A Los Angeles federal court issued a novel ruling about liability of social media influencers for trademark infringement. Petunia Products, Inc. v. Rodan & Fields (C.D.Cal. 2021) involved a standard trademark dispute between competitors in the cosmetics industry.   The trademark owner sued not only its competitor, but a social media influencer the competitor had hired to promote its product.  The district court denied the social media influencer’s motion to dismiss, holding that her actions constituted a “use in commerce” sufficient to hold her liable under trademark law.

Trademark owners should take note of another potential avenue to enforce their rights. Influencers and other social media promoters should also take note that in endorsing products, they may be exposing themselves to liability.

The Case
Petunia Products, Inc. owns the mark BROW BOOST for eyebrow conditioners, which it uses in connection with its “Billion Dollar Brows” eyebrow primer and conditioner product.  In mid-2020, Rodan & Fields, LLC (“R+F”), a competitor, began to sell a product called “Brow Defining Boost.”  Petunia Products had alleged that Rodan & Fields infringed its mark in various ways, including using its mark as an AdWord on Google’s service, and that it uses the hashtag #BROWBOOST to promote its competing product.

Molly Sims is a fashion model and media influencer who was hired by R+F to promote its allegedly infringing product.  She maintains her own website, where she promotes various fashion products. 
Sims authored a blog post in which she promoted the Allegedly Infringing Product and provided a link to R+F’s website, where the R+F product is available for sale. In the blog post, Sims opens by thanking R+F for sponsoring her post. She goes on to favorably review the Allegedly Infringing Product. At the end of the post, she includes a link to R+F’s website for those who want to “learn more about how to purchase Brow Defining Boost.” The end of the blog also includes an image of the Allegedly Infringing Product and its price.

The Court’s Ruling
A key issue on Sims’ motion to dismiss was whether she had made a “use in commerce” of the accused mark. The court held she did, and denied her motion, because commercial advertising falls within the scope of trademark law.  Her repeated use of images of R+F’s product, with its (allegedly) infringing mark, promotion and endorsement of the product, and links to R+F’s own site where the product could be ordered, all constituted use in commerce sufficient to bring her acts within the scope of the Lanham Act.   

This seems consistent with the Lanham Act, that subjects to liability any “use in commerce [of a confusing copy of a mark] in connection with the sale, offering for sale, distribution, or advertising of any goods or services . . .”  15 U.S.C. 1114(1).  Sims’ use appears to be a classic case of use of a mark in advertising.

The district court rejected Sims’ argument that such a ruling would stifle legitimate commentary, which would be protected by the First Amendment, because the case involved commercial promotion for pay, not merely independent review or product criticism. 

​The district court distinguished Bosley Med. Inst., Inc. v. Kremer, 403 F.3d 672, 676 (9th Cir. 2005), a case where someone used a company’s mark in the context of a website critical of the mark owners’ business.  There, the Ninth Circuit had emphasized that the accused website did not link to any commercial sites, did not offer any products or services, and did not contain any paid advertisement from any other commercial entity.  In the Molly Sims case, in contrast, her use of the accused mark was clearly part of a paid advertising campaign, and linked to a site that offered the accused product.

Implications
The use of social media influencers has exploded in the last several years, and many companies have exploited its marketing power.  Celebrities and other people with influence have likewise taken advantage of this new opportunity.  Both should be aware of the potential exposure to liability.  Influencers, for example, may wish to negotiate indemnification provisions in case of being accused of infringement.

​Trademark and other intellectual property owners should also be aware of another possible weapon in their enforcement arsenal.  

​The Supreme Court recently issued its long-awaited decision in Google LLC v. Oracle America, Inc. (2021), a copyright infringement case dealing with portions of Oracle’s highly successful Java programming software.  The Court assumed, without deciding, that the portion taken was copyrightable, but held that Google’s copying and use of it was a “fair use,” and hence not liable for infringement.

The Court’s fair-use analysis seems to break new ground in considering several of the four statutory fair-use factors.  But the opinion emphasized that it did not intend to “overturn or modify our earlier cases involving fair use” and that the unique nature of software in general, and the particular kind of software used by Google, led to the result, which could be very different in other copyright cases.
So, will the Google decision cause a sea-change in fair-use cases, or will it simply be ignored outside a narrow set of copyrighted works?  That remains to be seen.  In a recent Second Circuit case involving artwork by the late Andy Warhol, the plaintiff already petitioned for rehearing based on the Google decision. So we may soon learn the extent that the Google decision affects copyright law.

The Copying At Issue In Google

The case involves a type of software called an API, which stands for  Application Programming Interface.  It allows computer programmers to use pre-written code to build certain functions into their own, new programs.  This can be an enormous time saver, as a good API can call on a library of thousands of functions.

There are different parts of the API.  One part, termed the implementing code, tells the computer how to perform the function.  The programmer then uses “method calls” to call the particular desired function by name.  An intermediate part of the program, termed “declaring code,” connects the method calls to the implementing code, acting as a kind of address to inform the computer where the relevant implementing code can be found.

Google did not copy any implementing code.  It wrote most of its own declaring code as well, but copied a portion of Oracle’s copyright declaring code.  It used the declaring code to create the Android platform, an environment that allows programmers to create Apps that are used on smartphones utilizing the Android program. 

The Court’s Fair-Use Analysis

​As would be expected, the Court considered the four fair-use factors set out in the statute, 17 U.S.C. 107.  But the way it treated certain of the factors seems to deviate from prior opinions.

Uncharacteristically, the Court first considered the second factor, the “nature of the copyrighted work.”  It emphasized that not only was the portions of Java copied by Google highly functional, but they were “inherently bound together with uncopyrightable ideas (general task division and organization) and new creative expression (Android’s implementing code).”  Furthermore, much of the value of this part of the code lay not in its inherent creativity, but in the fact that Oracle encouraged programmers to become familiar with, and want to use, this particular set of commands.  All of this meant that the copyright owned by Oracle was “thin.”

On the first factor, the “purpose and character of the use,” the Court noted that this factor looks to whether the second user “adds something new, with a further purpose or different character, altering” the copyrighted work “with new expression, meaning or message.”  Google used Oracle’s declaring code to create “new products,” i.e., a programming medium for smartphones, an area of technology that Oracle had never entered.

On the third factor, “amount and substantiality of the portion used,” the court held this favored Google, because it only copied a very small percentage of the overall Java code.
Finally, turning to the fourth factor, “market effects,” the Court held (1) that Oracle’s own market was not affected, because it had never entered the market for programming of smart phones and (2) that there has to be a balancing of the public good from the use against the lost revenue.

A notable aspect of the decision is that the court emphasized that the fair use concept is “flexible, that courts must apply it in light of the sometimes conflicting aims of copyright law, and that its application may well vary depending upon context.”  The purpose of the Copyright Act is to foster and encourage creativity, and where copyright protection stifles that, courts should bend fair use so that copyright accomplishes its goal. 

Each of these seems to break new ground in fair-use law.  For example, one can discern a gestalt consideration that hovers above the four fair-use factors – will finding that the accused party’s use is “fair,” and hence not infringing,  inhibit or foster creativity?  If a court is convinced that denying fair use will stifle creativity, it should lean towards finding fair use.

Similarly, that Google used the very same type of expression (computer declaring code) simply to create a new product (smartphone programming platform), would until now generally not be thought of as transformative under the first factor. 

The Andy Warhol Case

Less than two weeks prior to Google, the Second Circuit decided The Andy Warhol Foundation for The Visual Arts, Inc. v. Goldsmith (2021), in which the Second Circuit rejected a fair-use defense in a case involving the late Andy Warhol’s use of a copyrighted photograph of Prince to create a piece of artwork.  After the Google decision came out, the Andy Warhol Foundation wasted little time in moving for rehearing, arguing that the Supreme Court decision required a different outcome.  Not surprisingly, the photographer opposed, arguing that the Google decision should be limited to software, as the Court seemed to hint.

So in short order, one federal appeals court may decide how much of an effect the Google decision will have on copyright law outside the context of software.

As e-commerce retailer behemoth Amazon continues to expand its share of the online market, so too do the legal issues created by its platform.  A recent New York trial court decision, State Farm Fire and Casualty Company v. Amazon.com Svcs. Inc., 70 Misc.3d 697 (2020) weighed in on one issue:  whether Amazon is liable under product liability law for defective products sold by third-party sellers on its platform.  Siding with the plaintiff, it ruled that because Amazon exercises significant control over products sold on its platform and distributed through its warehouses, it could be held liable under New York law for distributing the defective product.

This issue has divided courts throughout the country.  Many courts have held that because Amazon does not take title to the products sold on its platform by third party sellers, it is not a “seller” and thus cannot be held liable for the product if it turns out to be defective.  But other courts, including a Third Circuit opinion applying Pennsylvania law, and an intermediate California appellate court, have held that because Amazon exercises significant control over what is sold on its platform and distributed through its warehouses, it can be held liable under products liability law.

These decisions may also have implications for intellectual property cases, particularly counterfeits and other trademark infringements.

The State Farm Case

The case involved a thermostat purchased on Amazon, whose alleged defect caused a fire in a home.  State Farm, as subrogee of the home owner, brought a products liability claim against Amazon.  The thermostat was sold by a third-party seller on Amazon.  In such cases, Amazon does not take title to the product.  However, the seller has to register with Amazon, and the product must be approved for sale.   In addition, many third-party sellers use Amazon’s Fulfillment Services, whereby a quantity of product is sent to Amazon’s warehouse, which then fills the order after it is placed on the website.

Amazon moved for summary judgment, arguing that because it did not take title it cannot be held liable for distributing the defective product.  The trial court, however, rejected that argument.  It first noted that New York common law looks to hold the parties in the best position to maintain product safety.  As New York’s highest court held:

Where products are sold in the normal course of business, sellers, by reason of their continuing relationships with manufacturers, are most often in a position to exert pressure for the improved safety of products and can recover increased costs within their commercial dealings, or through contribution or indemnification in litigation; additionally, by marketing the products as a regular part of their business such sellers may be said to have assumed a special responsibility to the public, which has come to expect them to stand behind their goods.

Sukljian v. Charles Ross & Son Co., 69 N.Y.2d 89, 95, 503 N.E.2d 1358 (1986).

The State Farm court then noted that Amazon exercises tight control both over what may be sold on its website and what may be distributed through its Fulfillment Services warehouse network.  “Viewing these facts in the light most favorable to Plaintiff, Amazon exercises sufficient control over the product to be considered among ‘retailers and distributors’” who are strictly liable for defective products.

Other Courts

The State Farm court appears to be a more recent trend in product liability law.  For some time, many courts sided with Amazon and held that the fact that it does not take title means it is not a “seller” and hence had no liability.  But more recently, a few courts have ruled the other way.  In Oberdorf v. Amazon.com Inc., 930 F.3d 136 (3d Cir. 2019), the Third Circuit predicted that Pennsylvania would impose liability on Amazon.  That opinion was later withdrawn and the question certified to the Pennsylvania Supreme Court; the case was settled before that court could take up the issue.  Courts in Wisconsin and California have also ruled against Amazon.  State Farm Fire and Casualty Company v. Amazon.com, Inc., 390 F.Supp.3d 964 (W.D. Wisc. 2019); Bolger v. Amazon.com LLC, 53 Cal. App. 5th 431, 267 Cal. Rptr. 3d 601 (Cal. Ct. App. 2020), rev. denied, No. S264607 (Cal. 2020).

What these courts focus on is Amazon’s close involvement in and control of the third-party sellers’ market, including policies about what may be sold on the platform, customer service requirements, and pricing requirements.  For those third-party sellers who use Amazon’s Fulfillment Services, there are additional controls and restrictions.  And, Amazon takes a hefty percentage of the sales as a commission.

Implications For Intellectual Property Cases

Such decisions may also have an indirect impact on counterfeiting and other trademark cases.  Amazon has a very negative reputation among brand owners for having large numbers of counterfeits and other infringements.  While Amazon does have a counterfeiting takedown program, many brand owners consider the effort half-hearted at best.

Trademark law distinguishes between direct and secondary liability.  Someone who sells a counterfeit or infringing item is strictly liable for infringement.  A party that merely aids another in selling is only contributorily liable – and that requires a showing that the party knew that item was counterfeit and gave substantial assistance to the primary infringer.  That is a much more difficult showing.
Amazon has always insisted it cannot be held directly liable, for similar reasons as it has advanced in products liability cases. 

The Trademark Act holds liable any person who uses a trademark in commerce “in connection with the sale, offering for sale, distribution, or advertising of any [counterfeit or infringing] goods or services.”  Given the close control that Amazon exerts over products sold through its platform, it is likely that brand owners will argue that Amazon is using the trademarks contained in the offerings on its platforms in connection with “distribution or advertising.”  The decisions of courts like State Farm, though in a different legal context, give weight to such arguments. 

​The statute of limitations to foreclose on a mortgage in New York is six years.  CPLR 213(4).  This seemingly simple rule has created a great deal of confusion as to when the statute begins to run where the mortgage contains an acceleration clause – a provision that upon default, the lender may convert the entire set of payments into a single owed payment, and then foreclose on the mortgaged property.
 
In the recent decision of Freedom Mortgage Corp. v. Engel (2021), New York’s Court of Appeals consolidated four appeals, and issued an opinion in which it clarified the applicable law in various situations.
 
The Court of Appeals emphasized that the decision presented a situation where “the need for clarity and consistency are at their zenith: contracts affecting real property ownership and the application of the statute of limitations.”  In each of the decision, the need for clarity, simplicity and ease of application drove the outcome.
 
This decision is considered a sea-change over prior law, and numerous lenders have moved to reopen cases dismissed on statute-of-limitations grounds.
 
General Principles
 
Many loans have a schedule of payments that stretch over years.  Typically, when the borrower defaults, the loan agreement will provide that the lender has the option to accelerate the loan, meaning to call it in and have the entire balance due.  “Acceleration in this context is a demand for payment of the outstanding loan in full that terminates the borrower’s right to repay the debt over time through the vehicle of monthly installment payments.”  Acceleration is generally an option for the lender, not something that happens automatically. 
 
In 1932, the Court of Appeals held that, to be valid, an election to accelerate must be made by an “unequivocal overt act” that discloses the noteholder’s choice, such as the filing of a verified complaint seeking foreclosure and containing a sworn statement that the noteholder is demanding repayment of the entire outstanding debt.  A clear letter from the lender to the borrow also suffices.  But what is crucial is that the lender make it crystal clear that acceleration is sought.
 
Acceleration has a major effect on the statute of limitations.  When a borrower misses a payment, then the statute only begins to run with respect to the one missed payment; it does not run with respect to the rest of the payments which are not yet due.  Once an acceleration happens, however, the statute of limitations begins to run as to the entire loan.
 
Specific Rulings
 
The Court of Appeals made specific rulings on various situations involving accelerated loans.  In one case, the lender had previously filed foreclosure actions, but had failed to attach the proper documentation, leaving out modifications to the loan agreement that had been made in the interim.  The borrower succeeded in having these actions dismissed.  The Court of Appeals, reversing the Appellate Division, held that the dismissed action did not count as an acceleration, since the action was defective and the lender succeeded in having it dismissed.
 
In another case, the lender sent a default letter to the lender, noting the failure to make timely payments, proposing a cure period, and then stating that the lender “will accelerate” the loan if payments were not made current.  Resolving a split among the Appellate Divisions, the Court of Appeals held that a future-looking “will accelerate” statement is not itself an acceleration, and does not start the statute of limitations running.
 
In the last two cases, the Court considered whether an acceleration can be withdrawn.  The lenders each filed foreclosure actions that constituted an acceleration, but then voluntarily dismissed them.  Overturing decisions of the Appellate Division, the Court of Appeals held that a voluntary discontinuance automatically revokes the acceleration.  “In such a circumstance, the noteholder’s withdrawal of its only demand for immediate payment of the full outstanding debt, made by the ‘unequivocal overt act’ of filing a foreclosure complaint, ‘destroy[s] the effect’ of the election.”
 
The borrower might still avoid the revocation if it can be shown that “the borrower changed his position in reliance on that election by executing a new mortgage.”  It is therefore possible for an acceleration to be irrevocable.  And, of course, if the lender has done other acts of acceleration, such as an acceleration letter, that might require additional acts of revocation.
 
In the few months since the Engel decision, several Appellate Division decisions have relied on it in various circumstances to deny dismissal or summary judgment of foreclosure complaints.
 
The Engel decision, as it expressly explains, is grounded on the notion that lenders need flexibility to determine whether and when to accelerate a loan.  Often, it is in both the lender’s and borrower’s interest not to accelerate, to allow the parties to work out an accommodation. As in many areas of the law, parties are best served when they are clear about what they are doing and what rights they are exercising. 

The recently passed CARES EXTENSION ACT made several important changes to procedures in the Trademark Office and in federal court.   While some are favorable to trademark owners, others are meant to clear the Trademark Register of deadwood registrations that are not being used. 

Among other things, the new law establishes procedures to expunge the register of marks that have never been used for the goods and services listed (a growing problem with foreign-based registrations); a new basis for cancellation (the mark was never used for some or all of the listed goods and services); a procedure to provide information on a pending application that would be a basis for denial; and restoration of a presumption that trademark infringement creates irreparable harm.

The new law has several different features, not all of which are related.  It not only presents opportunities for both trademark owners and potential trademark infringers, but also presents new challenges and potential pitfalls.

Presumption of Irreparable Harm – Federal courts traditionally presumed that a showing of likelihood of confusion also meant that irreparable harm, a requirement for a preliminary or permanent injunction, should be presumed.  This is because the harm from infringement – confusion as to who makes the goods, and likely affect on good will – are difficult to quantify, and thus do not lend themselves to money damages.

But in a patent case eBay v. MercExchange (2006) the Supreme Court held that there is no presumption of irreparable harm.  Many courts applied that in trademark and copyright cases to mean that there is no presumption of irreparable harm as well.  The new law restores the presumption for trademark cases.

The new law will not likely have a major effect on litigation.  In the vast majority of cases, trademark owners have been able to show irreparable harm without a presumption.

But what trademark owners must be wary of is delay.  The presumption of irreparable harm is rebuttable. Before eBay, one of the most common ways to rebut the presumption was to show the trademark owner delayed.  This is particularly the case for preliminary injunctions; in one often cited Second Circuit case, a delay of only six weeks was enough to rebut the presumption.  So if a trademark owner wants an injunction, it should move quickly to stop the problem.

Expungement Procedures  -- The new law creates two ex parte expungement procedures, one directed to domestic registrations, the other to registrations based on foreign registrations.  Both allow any party to submit a petition to expunge a registration, along with evidence that the mark has never been used in commerce.  The Director of the Trademark Office reviews the petition, and if warranted, initiates a procedure, on notice to the trademark owner, that the registration may be cancelled.

This addresses a problem that has arisen in recent years involving fictitious registrations for marks that are never actually used.   Many such registrations have been obtained by foreign registrants, particularly from China and India.  The motivation seems to be Amazon, which requires a trademark registration in order to register as a merchant.

Parties who actively sell on Amazon often encounter such registrations as blocking their names, and the new procedures will be a way to expeditiously deal with the problem.

Third Party Submissions In Pending Applications
The new law also formalizes what previously was called a letter of protest.  If there is a pending application, anyone may submit evidence relevant to a possible refusal. The Director reviews it, and can include it in the examination file.

This can be an efficient and inexpensive way to oppose pending applications, without the expense and time of a formal opposition or cancellation procedure before the TTAB.  This is a good reason why trademark owners should use a monitoring service to be alerted to pending applications that may come too close to their trademarks.

Non-Use As A New Basis For Cancellation
The new law enacts a new basis for cancellation.  At any time after three years after registration, if one can show that the mark has never been used in commerce for all or even some of the listed goods, the registration is cancelled.  Unlike abandonment, intent is not a defense – no matter the owner's intent, if the mark has never been used, the registration can be cancelled.

This new section is a potential landmine for foreign registrations.  It is common practice in Europe to register for dozens or even hundreds of goods and services, intending to use the mark for only a few.  These registrations are then used to obtain U.S. registrations under either Section 44 of the Trademark Act or through the Madrid Protocol.  But the owner only uses the mark for a few of the listed goods and services.

Under this new basis for cancellation, the entire registration could be cancelled in that circumstance.  So foreign registrants must carefully consider the scope of their registrations, and omit any goods and services they do not intend to use.