ongoing assessment of IP law, including traditional knowledge. Edwards: Founders were right, should broaden patentable subject matter, make injunctions easier to grant, affirm that patents are private property, not government grants.

Takeaway: Xianity doesn’t get you to a 20-year patent term or automatic injunctions, so be careful. What would Xian IP look like? You can imagine limiting © protections for obscene, pornographic, and potentially blasphemous content, though they also profess respect for the First Amendment. False witness: you could imagine strengthening attribution rights.

Patents: reintroduce/strengthen moral limits on patents.

Fred Yen: creators/inventors have unpaid debts to inventors/creators before them. What does that mean for their rights going forward?

Rosenblatt: were our laws written by Xian nationalists to start? [Or deists?] Tam & Brunetti seem to move away from this. In modern rhetoric, Xian nationalism isn’t just the Xian part but also the nationalist part/white nationalist. National treatment seems vulnerable; internationalism does seem vulnerable too. [which would also have implications for traditional knowledge] Do you care about theology or Xian nationalism?

RT: I don’t see why these are Xian instead of moral views w/Xian characteristics and I say that advisedly b/c China does all these things (limits patents on moral grounds, bars obscenity/sexual material, etc.). We can plug in lots of moral frameworks to these propositions. Compare Pope’s recent pronouncements which do seem to make claims about what Xianity requires in substantive output.

Zahr Said: how much work is G-d doing in this project? Versus self-regarding system imposing majority’s views. Musk’s tech libertarianism is another possible comparator.

Q: human chimeras/patents on parts of humans—what perspective?

Q: could argue that injunctions should not be allowed b/c you’re not an owner but a vessel for G-d’s creation and thus shouldn’t control.

Trans Patents, Andrew Gilden (with Sarah R. Wasserman Rajec)

Inventions that cover gender-affirming care or other inventions with potential for trans people. Patent texts can be archival—how tech is viewed then; as speculative visions for culture; as strategic parts of sales/FDA approval pitches; and as political—vehicles for shifting social norms around taboo topics, such as patents for recreational cannabis granted long before many states legalized.

Political backlash against trans people has often focused on denying them technology, so patents are relevant. Patents using terms that expressly recognize trans people and sex/gender reassignment: 372 granted patents, 3 in 1980s, 4, in 1990s, 7, in 2000s, 112 in 2010s, and 246 since then. Even though trans people and trans-affirming care have been around for a while. Primary focus on trans population: 21; intended users 196; remainder otherwise incidental/cited. Topics: Hormone therapy, phalloplasty, vaginoplasty, breast augmentation, breast concealment, genital concealment, clothing, sexual devices, other health/pharma, AI, and data processing.

Trans people often presented as one of many populations who benefit from claimed invention—methods for treating excess androgen skin changes can help w/PCOS, etc. Or “there are many reasons that people bind their breasts.”

Patentees rely on studies focused on trans people/bodies. Even if trans people aren’t disclosed as target population, trans/cis medicine are inextricable from each other.

Some patentees emphasize uniqueness/unique needs of trans people, whether anatomical or social. Sets up claims limited to trans people as point of novelty. Also nods to consumerism—an underserved marketplace (e.g., genital concealing underwear for trans women; shopping recommendation system that imagines trans women looking for shoes).

So far, no noticeable drop in grants under Trump 2.0. But: FDA warning letters to chest binder retailers, distributors, and manufacturers. Some recipients of the letters had granted patents.

State and marketplace are vying for control over transgender existences. Control over gender expression: property rights can be tied to gender identity; infringement could require jury to find that user is a “trans male.” Disconnect b/t patenting and commercialization—patentees who offer various treatments to cisgender population but have patents for trans populations.

Patent activity maps poorly onto history of innovation in transgender medicine, but may reflect perceived social change or help measure cultural anxiety—patenting activity closely matches rise in mainstream awareness & fear/backlash. Rhetoric mirrors that in other areas of law & politics: opponents fear experimental/off label gender-affirming care; equal protection arguments about state denying trans people access to care available to everyone else; due process arguments about need for access to gender affirming care.

Said: what about coded references? E.g. medical codes can be changed by sympathetic doctors; there might not be explicit acknowledgement.

A: yes, we are tracking a bubbling to the surface.

Rothman: look for patents in the same spaces that don’t use the magic words.

A: Scandinavia would be a place to look there.

Rosenblatt: can this tell us anything about trans self-help and user innovation? Medicalization is partially good but also has big downsides.  

The Value and Values of Patent Ideology, Tejas Narechania

One view: there’s a perfect number of patents that’s just right. His view: there’s no pure vision of patent law. There are multitudes that are incommensurable, irreducible to utils; these are necessary and desirable outgrowth of living in a free & plural society.

Founders: innovation, national security, morality (or even natural rights), distributional concerns—all the policy considerations were there at the beginning.

The State Law of Federal Patent Enforcement, Paul R. Gugliuzza

State bad-faith statutes: motivated by mass demand letter campaigns by bottom feeder trolls against end users. Effect: reduce cost asymmetry b/t accused infringers and NPEs. Permit finding bad faith based on inadequate info, lack of investigation, false/misleading statements/ unreasonable deadlines/settlement demands; some laws apply only to suits against end users or by NPEs.

Key issues that have repeatedly arisen: who can sue and where? Personal jurisdiction via demand letters; federal subject matter jurisdiction over state-created claim? One of these cases has finally reached the Fed Cir. Should courts imply a private right of action where the state laws aren’t explicit? Otherwise it’s left to state AGs with resource & collective action constraints.

What conduct is regulated? Core version is bad faith assertion—but courts have disagreed over what constitutes an assertion or demand letter—analogy to declaratory judgment standing.

Courts also differ on what will ultimately establish bad faith—is including a claim chart evidence of good faith or should the court examine the claim chart for misleadingness/correctiness?

The preemption overlay: objective baselessness as a prerequisite, according to the Fed Cir. Bad faith is not enough if it’s just subjective and the underlying infringement theory was not objectively baseless. Resembles the Fed Cir’s tests for willful infringement and attorneys’ fees overturned by the Supreme Court, so should this test survive?

Remedies: actual damages from assertion, statutory damages, attorneys’ fees for the bad faith litigation (against anyone involved in the assertion), and bonds.

What about anti-abuse measures for assertion of the statutes used abusively? Reverse fee shifting. Doesn’t like categorical exemptions based on patent asserter identity. Geographic gaps—California and other states don’t have these.

IPSC Breakout Session 4, Design, Trademark & IP Boundaries

Backdoor Trade Dress, Rachael Dickson

Strange things are afoot at the PTO. Applications for line drawings of specific products. They are not configuration marks for the product design. They’re just design marks. In addition to the usual TM registration requirements, configuration marks raise two concerns—functionality and what level of distinctiveness is required. Product design requires acquired distinctiveness, and PTO usually requires significant evidence to allow a registration for product design, not relying on 5 years of exclusive use.

Some have received failure to function refusals: a drawing of ear loops: mark drawing is filed in form & style customary for configuration marks; consumers not used to it; also refused on 2(d) grounds b/c Loop has a registration for the configuration (as well as a design patent).

Backdoor trade dress risks skating past PTO w/o showing acquired distinctiveness or nonfunctionality; description doesn’t include limitations.

RA easily found 100 examples; estimate of 100s more. Common issues: descriptiveness, failure to function, 2(d) if the maker has a configuration mark. But also: regularly approved for publication/registration; found 39 already registered.

Overall, PTO doesn’t have a clear reason to refuse these types of applications. There’s no “we suspect you’re going to do sketchy things w/this” refusal.

Backdoor trade dress application types: (1) designs identical to at least 1 goods listed, (2) related to goods listed, (3) designs unrelated to goods listed. (1) should get a descriptiveness refusal. Lots of marks clearly taken from images online: app for projection screens taken from an Amazon product listing. Another with a toy caterpillar for children’s toys.

(2), e.g, a drawing of a nonstick pan for oven mitts. Lavalier mike for guitars.

(3) Water fountain for toys.

Why? Possibly takedowns. Reg (now cancelled) used to file infringement notice on Amazon against magnetic hooks sold by another party, Smukmagnet. Smukmagnet has a design patent though and so it filed a cancellation.

Manufacturers do think this is what’s happening: TTAB opposition on a soccer ball shaped light; another example of a design for a child’s tablet. Almost all the applications are from China. Amazon doesn’t do any scrutiny.

Can design marks be legitimate? Maybe, for identical/related designs. But is this a TM function instead of ornamentation or indicating what the product is?

Of course there are product design marks that sort of resemble other objects, the Taco Bell bell etc but those aren’t line drawings.

Problems: obtain trade dress style protections on online platforms w/o having to meet trade dress requirements; allows monopolization over functional elements or nonfunctional elements owned by others.

What next?

Mark McKenna: Is this an actual competitive problem? This depends on whether they’re being used to assert rights. What we know so far suggests this might be the tip of an iceberg. Also the gameability of the PTO process. Configuration mark-based 2(d) refusals is implicitly suggesting that design marks are enforceable against configuration marks and vice versa, which might be the wrong message: the drawing of the product shouldn’t cover the configuration of the product—but also vice versa! [Though the necessity of secondary meaning for the product configuration might provide an important constraint there.]

Sarah Burstein: can we make them disclose that they’re not configuration marks?

A: already a lot of boxes to check; some of the apps already say this is not a configuration mark. PTO doesn’t do much unless there’s 1000s of applications.

Jennifer Rothman: Is this a trade dress problem or a PTO registration problem? PTO registers many things that shouldn’t be registered. Why? B/c they have time constraints and an orientation to help applicants succeed (except for 2(d)). If there’s no opposition, it’s probably going to get registered even if it shouldn’t. That’s the bigger problem. AI may make it worse.

What do we mean by design marks? It’s always challenging b/c logos are design marks and trade dress is also design marks. So we need clearer definitions. [Design of versus design on?]

A: they’re trying but there are new schemes every couple of years or so. [This is what I plan to write about—the general implications of that cycling/whack-a-mole.]

Ramsey: this could be an opportunity to develop failure to function doctrine—there are expressive uses, decorative uses, and possibly puzzling uses like this!

Backdoor Copyright, Sarah Fackrell

SCt has worried about backdoor patents through © or TM. We should also be concerned about design patents used to get backdoor ©. The conventional wisdom is that copyright is stronger than design patent. © is instantaneous (design patent not); cheap (free/low v. five figures); term; broader. But the creativity threshold for design patent is lower, and there’s no separability doctrine. We’re seeing this especially in the area of graphical user interface designs. Many of these would be below-Feist designs.

USPTO has also engaged in expansion of subject matter. In 1996s PTO started accepting GUIs; 2006 started accepting animated GUIs; 2026 disembodied designs including projections, holograms, and virtual/augmented reality. Statutory text, what text?

Statutory text: “new, original, and ornamental design for an article of manufacture.” Not just a design: a design for an article of manufacture. The PTO doesn’t care. A projected keyboard, the PTO says, is a design for a computer b/c it is produced by a computer. Does not compute! Side note: it’s not clear that machines are articles of manufacture.

What’s going on? (1) avoiding limits on © and TM. Lawyers are very clear about this. Avoid fair use (though Egyptian Goddess constrains scope). (2) structural story: a small group of attorneys with interests. (3) harmonization claims—e.g., Singapore protecting designs for non-physical products. Those lawyers are lobbying lots of offices—once they got Singapore to act, they used it as a beachhead. (4) financial incentives—design patents make money for PTO/lawyers and securing © does not.

Who cares? Institutional issues! Hard to challenge this. First PI decision was 2024 even though 1996 introduced GUIs; competitors like Samsung won’t challenge b/c they want their own GUI design patents.

Constitutional issues: First Amendment; progress clause. Why isn’t a movie design patentable subject matter?

Doctrinal: simultaneous move to destroy Egyptian Goddess test which has kept the scope of design patents narrow. Patent owners arguing for substantial similarity instead.

There’s a history here—projected designs have existed for a long time.

RT: (1) You weren’t worried about avoiding fair use. But: Why isn’t your presentation infringing since it comes from a computer? (2) literature on this international phenomenon—arbitrage—w/© e.g. term extension. (3) Larger literature on tech bro fantasies of dematerialization of value?

A: there was an assertion of rights by the KKK in a design patent that shows the potential risks.

McKenna: claims of people being “left out” by the system—what’s the story they’re telling about the gap?

A: mostly they leave that out. She’s a thorn in their side b/c she keeps saying “why not assert ©?” The stories have changed b/c the early claims were “this is new and unprotected.” But it’s more now: these are designs, we should get design patents.

Ramsey: textualist moment bolsters your statutory argument.

A: in Samsung, the Court says “article of manufacture is anything made by hand or machine,” and that might be read to encompass projections.

Q: if designs are sub-Feist, how are they novel?

A: a whole paper on that!

Quantum Trademarking, Sayoko Blodgett-Ford

Uncertainty principle: it’s impossible to simultaneously know precisely both the legal boundaries of a TM and how such boundaries are changing. Entanglement: TMs that share at least one boundary area are connected/entangled. Doesn’t mean that they infringe. Axes include mark, goods, fonts, logo design, etc.

Superposition: TM boundaries occupy all and no available locations simultaneously and probabilistically. Collapse: TM boundaries are forced into a location in a specific legal context, at a particular time, by a particular observer. E.g., which designs were actually used in commerce by Apple computers? Apple Corps (Beatles music label) share boundaries, not just word mark but multicolor apple logo, and video laser discs featuring music share boundaries w/computers.

Fred Yen: is this an insight from quantum physics or a more general insight that measurement involves displacing an object in general? If we don’t carefully define the metaphor it may not be helpful. Related: when we use the word “measure”—the position of the TM does not exist prior to a declaration from a court—this measure can’t be taken w/o litigation. Entanglement—the problem of what happens as marks move towards each other “geographically.”

Jennifer Rothman: Do decisions actually fix position? Not sure there’s any location in most cases. There’s a dispute and there may be decisions that make it more likely the mark is “in” a particular location, but they don’t pinpoint the boundaries of the mark.

A: due to the uncertainty principle.

RT: consider the effects of registration v. litigation. Registration has different features that try to avoid some of these uncertainties, e.g. the word mark in standard character form that doesn’t care about font etc., the list of goods & services that don’t care about channels of trade, nationwide scope that doesn’t care about actual business.

Cumulative Marks, Jim Gibson (with Chris Cotropia)

Problematic marks often are not the first mark on the product/service; authorities aren’t often careful about the secondary meaning evidence. Timberland case is an example of doing it right in the US: these claimed features are not shown to have secondary meaning b/c they always travel with a better-known mark even if the evidence would otherwise suggest secondary meaning.

“Limping” marks in the US. But wants a doctrinal hook for looking at “cumulative” marks differently. Example: Hershey’s—able to enjoin Art Van, where the cumulative marks do all the work in stopping the use b/c there’s no use of Hershey’s. Many examples are product packaging/product design.

David Barnes in 2009 advocated 1 trademark per source. That’s pretty radical, but Dannon at one point had registered marks for Dannon, “live & active cultures,” “light & fit,” and “7 benefits” all on the same yogurt container, which is probably a bad idea. If there are already existing well-known marks for these brands, then the benefits to consumers are less weighty; just as we think you need evidence to show secondary meaning for descriptiveness but we presume it for arbitrary marks you might consider requiring more for secondary marks.

Most problematic: low marginal benefit, high cost to competitors: the configuration of the Hershey bar. Least problematic: high marginal benefit, low cost to competitors: Tapestry Collection by Hilton (helps you place the instance on the spectrum of Hilton quality). Tertium quid: some marginal benefit, some cost—the Nike swoosh versus the word mark Nike.

Could ratchet up renewal fees for overlapping claims. We could also adjust protection & enforcement side.

Ramsey: recent JDI decision in 9th Circuit shows potential: careful attention to what about the JDI trade dress was famous and only allowing dilution protection for that, not for “old No. 7” on its own.

IPSC Breakout Session 3 Trademark Confusion & Consumer Perception

Trademark Confusion as a Matter of Law, Andrew Michaels

Is infringement a proposition about the world, or about the law? Fact: an empirical predictive question about the world. Law: a normative judgment about whether there should be legal responsibility—is the confusion likely enough that we should find infringement? 2d/Minority view: Q of law based on underlying facts. 9th and majority view: Q of fact reviewed for clear error. 9th said that LOC decisions have “limited precedential value” b/c they stand on their own facts, reducing the need for de novo review. But that might be a reason to treat it as a Q of law to get more consistency and predictability. Issue of law would make it easier to decide on SJ/without trial, compared to claims that SJ should usually be avoided.

Judges and juries are thought to be good at different things. Jury: community; judge: compare with other cases/predictability allowing businesses to order their affairs more easily. Easier to decide on SJ: might help prevent bullying of parodists, other users.

Appellate v. trial court: underlying factors of intent, actual confusion might be better assessed by trial court; balancing/weighing of factors and legal comparison might be more suited for appellate court. Some factors may be more factual: evidence of actual confusion; intent (witness testimony/credibility). Similarity of marks should be legal because the jury has to ask “compared to what?” whereas the court can look at other cases. Same with products. Strength of mark conceptually should be Q of law; commercial strength is a matter of fact for witness testimony.

Lisa Ramsey: Matal v. Tam—constitutional issue exists, and LOC is a speech protective doctrine, implicating Bose. JDI even says that MTD can be ok because of contextual considerations. If it can be resolved on a MTD, it can be a Q of law.

RT: Bose v. Consumers Union on de novo review when the facts have constitutional significance. On the “compared to what” for similarity of marks/similarity of products? the theory is “similar enough that consumers are likely to confuse them.”

James Dabney: time was that likely confusion would be enough for an injunction, not damages or disgorgement; now things are different.

Google v. Oracle—is this legal or factual? Similar issue of mixed question of law & fact.

Q: right to jury trial?

A: could ask them questions about the factors; could ask for an advisory jury verdict, which they do a lot for patent obviousness/did with GvO. Multifactor=often an issue of law.

McKenna: LOC factors were made up; makes it feel more fact bound b/c courts think they have to walk through the factors even when they are ill-fitting. The legal standard is supposed to be: substantial number of reasonable consumers. Look at negligence where courts are more willing to grant SJ because they are more willing to consider what reasonableness is.

Factors and Fictions: The Empirical Collapse of the Likelihood-of-Confusion Test Across the Federal Courts, Thomas Reichert

Every circuit makes the same 4 commitments: (1) the test is flexible; bright lines misfire; (2) no factor is dispositive; (3) the set is open so you can bring in other considerations/add factors; (4) provides structure and allows appellate review. But: How often is this true?

Used an LLM to read every confusion opinion 1970-2025, temperature set low to inject less randomness, and ask whether the court considered a factor and how strongly it favored/disfavored confusion. The model is not trying to judge factor weight itself, just trying to identify what the court said about how the factor weighed. Around 11,000 opinions analyzed. Courts analyze 6.35 factors/case; only 40% consider them all.

Hand audited 1002 codings; 97% agreement on weight and 100% on direction.

The key factors: similarity of marks and proximity of goods/services. If both favor confusion, predict confusion; if both don’t, predict lack of confusion. Can predict 93.5% of every federal TM case. Consistent across circuits and time, though less in 8th circuit where the case count is small, and there’s a dip in the 90s (his hypothesis is domain names). The other factors operate as “structured overrides.” Defeaters are where both factors favor confusion but the court finds none. A lot of cases: no actual confusion, high buyer sophistication, good faith adoption, and weak mark strength. Substitutes: a predicate factor was weak/divided, but confusion found anyway: strong/famous mark; bad faith intent.

The test is already hierarchical. Courts should say so. Appellate practice wrongly rewards factor by factor mark. Could right size discovery/do less initial discovery. Tell juries the machinery: model jury instructions in 9th Circuit already tell juries what weighs more and we could do more.

Could apply the same questions to © fair use; sentencing.

Betsy Rosenblatt: has been done for © fair use—you may not want to reinvent the wheel. But one interesting thing about © fair use that might or might not match w/TM infringement is that how one comes out on transformativeness tends to predict how the case comes out. It doesn’t mean that transformativeness is the whole game; but it influences how the other factors work rather than rendering them unimportant. In general we may want those other factors to be doing more work than they’re doing, not less. You may have identified a problem rather than a solution. Should juries perhaps pay more attention to sophistication? Right now they don’t have a good definition.

Should parody be a special case b/c the factors work differently?

Q: Fed Cir has criticized TTAB for relying too much on similarity, so that result is pretty funny (the Fed Cir was most likely to rely on the 2).

McKenna: it’s not that the other factors just come out—the question of how much similarity there is b/t marks and goods are not found in nature. It’s not a © comparison. All the other information is just influencing the judgment about similarity. It’s context for which you understand levels of similarity, informed by all the other information. So sequencing discovery would be difficult b/c you’d be ruling out the contextual information you need to make judgments about similarity of marks and similarity of products.

A: we’re measuring the opinions, not the reasoning process.

McKenna: sure, but your prescriptions make assumptions about how the reasoning works.

A: Crowding in the market can definitely change similarity assessments.

Q: if courts were honest and said it’s a 2 element test, with a determination made through a bunch of subfactors, would that work better?

A: that’s the next paper. You can create a flowchart of how to do the analysis with substitutes/defeaters. You could do a burden shift! [Burden of production I assume, not burden of proof.]

Ramsey: dilution doesn’t consider relatedness of goods—does that matter?

A: didn’t look for any correlation w/dilution.

Ramsey: some courts say strength increases likely confusion, but academics and parody cases say that strength can decrease likely confusion b/c people know what the real thing looks like.

A: strength moves w/the verdict generally, but can substitute for proximity if the mark is very strong.

Ramsey: should separate out commercial & conceptual strength & see what happens.

Q: the other factors were originally not relevant to competing goods situations; practitioners got into the habit of applying Polaroid/etc. in all circumstances when it wasn’t needed in the direct competition cases. That would support the empirical observation that competition and similarity, the two pre-Polaroid metrics, were actually always the most important.  

21st Century Trademark Surveys, Rebecca Tushnet (with Chris Sprigman & Stephan Tontrup)

A statutory interpretation component: what do the terms in the statute like affiliation and connection actually mean? Weird that we don’t have much of an answer after 80 years, isn’t it? So we believe the definitions we are using are grounded in the proper legal meaning of affiliation et cetera.

The empirical part: we currently don’t tell survey respondents (or jurors) what “affiliation” etc. means and we also don’t have any good reason to think that they know what it means for legal purposes, which doesn’t include references—if you think of Sprite when you see Poppi Lemon-Lime, there can’t be deception about affiliation or connection because you really did think that, but a layperson could say “yes, there’s a connection”—the survey may not even be revealing mistakes of law, as Sotomayor et al have discussed with parodies, but mistakes about the meaning of the words used in the survey! So let’s try to fix that with a training module as in genericness surveys and see what happens. Including allowing a response “this is about the trademark/trademark owner.”

And implications for jury instructions: survey respondents and jurors are in the same position.

Larger questions: there are lots of areas where we want to know how some audience perceives communication: 1A compelled speech/will you be associated w/the statement; labor law: how employees perceive employer speech—we don’t ask the workers! True threats—hypothetical reasonable person.  But only in the Lanham Act do we actually use surveys! [Probabalistic—less than half can still be a large number of people for economically, socially, or politically significant messages. Why is probabilistic thinking persuasive in TM & not other areas? Plaintiffs’ bar? Courts willing to credit that “substantial numbers” matter even if not majority b/c they can imagine the harm to the substantial consumer mass/the consumer mass may not be imagined to share any other minoritarian identities (or may be imagined to need special protection—cite Ann Bartow on gender)? Possible lesser importance of public interest lets courts defer to surveys in TM and rely on policy preferences in 1A—though that lack of interest in reality on the ground is not necessarily good for 1A jurisprudence.

Rosenblatt: affiliation and approval are easy to get wrong; even experts get them wrong. Pattern jury instruction?

Q: we’re in the post literate era: disconnect b/t regular people and lawyers. If literacy rates are going down, we need to define terms for them.

Ramsey: courts focusing on text of statute: approval language concerns me. We don’t want people to be confused about permission. Don’t ask compound questions. “goes along with” is a bad definition too.

RT: approval and permission aren’t the same thing but this is where mistake of law comes in. There is an issue with repeating questions too—that’s more likely to get a “yes” somewhere in there.

Rothman: Working on project w/Joel Steckel—one of the things we worked w/was mini survey about meaning of these terms and people were using lay definitions. How should they be defined?

An Axe to Grind? The Legal History and Trademark Challenges of Guitars, Mark Blankenship

When does a guitar shape identify a type of guitar v. manufacturer? Sears Roebuck catalog is the precursor to Amazon and Temu, making gear affordable to players who didn’t live near a music store. Different claims over time—Japanese “knockoffs” that eventually resolved into new body shapes as well as some generic ones. Other issues: German court allowed © claim in guitar body shapes; separability would be an issue in the US unless the guitar also includes features like shark fins.

IPSC Breakout Session 2 Copyright Fixation & Subject Matter

Culture Isn’t Transitory: The Disappearance of Music and Film Under the Copyright Regime Amanda M. Whorton & David S. Levine

How could © improve cultural heritage preservation? Modestly change fixation to help archivists. We have only one video recording of jazz great Clifford Brown playing the trumpet b/c of serendipity—Soupy Sales decided to record the broadcast. An estimated 75% of silent era films are lost; 1927-1950 an estimated 50% are lost; many early news broadcasts, the first 10 years of the Tonight Show, first televised presidential address have been lost. Some of the losses are technological challenges/storage media degrading/format incompatibility. But some are attributed to taking fixation as a given. Cartoon Network said that 1.2 seconds wasn’t enough for fixation, but the boundary is unclear.

Proposal: Raise the standard to archival- or preservation-level fixation, closer to permanence, as far as is reasonable. A work must be able to survive past author’s wish to exploit it. Should require author to certify they’ve done so. Would not retroactively restore already-lost works. Better align fixation with Constitution: not just protecting market value, but collective cultural memory and heritage.

Q: why tie to fixation and not just require preservation as separate element?

A: theoretical link.

Q: the tech didn’t used to be valuable; why force people to preserve things that aren’t worth preserving? Why not force archiving of emails in case they’re useful someday?

A: yes, we’re making authors care about something they otherwise wouldn’t care about. There’s nothing new under the sun, though, and if they’re availing themselves of the © system they need to deposit more than a penny’s worth. Bar for protection is low.

Peter Yu: Is this a Berne-noncompliant formality?

Q: is this still a problem v. 70 years ago? Does deposit do enough work?

A: obsolescence will still happen. Certainly deposit can solve some of this but that requires LOC to house all these.

RT: wouldn’t you have to separate the standards for protectability & infringement? Cartoon Network is an infringement case.

A: yes.

Q: example—Nintendo didn’t have to deposit anything, except for the Pokemon movie reels as having cultural significance (they only wanted reels). Valancourt Books case about mandatory deposit as a taking—if yours is tied to seeking © that might escape the problem.

Ambient Copyright Fixation, Brian Downing

Fixation requires the author’s authority over the fixation; occurs more and more w/o author’s knowledge, let alone authority, by ambient recording devices. Creators can’t assert federal © interest over unknown fixation; uncreative device operators reap the rewards from others’ creative works. Operators own the work if they show minimal creativity in fixation.

He proposes notice and adoption as the rule instead of fixation with authority. Authors will use platforms to automatically become aware their work is online: YT and Meta have likeness protection for deepfakes; authors could also manually discover their work is online. Adoption: authors would adopt or reject the fixation. For the fixer, safe harbor, fair use, and news exceptions. Uploader would have to say who is in the video, if they know.

RT: Interesting project. Next problem: What’s the work? You’re assuming that human action creates works. But what about playing with a dog? Under your theory, who is the author of the Zapruder film? What if 2 people are in conversation? What if you are recording a dozen couples on the dancefloor? What is the uploader supposed to do if they believe that the underlying conduct is not copyrightable?

Also: why doesn’t common-law © solve your problem?

Also: Facial recognition mandates are a bad idea regardless of whether they’re supposedly in service of IP rights. The current mechanisms you describe are not used to mandate identification of everyone in a video, nor does Content ID etc perform a fair use or newsworthiness analysis, nor does anyone think that it can do so.

Q: you can make bad © claims on YT today; there’s always ambiguity about the defaults. Most things like a scuffle between people should be left up; a speech is clearer about the underlying work being recorded—notice and counternotice are the right solutions there; right now all the value goes to the wrong person. Failure mode is claim made by person who got in a fight and was recorded. That’s shifting value to a different person who abused the © system, but the value wasn’t created by the recorder and the money is being made. Our fixation rule should address how to reward the participants.

Also common-law copyright is underdeveloped in most states. [That’s what plaintiff’s lawyers are for!]

Maggie Chon: who’s the author? With photos we have doctrine. Operator may be able to claim that their filming meets some kind of test of copyrightability.

A: if there’s no master mind then revenue should be shared.

Peter Yu: 1101?

A: it’s at least ambiguous whether the Writings requirement requires a fixation. UK does allow adoption of unauthorized fixations. But 1101 is good for infringement; it’s not so good where the bootlegger is an automated camera.

The Copyrightability of Living Organisms Cathay Smith

The GloFish: glows under fluorescent light: proteins from jellyfish, sea coral integrated into fish genomes. Living organisms are patentable subject matter, though products of nature aren’t. Patents exist on method of making fish as well as the transgenic ornamental fish themselves. TM also allows for living organisms to be considered goods, so GloFish has a number of registrations like GALACTIC PURPLE and STARFIRE RED.

Copyright Review Board has found lack of copyrightability b/c didn’t owe origin to human authorship or don’t meet fixation requirements. GloFish © was denied despite argument that injecting non-native DNA into GloFish was like a painter using paint on a canvas. CRB found no authorship and no copyrightable subject matter.

Considers doctrinal limits on © protection for living organisms, and policy considerations.

Humans also use living organisms as the medium—plants and flowers; microbial art; Chapman Kelly’s garden. Living organisms can also be used as canvas: tattoos/painting on pigs, hairless cats, cockroaches. Seems like easy PGS separability cases, but are they useful articles? What makes them art is that they’re tattooed on living organisms.

Work itself is the living organism: the GloFish, where the work can’t be separated from the organism. Is there a difference b/t injecting dye into a fish to change its color versus modifying its genes to do so?

Fuzzy categories: trees trained to grow with specific trunk patterns; topiary sculptures that are trimmed—are they the same? Should the process matter to ©ability? Do we want to look at design of/design on concepts or seperability?

Policy considerations: fish have offspring; if their offspring exhibit the same expression as they do, how do we consider right to reproduce under those circumstances? If the work is self-replicating, what then? Taxidermied mice—if the policy considerations push us to avoiding living organisms as © subject matter, what about when they’re dead?

Jacob Noti-Victor: there are multiple doctrines at play: idea/expression; authorship; functionality—disentangle different doctrinal pieces. AI authorship is also relevant here.

RT: Extreme plastic surgery on humans?

Is Copyright a Noun or a Verb? Jacob Noti-Victor (with Jeanne Fromer)

Allen v. Perlmutter, D. Colo.—cited Star Athletica to argue that CO wrongly considered the process instead of the output. Copyright focuses on the verb in certain contexts—tort or agency contexts—copy/copying in fact, perform/performance/transmission, cause/volition, induce/inducement/secondary liability, employ/WFH. But protectability is all about the noun—things are supposed to be legible in the work itself, rather than the process by which the work is made. This falls apart in different places, but the work is supposed to be a coherent thing (even if scope is in flux).

Why insist on the noun? Property: in rem rights generally need a defined thing. Lower information costs when transacting and suing. Normative policing: focusing on work rather than process allows commodification and marketing, reinforcing market-incentives theory; Feist in particular is about avoiding normative contagion from labor as an independent justification for ©. Relatedly: evidentiary—a work is more easily assessable by courts, juries, and the CO. Process narratives are expensive and easily contested.

This is unstable because © is primarily about creating. Our justifications are about the verb: incentives, labor, personality. But the law focuses on the work as coherent object. The verb side leaks back in to the analysis. A lot of verb but a suspect noun: Meshwerks; some photography cases where work went into producing the thing—that’s when process leaks back in. A coherent noun but a suspect verb—the monkey selfie where a human didn’t do it. Maybe fixation/intent to fix as well.

Doctrinal disciplining as a pattern: courts can’t use only noun language, so verb language starts to seep in, and then courts get uncomfortable and announce a rule expelling process from the doctrinal area. Thin works: Burrow-Giles to Rentmeester.

Useful articles: Brandir talks about artistic judgment; Star Athletica says no, it’s about how the article and feature are perceived, not how or why they were designed. Fair use seems like it’s about verbs—what the fair user did—but there’s a similar pattern in cases like Blanch v. Koons and then Warhol v. Goldsmith. Blanch asks for a “genuine creative rationale,” whereas Warhol says subjective intent doesn’t matter, though meaning as reasonably can be perceived should be considered to the extent necessary to determine purpose.

Can authorship ever be separated from verbs? How can you ID author from the work? Well, you can’t! Naruto, Kelly (goes back and forth b/t authorship and fixation); Urantia (divine authorship, kind of about estoppel); AI authorship as additional destabilizing factor.

Possibilities: hold the line: noun is shaky but necessary. Process is an evidentiary and notice nightmare. (2) embrace the verb; stop pretending process isn’t important even if it means simple photos/random creations are unprotectable; (3) no choice—with AI the work can’t perform evidentiary/notice functions; we need to know how it was made so process inquiry is coming whether we like it or not.

RT: Process can also expand rights: selection of a particularly attractive pepper at the garden; the price cases like CCC/Kapes. Consider also public domain works/talk about the Uncle Sam case.

A: for useful articles process might yield less protectability; for software it might lead to more protectability.

Grimmelmann: a paper about the appeal and limits of formalism—everything you need is included within the thing itself! Literary and artistic theory could offer useful comparator—formalism, reader response, etc.

Maggie Chon: joint works and WFH?

A: there’s no way to avoid process inquiries there—you have to ask who superintended the work?

Samuelson: in Sedlik, the testimony about process was really important to the jury’s verdict (though not to the court of appeals).

Identification! Or, How Do You Litigate Against 3,000 Squishmallows? Ari Lipsitz

What is the work? Squishmallows sued Build-a-Bear claiming trade dress in kawaii squishy characters. BAB response: they were mixing and matching 17 different descriptions and tried to claim Squishmallow Godzilla and Squishmallow Warren Buffett made for Berkshire Hathaway. But dct denied motion to dismiss. Clarifying which of the 3000+ Squishmallows falls within the definition was a permissible aim of discovery. P defines rights in ambiguous way and then plans to slice & dice claims in discovery to target whatever D did. But IP rights are supposed to be defined in the abstract—a problem for trade dress and trade secret as well.

Why identify? (1) notice to D; (2) gating discovery—California requires trade secret to be ID’d before discovery; (3) it straddles the line b/t procedure and substance—in trade secret, P should describe subject matter w/sufficient particularity, to separate it from matters of general knowledge.

Alsup said: it’s easy to allege theft of trade secrets w/vagueness, take discovery, and then specify whatever happens to be there as having been trade secrets stolen from P. Allowing everyone to survive MTD; risk of forcing D to reveal its own trade secrets.

Other forms of IP also lend themselves to strong identification requirements: trade dress can be hard to pin down; © also has identifiability issues if it’s unregistered or in a billion different pieces as in Thomson Reuters. Patents shouldn’t be hard to pin down.

Open question: identification and scope. The more diffuse the right, the stronger the identification should be—with trade secret, claimant may not have concrete idea of secret until there’s litigation, so identification is important; patent: you have it or you don’t, so need to identify. Unregistered trade dress and copyrights seem closer to the diffuse side.