For the serious handbag collector, purchasing a Hermès Birkin is a pinnacle achievement. Beyond the eye-watering price tag for the purse alone, being offered a Birkin for purchase is often the result of several thousand dollars in prior sales with Hermès and a positive, established relationship with a salesperson who is authorized to offer Birkin bags to hand-selected customers deemed worthy of carrying a Birkin. The pre-owned market does not offer any price relief, as a Birkin is one of the few fashion items capable of appreciating in value. So, it comes as no surprise that handbag enthusiasts who don’t have the cash for the bag itself, or the time, money, or desire to spend on purchasing other items, look to other options.

And those shoppers with a TikTok account may have learned this past week of a new “option” to source dupes of Birkins and other luxury brand products direct from Chinese manufacturers. In these viral posts, the content creator claims that—in direct response to the U.S. Tariffs—they are exposing many of the world’s costliest luxury goods are made at a sliver of the retail price in China. Many of the videos include links to where individuals can purchase goods directly from the Chinese manufacturer.

To be clear, regardless of whether the goods are purchased directly from a third-party manufacturer that is involved with the designer item’s production, the item purchased is still a counterfeit. But that hasn’t stopped many shoppers from buying $5 leggings said to be produced from the same manufacturer as Lululemon or $50 Louis-Vuitton style bags. As a direct result of these viral videos, the Chinese wholesale marketplace app DHgate jumped to the No. 3 spot of free iPhone apps in the U.S. Apple App Store, from No. 352 less than a week ago.[1] Of course, using DHgate is not a solution for avoiding tariffs, as any Chinese imports to the U.S. will still be impacted by the tariffs and potential seizure by customs officials. But those details may be less important to a buyer who is only shelling out $1,400 for a Birkin dupe instead of the minimum $15,000 for an authentic Hermès purchase.

With this swarm of videos has come the question—how are brands able to identify as being made in one country when a significant portion of the bag is made elsewhere? Much confusion often exists at the consumer level about what the “made in” tag means, with ordinary consumers assuming that if a product is identified as being made in a specific country, say Italy, a consumer may assume that a product bearing a “made in Italy” tag includes all Italian-made materials and craftsmanship. Yet, due to the often-complex supply chain in the fashion industry, there is often not a clear through-line about the country of origin. And the fact that this murkiness exists is no secret. For example, in 2017 the luxury brand Louis Vuitton received heat after an investigative report claimed that near-entire shoes for the Louis Vuitton Moet Hennessy (LVMH) luxury group were made in Romania, with only the soles being added in Italy.[2]

Generally, every article of foreign origin into the U.S. must be marked with its country of origin. The country of origin means the country or territory where the good originates. The marking rules and regulations can vary across different products and industries, particularly when a good’s manufacture begins in one country and is completed in a second country. For textiles and apparel products, the country of origin to be marked is:

(1) The country where the good was wholly obtained or produced;

(2) If (1) is not applicable, the country where the product underwent a change in tariff classification or met other requirements specific to the article, which are specified in 19 USC 102.21 (e);

(3) If (1) nor (2) apply, then the country is where:

a. 50% or more of the exterior surface area is formed by major parts that have been knitted or crocheted directly to the shape used in the good (“knit to shape”); or

b. The good was wholly assembled;

(4) If (1)–(3) do not apply, the country of origin is where the most significant assembly or manufacturing process occurred; OR

(5) If none of the foregoing apply, the country of origin is the last country where an important assembly or manufacturing process occurred.[3]

In luxury brands, the most labor-intensive part of the production, such as elaborate embroidery and embellishments, are done on panels and then assembled to create the finished product in the “made in” location.[4] High end designers, such as French designer Isabel Marant, have acknowledged that much of haute couture originates with Indian artisans.[5]

Considering the murkiness in the marking rules being an open secret for fashion—assuming what they claim is true, why are these manufacturers now speaking up? Based on comments from the content creators, the videos appear to be a response to the U.S.’s ongoing trade war with China after Trump placed 145% tariffs on Chinese imports, to which China retaliated with its own 125% tariffs on U.S.-imported goods.[6] It’s reasonable to believe some are chasing new buyers, but other commentators suggest that the videos are intended to communicate to U.S. consumers how much they actually rely on China for goods they believe originate in other countries. Indeed, the Peterson Institute for International Economics determined in 2024 that 40% of all footwear and 25% of all textiles and clothing imported into the U.S. came from China.[7] Some of the videos also directly attack “made in China” stigma, where Chinese-manufactured goods are treated as synonymous with being cheap or poor quality. Whatever the reason, the claims made should all be taken with a massive grain of salt—as the claims are directly contrary to the full-throated representations made by various European fashion houses and could be merely a well-timed effort by counterfeiters to sell counterfeit luxury items under the guise of being the “same” as a quality product.

Though the claims by content creators may not change the minds of consumers that believe luxury pieces are a worthwhile investment, there is no doubt that these claims continue to draw negative press to luxury brand markups and the actual “worth” of such products. The surge in popularity of apps like DHgate also illustrate how effectively Chinese creators can leverage platforms like TikTok to influence U.S. consumers.[8] A scandal last year about Dior and Armani utilizing sweatshops in Milan threw water on claims that the high price tags are justified by the brands avoiding human rights violations, and Q1 stock reporting from brands like LVMH signal tough times ahead.[9] It’s undoubtable that these videos call into sharp question, for many U.S. consumers, where these luxury brands are actually crafted and whether the “made in” tags actually reflect the quality they have come to expect with the high price tags. It also calls into question what can be done about these luxury counterfeiters that are offering purportedly Chanel Boy Bags and Hermès Birkins without the hefty price tags.

What can be done all comes down to the tricky situation of trademark and trade dress enforcement ability in China. Just last year, luxury retailer Burberry won a significant infringement claim and was awarded £675,668 in damages by the Jiangsu Provincial High People’s Court relating to infringement by the holder of a Chinese trademark registration for Baneberry, under which it sold in approximately 40 pop up shops products bearing Burberry’s iconic checked pattern and a logo resembling Burberry’s Equestrian Knight design.[10] In addition to the infringement, the Chinese court found that Baneberry had registered their trademarks in bad faith. This ruling signified to many an increasing ability to bring actual enforcement action against infringers in China, which, according to the Organization for Economic Cooperation and Development, accounts for more than 80% of the world’s counterfeit goods.[11]

But China’s willingness to support U.S. efforts to weaken counterfeiting are inseparably connected to the two countries’ trading relationship. Beijing has taken steps since 2019 to increase IP enforcement efforts by introducing stricter trademark laws, increasing penalties for infringement, and establishing specialized IP courts. But those efforts could stall as tensions increase, putting at risk the ability to challenge infringers, increased counterfeiting, and increased cyber espionage.[12] There are theories that China is considering giving free reign to counterfeiters that target American brands (such as often-counterfeited brands like Ralph Lauren, Tom Ford, Calvin Klein, Coach (Tapestry), Jimmy Choo (Capri), The Row, Estée Lauder, and La Mer) in response to the trade war. Social media content creators have also noted Zara, H&M, and Stanley as at risk since many of their products are known to originate in China. Such a strategy would not be without precedent. In 2023, in response to U.S. and E.U. sanctions related to the Ukraine invasion, Russia ended the requirement for goods entering Russia to provide a certification of IP ownership of the goods.[13] While the products could still be inspected by local authorities, reports reveal these inspections to be relatively lax and many of the certifications clearly faked.[14] As a result, many counterfeit luxury goods have flooded the market in the EU through Turkey. Also at risk is the possibility of retaliatory measures; China has previously threatened to cancel trademark registrations and patents in response to past trade disputes.[15]

For U.S. companies watching these developments, a few measures can be taken to best arm themselves against these volatile times. First, companies seeking expansion in and manufacturing in China should seek to register patents and trademarks early. China has a “first to file” trademark application process that can allow bad faith actors to apply for trademark applications on established brands, complicating the enforcement process when the time comes. Second, companies should also be sure to invest in their security infrastructure to protect proprietary designs and trade secrets.


[1] Sarah Perez, “Chinese marketplace DHgate becomes a top US app as trade war intensifies”, TechCrunch (Apr. 14, 2025) (https://techcrunch.com/2025/04/14/chinese-marketplace-dhgate-becomes-a-top-us-app-as-trade-war-intensifies/).

[2] Alexandra Lembke, “Revealed: the Romanian site where Louis Vuitton makes its Italian shoes”, The Guardian (Jun. 17, 2017) (https://www.theguardian.com/business/2017/jun/17/revealed-the-romanian-site-where-louis-vuitton-makes-its-italian-shoes).

[3] 19 USC 102.21(c). See also “Marking of Country of Origin on U.S. Imports”, U.S. Customs and Border Protection (last accessed Apr. 15, 2025) (https://www.cbp.gov/trade/rulings/informed-compliance-publications/marking-country-origin-us-imports).

[4] Emily Chan, “What Does the ‘Made In’ Label On Our Clothes Actually Mean?”, British Vogue (Jan. 18, 2024) (https://www.vogue.co.uk/article/what-does-made-in-label-mean-clothes).

[5] Kai Schultz, et al, “Luxury Fashion Relies on Indian Artisans. The Labels Tell a Different Story”, Bloomberg (Sept. 28, 2023) (https://www.bloomberg.com/features/2023-india-luxury-fashion-supply-chain/).

[6] Dhani Mau, “Hey, Quick Question: Why are Chinese Fashion Manufacturers Going Viral on TikTok Amid the Tariffs Chaos?”, Fashionista.com (Apr. 14, 2021) (https://fashionista.com/2025/04/chinese-suppliers-manufacturers-tiktok-luxury-fashion-brands).

[7] Id.

[8] Sarah Kent, “’Trade War TikTok’ Takes Aim at Luxury”, BusinessofFashion.com (Apr. 15, 2025) (https://www.businessoffashion.com/articles/sustainability/trade-war-tiktok-luxury-brands-chinese-factories/).

[9] Id.

[10] Maura O’Malley, “‘An important moment in Chinese IP litigation’ – Burberry celebrates trademark infringement win in ‘key market’”, Global Legal Post (May 8, 2024) (https://www.globallegalpost.com/news/an-important-moment-in-chinese-ip-litigation-burberry-celebrates-trademark-infringement-win-in-key-market-1386955387).

[11] “Behind the U.S.-China Trade War is a High-Stakes IP Battle”, The Fashion Law (Apr. 9, 2025 (https://www.thefashionlaw.com/behind-the-u-s-china-trade-tariffs-is-a-high-stakes-battle-over-ip/).

[12] The Fashion Law, note 11.

[13] Noëmie Leclercq, “Putin weaponises counterfeit luxury goods amid Ukraine war”, Glitz.com (Sept. 2, 2023) (https://www.glitz.paris/en/entourage/2023/02/09/putin-weaponises-counterfeit-luxury-goods-amid-ukraine-war,109911779-evg).

[14] Id.

[15] The Fashion Law, note 11.

On October 10, 2024, during an oral argument, the Federal Circuit was presented with an argument that patent applications, filed before the filing date of the challenged patent but not published until after the filing date of the challenge patent, are not “printed publications.” The panel admitted that the argument carried weight, creating worry in the patent community of potential impacts of such a ruling. However, the Federal Circuit issued its opinion on January 14, 2025, finding that a patent application is prior art and that the effect of the application as prior art stems from its filing, not its publishing date.[1]

The case is Lynk Labs, Inc. v. Samsung Electronics Co., Ltd., which followed a finding of unpatentability by the PTAB in an inter partes review (IPR). The Board’s ruling, challenged by Lynk Labs, found that a patent application filed before Lynk Labs’ priority date, but published after, could be used as prior art to invalidate the challenged claims. The prior art application introduced by Samsung was abandoned and a patent was never issued. Nevertheless, the Federal Circuit agreed with the PTAB’s decision.

Section 311(b) of the America Invents Act (AIA) states that a patent can be challenged in an IPR “only on the basis of prior art consisting of patents or printed publications.”[2] Lynk argued that a patent application that was not published prior to a challenged patent’s filing date is not included in “printed publications” because it was not publicly available at the critical time. Lynk cites to a handful of case law in saying that “the statutory phrase ‘printed publication’ has been interpreted to mean that before the critical date [of the challenged patent] the reference must have been publicly accessible to the public interested in the art.”[3] Lynk argued that had Congress intended patent applications to be used as prior arts in IPR, it would not have expressly left them out as a category in 311(b).[4]

The Federal Circuit agreed with Lynk: “The touchstone of whether a refence constitutes a printed publication is public accessibility.” citing Weber, Inc. v. Provisur Techs., Inc., 92 F.4th 1059, 1067 (Fed. Cir. 2024). However, the Court goes on to critique Lynk’s focus of printed publications under 35 U.S.C. § § 102(a) and 102(b). The Court points out that Congress created a “special rule” in § 102(e)(1)[5] for published patent applications. Under § 102(e)(1), a person shall be entitled to a patent unless “(e) the invention was described in – (1) an application for patent, published under 122(b), by another filed in the United States before the invention by the applicant for patent.” (emphasis added).

Because patent applications are published and publicly accessible, they are printed publications. Further, as stated by the Court, “102(e)(1) treats this type of printed publication as prior art as of a time before it became publicly accessible—i.e., as of its filing date.” The term “printed publications” does not have its own temporal element. Rather, such element is found in the applicable statute for each type of printed publication. The rule for patents as prior art presents similar wording, granting a patent prior art status as of its filing date.[6] Thus, in a similar IPR situation, 311(b) can include as prior art a patent that did not become a patent until after the challenged patent’s priority date.

Ultimately, the Court concluded that “the plain language of §§ 311(b) and 102(e)(1) permits IPR challenges based upon published patent applications, and such published patent applications can be deemed prior art in IPRs as of their filing date.” Additionally, the Court went on to analyze the historical context of 311(b), finding nothing to sway its interpretation. Finally, the Court further confirmed its interpretation, finding that making patent applications available as prior art under 311(b) is fully consistent with Congress’s purpose in limiting the types of patentability challenges in IPRs.

The Federal Circuit’s ruling eases the worries of the patent community and maintains the status quo. A contrary decision would call into question previous IPRs which relied on applications as printed publications to invalidate challenged claims. Fortunately, the Federal Circuit avoided such consequences, following years of precedent and the clear language of the law.


[1] Lynk Labs, Inc. v. Samsung Elecs. Co., Ltd., Case No. 2023-2346, (Fed. Cir. 2025) (23-2346.OPINION.1-14-2025_2450365.pdf).

[2] 35 U.S.C. § 311(b).

[3] In re Klopfenstein, 380 F.3d 1345, 1348 (Fed. Cir. 2004); see also Acceleration Bay, LLC v. Activision Blizzard Inc., 908 F.3d 765, 772 (Fed. Cir. 2018) (explaining that public accessibility is the “touchstone in determining whether a reference constitutes a ‘printed publication,’” and then noting the Board’s finding that the reference “was not publicly accessible before the critical date” (emphasis added) (cleaned up)); In re Lister, 583 F.3d 1307, 1311 (Fed. Cir. 2009) (“In order to qualify as a printed publication within the meaning of § 102, a reference ‘must have been sufficiently accessible to the public interested in the art.’” (citation omitted));

[4] Stephen Schreiner, attorney for Lynk Labs, details his argument at Patent Applications Published After the Priority Date of a Challenged Patent Are Not ‘Printed Publications’ for IPRs.

[5] This case was governed by pre-AIA language.

[6] See pre-AIA 35 U.S.C. § 102(e)(2) and current 35 U.S.C. § 102(d).

This article was originally published by Biz New Orleans.

NEW ORLEANS – Devin Ricci and Mary Love, Intellectual Property Lawyers with Kean Miller LLP, have issued a statement urging businesses and individuals to exercise caution to avoid unauthorized use of NFL-related branding or while streaming game broadcasts to avoid hefty fines or legal action.

Ricci and Love highlighted that the NFL maintains strict control over its intellectual property, including game broadcasts, logos, team names, and slogans. “Many facets of the Big Game are protected by copyright and trademark laws,” Ricci explained. “The NFL actively enforces these rights through measures such as takedown notices, cease-and-desist letters, and lawsuits.”

Businesses planning to stream the game must ensure they have the appropriate licensing. A commercial license is typically required for public screenings in establishments such as bars, restaurants, or event venues. Relying on personal streaming subscriptions for such purposes is a violation of the NFL’s licensing agreements and could lead to significant penalties. Love advised, “Check with your service provider to confirm that your streaming package includes the necessary commercial rights.”

The NFL also owns trademarks on terms like “Super Bowl” and associated branding, which cannot be used for promotional or advertising purposes without explicit permission. “Unauthorized use of NFL trademarks is prohibited at both state and federal levels,” Ricci noted. “Businesses can avoid infringement by opting for general terms like ‘The Big Game’ instead.”

The financial consequences of violations are substantial. Under the Copyright Act, statutory damages range from $750 to $30,000 per work, with potential increases up to $150,000 for willful infringement. Similarly, the Lanham Act allows for damages between $1,000 and $200,000 per counterfeit mark, which can escalate to $2 million for willful violations. Both statutes permit the recovery of attorney’s fees, further increasing the costs of non-compliance.

The lawyers stressed that the NFL is particularly vigilant about protecting its brands during high-profile events like the Super Bowl. For businesses unsure about compliance, consulting legal counsel is strongly recommended. “The costs of obtaining the proper licenses pale in comparison to the potential damages of a legal dispute,” Love said.

Kean Miller LLP’s guidance serves as a reminder of the critical importance of respecting intellectual property rights, particularly as New Orleans gears up for a landmark Super Bowl celebration.

An owner of a trademark or service mark used in commerce may request registration on the principal register by filing an application with United States Patent and Trademark Office.  The registration provides certain benefits including prima facie proof of ownership and validity and constructive use throughout the country.  While registration is great for brand management and protection, once the registration is obtained notice should be provided.  

15 U.S.C. 1111 provides that a registrant may give notice that its mark is registered by displaying with the mark the words “registered U.S. Patent and Trademark Office” or “Reg. U.S. Pat. & Tm. Off.” or the R enclosed within a circle ®.  The statute further provides that in any suit for infringement by a registrant failing to give notice of registration, no profits or damages shall be recovered under the provisions of this chapter unless the defendant had actual notice of the registration.  While this statute does not create a defense in an infringement action, it does limit the remedies a registrant may pursue. A registrant who fails to provide notice of registration may still obtain an injunction but will not be able to recover damages and profits, unless the registrant can prove the defendant had actual notice of the registration.    

A registrant should be sure to put a notice adjacent to the mark when used in connection with the brand of goods or services subject to the registration. The notice should be put on a label, box, or website, as may be appropriate, to give this notice of registration.  If an infringement action becomes necessary, the registrant will want the be able to pursue all possible remedies, and not be limited by registrant’s own failure to give notice.   

In the fall of 2023, the Biden Administration issued the Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence.[1] In section 5.2 of the order, the administration charged the Under Secretary of Commerce for Intellectual Property and Director of the United States Patent and Trademark Office (USTPO) with three directives:

1. Publish guidance to USPTO patent examiners and applicants addressing inventorship and the use of AI, including generative AI, in the inventive process and how inventorship issues ought to be analyzed;

2. Issue additional guidance to address other considerations at the intersection of AI and IP, which could include updated guidance on patent eligibility to address innovation in AI and critical and emerging technologies; and

3. Consult with the Director of the United States Patent Office and issue recommendation to the President on potential executive actions relating to copyright and AI.

Accordingly, the USPTO responded to its first directive and issued its Inventorship Guidance for AI-Assisted Inventions on February 13, 2024.[2] The guidance provides that AI-assisted inventions are not categorically unpatentable. Instead, an analysis utilizing the Pannu factors determines if the human inventor(s) has made a significant contribution to each claim in the patent or patent application. The USPTO also provides examples in its guidance, as well as more specific examples on its website.[3]

In the fall of 2022, the Federal Circuit in Thaler v. Vidal affirmed the holding “that only a natural person can be an inventor.”[4] This holding remains aligned with the policy that patents function to incentivize and reward human ingenuity. However, the court did not answer the question of AI as a joint inventor.

In its guidance, the USPTO makes clear that the term “individual” throughout the MPEP refers to a human being.[5] When naming the inventor(s), the inclusion of AI or any system will render the invention unpatentable.[6] While there is no requirement for a named inventor to contribute to every claim in an application,[7] at least one named inventor must have invented each claim.[8] Therefore, in the case of an individual working solely with an AI system, the individual must significantly contribute to every claim. The guidance tells us that to significantly contribute means to meet the Pannu factors.

Under the Pannu factors, each inventor must: “(1) contribute in some significant manner to the conception or reduction to practice of the invention, (2) make a contribution to the claimed invention that is not insignificant in quality, when that contribution is measured against the dimension of the full invention, and (3) do more than merely explain to the real inventors well-known concepts and/or the current state of the art.”[9] In its footnote, the guidance explains that the mention of reduction to practice in the Pannu factors is an acknowledgement of the simultaneous conception and reduction to practice doctrine used in unpredictable technologies.[10] Applicants are reminded that the main inquiry is that of conception.[11] The Pannu factors are not a basis to conclude that reduction to practice, alone, is sufficient to demonstrate inventorship.[12]

To assist applicants and USPTO personnel in determining inventorship, the guidance provides a non-exhaustive list of “guiding principles” for the application of the Pannu factors in AI-assisted inventions:

1. The use of AI does not negate a person’s contributions as an inventor if the person contributes significantly to the invention.

2. Merely presenting a problem to an AI system may not be a significant contribution. However, a significant contribution could be found in the construction of the prompt in view of a specific problem to elicit a particular solution from the AI.

3.Although reducing an invention to practice alone is not a significant contribution, a person who makes a significant contribution to the output to create an invention may be a proper inventor, even if that person is unable to establish conception until the invention has been reduced to practice.

4. The designing, building, or training of the AI system in view of a specific problem to elicit a particular solution may be a significant contribution to the invention created with the AI system.

5. A person simply owning or overseeing an AI system that is used in the creation of an invention, without providing a significant contribution to the conception of the invention, does not make that person an inventor.[13]

These guiding principles attempt to alleviate some uncertainty surrounding “a significant contribution” under the Pannu factors. Applicants would be wise to keep the Pannu factors, as well as these principles, in mind early in the invention process. In addition, an inventorship analysis will assist an applicant in drafting the claims, as the USPTO requires an inventor (or one of the joint inventors) to significantly contribute to each claim.

As for the various duties of patent owners and applicants, the USPTO claims it is not changing or modifying its duty of disclosure.[14] However, the USPTO reminds those of their existing duty of disclosure and its applicability to the inventorship inquiry. Parties have a duty to disclose to the USPTO information that raises a prima facie case of unpatentability or that is inconsistent with a position an applicant takes in asserting inventorship or opposing its rejection.[15] This information could include evidence of an inventor’s lack of a significant contribution to the invention because the person’s purported contribution(s) was made by an AI system.

According to the USPTO, the duty of reasonable inquiry will remain the same, as well.[16] Identical to FRCP 11(b), the duty calls for an “inquiry reasonable under the circumstances” to ensure that (1) the paper is not being presented for any improper purpose; (2) the legal contentions are warranted by law; (3) the allegations and other factual contentions have evidentiary support; (4) and the denials of factual contentions are warranted on the evidence.[17] As for AI-assisted inventions, this inquiry should include questions about whether and how AI is being used in the creation process. Applicants and patent practitioners should conduct their own assessment of whether the contributions made by the natural person(s) rise to the level of inventorship as expressed in the USPTO’s guidance.

While the continuous evolution of AI will likely require further modifications to the idea of “inventorship,” this guidance will likely lead the courts’ analyses for now. The USPTO’s five guiding principles given above aim to counteract the ambiguity of a significant contribution in the Pannu factors. Applicants and practitioners should utilize these principles early in the application process as they are likely to appear in the litigation of AI-assisted inventions. For further assistance, the USPTO provides more specific examples of significant and insignificant contributions on its website.[18] The period for public comment on this guidance will remain open until May 13, 2024.


Special thanks to Michael Doggett, LSU Law Class of 2024, for his research and assistance in writing this article.

[1] Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence | The White House

[2] 89 Fed. Reg. 10043 (Feb. 13, 2024).

[3] AI-related resources | USPTO (Example 1 and Example 2 under “Inventorship”).

[4] Thaler v. Vidal, 43 F.4th 1207, 1213 (Fed. Cir. 2022), cert denied, 143 S. Ct. 1783 (2023).

[5] 89 Fed. Reg. 10046 (Feb. 13, 2024) (citing Thaler v. Vidal, 43 F.4th at 1211).

[6] Id.

[7] MPEP 2109.01.

[8] 35 U.S.C. 115(a) (“An application for patent that is filed under section 111(a) or commences the national stage under section 371 shall include, or be amended to include, the name of the inventor for any invention claimed in the application.”) (emphasis added).

[9] Pannu v. Iolab Corp, 155 F.3d 1344, 1351 (Fed. Cir. 1998).

[10] 89 Fed. Reg. 10047, fn.32 (Feb. 13, 2024).

[11] Id.

[12] Id.

[13] 89 Fed. Reg. 10048 (Feb. 13, 2024) (paraphrased).

[14] 89 Fed. Reg. 10049 (Feb. 13, 2024).

[15] 35 U.S.C. 101.

[16] 89 Fed. Reg. 10050 (Feb. 13, 2024).

[17] 37 CFR 11.18(b)(2).

[18] AI-related resources | USPTO (Example 1 and Example 2 under “Inventorship”).

On December 11, 2023, the Copyright Review Board affirmed the Copyright Office’s decision to reject Ankit Sahni’s application to register the AI-generated work depicted above.  The artwork, entitled “SURYAST” was based on a photograph taken by Sanhi, which Sanhi submitted into the RAGHAV Artificial Intelligence Painting App (“RAGHAV”) and instructed to paint in the style of Vincent van Gogh’s “Starry Night”.  Sanhi submitted to the Copyright Office that he also chose “a variable value determining the amount of style transfer” for the program to apply. In effect, Sanhi was attempting to register the artwork as a derivative of his photograph.

The Copyright Office initially rejected the application on the premise that it lacked the human authorship necessary to support a copyright claim, specifically rejecting Sanhi’s assertions that the work contained some human creative input. Instead, the Copyright Office stated that the “human authorship cannot be distinguished or separated from the final work produced by the computer program.”

Sanhi sought reconsideration, which the Copyright Office rejected again.  The Copyright Office opined that the work was a classic example of a derivative work in that it was a digitalization of a photograph. Because copyright in derivative works is in the transformative elements of the work (rather than the work as a whole), current guidance would require that the derivative work’s transformative elements be created by a human author.  Under this framework, the Office noted that derivative works are analyzed to determine whether the new authorship of the derivative work meets the statutory requirements for protection. Here, the Office found that the RAGHAV app contributed the new elements and, therefore, the derivative authorship was not the result of human creativity or authorship.

On July 10, 2023, Sanhi requested a second reconsideration, this time asserting the human elements of the work and attempting to downplay RAGHAV’s role as that of an “assistive software tool” akin to Adobe photo editing software. The Board also rejected these arguments, but its reasoning is instructive as to how AI-generated work may be found to be protectable in the future.

The opinion heavily cites the relatively recent decision by the U.S. District Court for the District of Columbia in Thaler v. Perlmutter, 2023 WL 5333236, which reasoned that the originator of a copyrightable work must be human to be protectable under the law.[1] For more information on the Thaler decision, please see our prior article at Thaler v. Shira Perlmutter, et al.: The Intersection of Human Control Over Artificial Intelligence and Human Authorship as a Necessary Requirement of Copyright .

The Board also cites to recent guidance promulgated by the Copyright Office. In effect, when analyzing AI-generated material, the Copyright Office must decide whether the human or the computer program is the “creator” of the generated work.

[W]hether the ‘work’ is basically one of human authorship, with the computer [or other device] merely being an assisting instrument, or whether the traditional elements of authorship in the work (literary, artistic, or musical expression or elements of selection, arrangement, etc.) were actually conceived and executed not by man but by a machine.[2]

This determination must be made on a case-by-case basis.  If all of a work’s “traditional elements of authorship” are generated by AI, the work cannot be registered due to a lack of human authorship.  However, if the work containing AI-generated material also contains sufficient human authorship to support a claim to copyright, then the Office will register the human’s contributions. For this reason, an applicant seeking to protect the human elements of a co-generated work should submit a limitation in the copyright application noting which materials are AI-generated and not protectable.

The decision by the board was very specific to the facts presented. Sanhi made a compelling argument to aggrandize his involvement, captioning it as “conceiving, creating and selecting an original [base] image,” “selection of the style image,” and “selecting a specific variable value determining the amount and manner of style transfer” which “cumulatively resulted in the [Work], which is the direct outcome of [his] creative expression and contribution.” But the Board was unconvinced. Rather, the Board simplified his contributions to three inputs (selecting a photo, providing a style, and providing a variable to the amount of style to which the program would apply), which ultimately did not have any control over where elements were placed or how they would ultimately appear. The Board rebuffed Sanhi’s argument that RAGHAV acted akin to “a camera, digital tablet, or a photo-editing software program,” instead finding that the program (and not Sanhi) made the determinative and creative steps in creating the work. The important takeaway is that Sanhi’s selection of inputs does not render the work copyrightable. Copyright law only protects the particular expression of an idea and not the idea itself.  Thus, the  inputs represent an unprotectable idea or concept, and not the particular expression of that idea, which was created by the AI program.

This opinion stands as another warning to those creating works with generative artificial intelligence. The Copyright Office and Courts continue to rule against the protection of works created by AI. As these opinions are rendered, we hope a line will solidify definitively showing what is protectable. In the meantime, it should be noted that nothing prevents Sanhi from registering his original photograph which he took with a camera. 

The Copyright Review Board’s decision is available at https://copyright.gov/rulings-filings/review-board/docs/SURYAST.pdf.


[1] See id. At *4 (“By its plain text, the 1976 Act . . . requires a copyrightable work to have an originator with the capacity for intellectual, creative, or artistic labor. Must that originator be a human being to claim copyright protection? The answer is ‘yes.’”)

[2] Copyright Registration Guidance: Works Containing Material Generated by Artificial Intelligence, 88 Fed. Reg. 16,190, 16,192 (Mar. 16, 2023) (quoting U.S. COPYRIGHT OFFICE, SIXTY-EIGHTH ANNUAL REPORT OF THE REGISTER OF COPYRIGHTS FOR THE FISCAL YEAR ENDING JUNE 30, 1965, 5 (1966));

The big patent news this holiday season involves a purported ban on the new Apple Watch. In October 2023, the International Trade Commission (the “ITC”) ruled that the blood oxygen monitoring features in these devices infringed a patent from Masimo, a medical device company. (Inv. No. 337-TA-1266). In response, Apple has announced that it will stop selling the affected watches at its stores and online effective December 24, 2023. This ruling will impact Apple’s watch sales, which news outlets report to be an estimated $16 billion dollar industry. Why then do the headlines on this story notably lack the large damage awards one would typically see in a finding of patent infringement? The answer highlights some of the pros and cons of using the ITC to enforce patents over federal district courts.

The ITC is a quasi-judicial federal agency with broad investigative responsibilities on matters of international trade. An increasingly popular function of the ITC is serving as an alternate forum to litigate intellectual property disputes arising under Section 337 of the Tariff Act of 1930, as amended (19 U.S.C. § 1337), commonly referred to as a “Section 337 investigation.” Section 337 vests the ITC with authority to investigate and issue decisions involving importation of articles that constitute unfair competition, which includes the importation of articles infringing on a valid patent, trademark, or copyright.

The ITC provides a lot of practical benefits over a traditional district court litigation. Section 337 investigations are assigned to an Administrative Law Judge (“ALJ”) that oversees the entirety of the proceeding, beginning once the ITC’s investigation begins. Compared to district court litigation, which can (and often does) drag on for years, a trial-like hearing in ITC proceedings generally occurs within one year from the IP owner’s complaint of unfair competition. Indeed, here, Masimo obtained the ultimate ban in less than two years from instituting the proceeding. And it did so in front of administrative law judges and not a lay jury which is likely to have some semblance of bias for Apple and its products. This has saved Masimo a lot of time and resources.

However, the ITC is very limited in the types of relief it can grant. The headlines on the Apple Watch ban notably lack any discussion about damages because the ITC has no authority to award damages for infringement. The ITC only has the power to grant injunctive relief, generally in the form of an Exclusion Order that prohibits the importation of infringing goods into the United States. Specifically, the ITC has no jurisdiction over products that are already in the United States. Only a U.S. District court can enjoin the sale of goods within the United States. As such, the ban does not prevent Apple or its suppliers from continuing to sell products that have already been imported into or otherwise are manufactured in the U.S.

If the Final Determination includes a finding of infringement, like it did against Apple, the ITC has the power to issue an Exclusion Order, either Limited (“LEO”) or General (“GEO”). A LEO excludes the importation of infringing goods by the infringing party named in the Section 337 investigation. A GEO, however, provides for the widespread exclusion of the importation of infringing goods regardless of the importing party’s identity or involvement in the Section 337 investigation. The orders are provided to U.S. Customs and Border Protection (i.e., “Customs”), which is tasked with the enforcement of these Exclusion Orders and preventing the importation of infringing goods into the U.S. Exclusion Orders are prospective in nature, however, and do not impact the use or sale of goods that are already in the United States.

Finally, because the ITC is part of the executive branch, the U.S. President has the power to disapprove a ban. The President has 60 days from the order to review and effectively veto the order. While this veto power has only been used five times in the history of the ITC, it was last used in 2013 when President Obama overturned a ban related to the Apple iPhone. (Inv. No. 337-TA-794). If the President does not veto the ban, it will be finalized, and Apple will be entitled to appeal the determination to the Federal Circuit.

This blog is an update to “Legal Issues with Using AI to Create Content – Written with Help from AIby Devin Ricci on April 28, 2023

On August 18th, the United States District Court for the District of Columbia issued an opinion stating that Artificial Intelligence (AI) generated artwork lacks “human authorship,” thus it cannot be the subject of a valid copyright claim. This decision raises many issues regarding copyright ownership that will require further court involvement and/or policy reform.

The primary challenge arising from AI-generated artwork pertains to copyright existence and ownership. Copyright law traditionally assigns authorship to individuals who create original works. However, in the case of AI, determining authorship becomes complex. Some argue that since AI systems are essentially tools programmed by humans, the programmers should retain authorship rights. Others believe that if AI can autonomously create something new without direct human intervention, it should be granted certain rights. This debate challenges the very essence of copyright law, which is built around the concept of human creativity.

Case Summary

The plaintiff, Stephen Thaler, used the “Creativity Machine,” a generative AI technology, to generate a piece of artwork. Thaler was unsuccessful with obtaining a copyright registration for the AI-generated artwork. In the copyright application, Thaler identified “Creativity Machine” as the author. The United States Copyright Office (“USCO”) denied the application because the work “lack[ed] the human authorship necessary to support a copyright claim.” Thereafter, Thaler filed a complaint in the D.C. District Court against the USCO and its director requesting the refusal be set aside and the AI-generated artwork be registered.

Thaler filed a motion for summary judgment arguing that AI-generated work is copyrightable because the Copyright Act provides protection to “original works of authorship.” This argument was premised on Thaler’s assertion that “author” is not explicitly defined in the Copyright Act and that the ordinary meaning of “author” encompasses generative AI. Ultimately, the D.C. District Court disagreed. The Court held the Copyright Act plainly requires human authorship. As explained by the Court, an “author” is “an originator with the capacity for intellectual, creative, or artistic labor,” which is necessarily a human being.

Implications and Considerations

This decision raises a host of questions and demonstrates that a more comprehensive legal framework is required as AI generated content becomes more sophisticated and prevalent. AI has revolutionized various industries, and the realm of creative expression is no exception. AI-generated artwork has gained significant attention in recent years, raising fascinating questions about the intersection of technology, creativity, and intellectual property rights. As AI systems create artwork independently, it becomes imperative to analyze the implications of this emerging trend on copyright, ownership, and the very definition of creativity. The legal framework is continually evolving and there are many issues that content creators, artists, and marketing companies need to be cognizant of as the legal framework develops.

If this ruling is upheld, a work created solely by AI theoretically is not susceptible to copyright protection at all. Because copyright law is preemptive, meaning it exclusively governs the subject matter of claims that fall within the purview of the Copyright Act, this could severely limit the ability to prevent infringement of an AI-generated work. In theory, because the work would not be protectable, there is no property right to infringe and may not be a legal basis to prevent third party use of the material. 

It is important to note that this recent decision may not stretch to underlying works created by a human, or to the extent a human could be considered a co-author of AI-generated content. In any event, it does implicate works where AI is fully creating the work with little to no human involvement. For example, if you use a program similar to the Creativity Machine and type into the program: “create a picture of Santa getting run over by a reindeer with cookies flying everywhere and a dog laughing,” the resulting image would not be protectable under this decision. In particular, advertising companies should be aware that AI-generated advertisements may not be subject to copyright protection.

However, there must be some middle ground between complete human authorship and complete AI-generated content. AI might be utilized in developing a work, but if there is enough human involvement it should be fair to say there is human authorship. Perhaps a photographer snaps a photograph and uses an AI editing tool to filter/edit the photo. Is the photographer’s involvement enough to make the edited photo a human-authored work? How much human involvement is required to constitute authorship? Courts will have to wrestle with the intersection of AI’s involvement in creative works to sort out these questions. Otherwise, it will be up to Congress to create a new framework for addressing AI generated or augmented works.

Conclusion

AI-generated artwork represents a groundbreaking fusion of technology and creativity that challenges established norms in the art and legal worlds. The complex questions it raises about copyright, authorship, and the essence of creativity underscore the need for collaborative efforts among legal experts, artists, programmers, and policymakers. Balancing the rights of human creators and the capabilities of AI will shape the future landscape of artistic expression and intellectual property rights.

On June 29, 2023, the Supreme Court adopted a restrictive view of the extraterritorial application of the Lanham Act, holding that federal trademark law cannot support a claim for trademark infringement against solely foreign conduct.

The case is Abitron Austria GmbH v. Hetronic International, Inc. Hetronic, an Oklahoma-based corporation, sells a wide range of radio remote controls in over forty-five countries, all of which are marked by their distinctive black-and-yellow trade dress. In 2006, Hetronic entered distribution and licensing agreements with Hydronic Steuersysteme GmbH (later purchased by Abitron Austria GmbH). These agreements authorized Hydronic to build and sell Hetronic-branded products so long as the parts were purchased directly from Hetronic.  In 2011, Hydronic began reverse-engineering Hetronic parts and contracted new suppliers to source them. When Abitron purchased Hydronic in 2014, they began selling products identical to Hetronic’s remote controls in foreign markets with the recognizable black-and-yellow coloring. Prior to litigation, Abitron’s global sales revenue from infringing products amounted to over $90 million, of which $240,000 came from U.S. sales.

A jury in the District Court for the Western District of Oklahoma awarded Hetronic $90 million in damages for Lanham Act violations to account for their lost revenue in both the U.S. and abroad. The District Court also granted Hetronic’s motion for a permanent injunction against Abitron’s infringement activities worldwide. The Tenth Circuit affirmed the injunction but narrowed it to the countries in which Hetronic marketed and sold products. 

The Supreme Court granted certiorari on the issue of whether the Lanham Act applies to infringing uses of trademarks outside of the United States. To assess the issue, the Court applied a two-step framework. The Court first asked if the statute explicitly applies to foreign conduct. The Lanham Act applies to persons who, without the consent of the trademark registrant, “use in commerce any reproduction, counterfeit, copy, or colorable imitation of a registered mark…[if] such use is likely to cause confusion…”. No mention is made of foreign versus domestic use of a mark, so the Court found the Lanham Act did not explicitly indicate it applies to foreign conduct.

Second, the Court considered the “focus” of the statute and whether the “conduct relevant to the focus” occurred in the U.S. Hetronic argued that the goal of the Lanham Act is to avoid consumer confusion or mistake, and so long as consumer confusion is likely to occur in the U.S., it does not matter whether the infringing products are sold in the U.S. or abroad. In response, Abitron contended – and the Supreme Court ultimately agreed – that the focus of the Lanham Act is to avoid the unauthorized use of a trademark in U.S. commerce, meaning that profits lost to trademark infringers abroad cannot be recovered in U.S. courts. “Use in commerce” is the “dividing line between foreign and domestic applications of these Lanham Act provisions.” Therefore, the basis of Abitron’s liability for infringing product sales is limited to the goods sold directly to U.S. customers, a mere $240,000.

The Court unanimously agreed that the infringement scope was limited to domestic activity but split 5-4 on the issue of when an infringing activity becomes “domestic”. Writing for the majority, Justice Alito opined that an infringing use is only domestic when the use in commerce occurs within a U.S. state or territory. Justice Sotomayor penned a concurrence in which she agreed with the overall finding but suggested that the trademark infringement provisions of the Lanham Act could extend to foreign sales when there is a likelihood of U.S. consumer confusion. 

During oral argument Justice Jackson posed hypothetical questions about how the Lanham Act could apply to foreign sales of infringing goods later resold in the U.S., such as when an American traveler purchases counterfeit designer purses in another country and then resells them inside of the U.S. Justice Jackson continues this line of thinking, writing that the Lanham Act could apply to a foreign company whose goods are resold in the U.S. or who engages in online activities that penetrate U.S. commerce without a physical presence in the country.

In many ways, the decision is unsurprising. The Supreme Court’s decision aligns with the international norm of each country granting and enforcing their own trademark rights within their own borders. In the future, U.S. trademark owners will need to seek relief in foreign jurisdictions if the infringement occurs outside of the U.S. The Supreme Court’s focus on “use in commerce” also aligns with prior U.S. case law in which courts have established jurisdiction over foreign companies that sold infringing goods in the U.S. Yet, the opinion also leaves open questions. The majority opinion explicitly limited the scope of the decision, stating that the case did not present the opportunity to deeply explore the meaning of “use in commerce”, so the particulars of that phrase will be left to the lower courts, leaving some uncertainty as to what would constitute actual participation in domestic commerce.

Read the opinion here.


Special thanks to Mattie Kleinpeter, Tulane Law Class of 2024, for her research and assistance in writing this article.

United States District Court Judge P. Kevin Castel issued an opinion on June 22, 2023, imposing sanctions and other penalties on the attorneys who relied on the artificial intelligence application, ChatGPT, in citing to fake cases in pleadings submitted to the court earlier this year.

Judge Castel’s thirty-four page opinion details the missteps of the lawyers, including filing the submission citing the fake cases, failing to withdraw the submission after opposing counsel identified the fake cases, doubling down on the existence of such cases when their validity was called into question, offering false information to the court in order to obtain an extension to a court ordered deadline, and providing “shifting and contradictory explanations” as to how and why the bogus case citations were submitted to the court.

The opinion makes clear that the real issue was the fact that the lawyers continued to mislead the court. Judge Castel wrote that, had the lawyers come clean about their actions shortly after they received the opposing parties’ brief questioning the existence of the cases, the outcome may have been different. While “poor and sloppy research” would amount to mere “objectively unreasonable” actions, the court found that the lawyers acted with subjective bad faith in violation of Federal Rule of Civil Procedure 11 based on all the subsequent failures to disclose.

Luckily for the offending attorneys, the court found that their submission of the fake opinions did not constitute a violation of 18 U.S.C. § 505. The statute states that it is a crime to knowingly forge the signature of a United States Judge or the seal of a federal court. Because the fake opinions did not include a signature or seal, the statute was not violated. But the court noted that the submission of fake opinions raises concerns with protecting the integrity of federal judicial proceedings and is an abuse of the adversary system.

Ultimately, the implicated lawyers were required to pay a penalty of $5,000 and to send letters to each individual judge falsely identified as the author of the fake opinions. Judge Castel’s opinion is a reminder that while a court may be forgiving of a lawyer’s choice to cut corners, lying to the court is still sanctionable.


Cite: Mata v. Avianca, Inc., No. 22-cv-1461, slip op. (S.D. NY. June 22, 2023).