by Samantha O’Connell*
Originating in common law privacy torts, the right of publicity now protects the economic and reputational value of identity instead. However, current holdings by courts have produced vague and inconsistent standards, particularly in the age of AI-generated deepfakes. This Contribution argues that the right of publicity should be re-oriented from copyright-inspired frameworks, such as the transformative use test, toward the trademark-based analysis established in Rogers v. Grimaldi which is centered on misattribution of source. A trademark-inspired analysis of right of publicity cases would yield more consistent results while more clearly targeting the core harms of this tort: misappropriation of goodwill and unjust enrichment. By emphasizing consumer perception over subjective aesthetic judgment, the Rogers test fills loopholes in the current doctrine that infringers might exploit, safeguards valuable social commentary, and better equips courts to address new forms of identity exploitation in digital media.
The right of publicity is, at its most basic, a doctrine that enables an individual to recover for the exploitative use of their identity. Originally an outgrowth of the common law right of privacy, it has steadily developed into an independent cause of action concerned less with dignity and more with the economic value of persona.1 This persona can take the form of a celebrity’s literal name and image or more abstract signifiers including their catchphrase, famous attire, or details such as a jersey number that point to a specific individual. The American Law Institute—an influential organization of scholars, judges, and lawyers dedicated to clarifying and summarizing common law rules—recommends liability against “one who appropriates the commercial value of a person’s identity by using without consent the person’s name, likeness, or other indicia of identity for purposes of trade.”2 The actual tort, however, is enforced by a patchwork of judicial holdings and state statutes without any federal counterpart.3
High-profile disputes, such as Scarlett Johansson’s objections to OpenAI’s chatbot using a voice “eerily similar” to her own, highlight both the continued relevance of the right of publicity and the uncertainty surrounding its boundaries in the age of artificial intelligence (“AI”).4 More broadly, what was once a problem largely cabined to celebrities is now a risk for anyone online, as the internet and social media advertising make identity ripe for exploitation.5
A central difficulty of these cases lies in balancing the plaintiff’s intellectual property claims against the right to free expression. When a company uses a celebrity in an advertisement for a product unrelated to the individual’s persona and without consent, the misappropriation is clear and actionable.6 That would be an obvious example of commercial speech—a communication that does no more than propose a commercial transaction and thus is subject to lower constitutional protection.7 However, the lines blur in cases where the identity supposedly being appropriated exists within expressive works that no longer qualify as commercial speech. As the Ninth Circuit emphasized in Sarver v. Chartier, the First Amendment “safeguards the storytellers and artists who take the raw materials of life—including the stories of real individuals, ordinary or extraordinary—and transform them into art, be it articles, books, movies, or plays.”8
To ensure the right of publicity never morphs into outright censorship, the Supreme Court has explicitly recognized an affirmative defense rooted in free speech doctrine that counters a prima facie misappropriation claim.9 Under this free speech defense, liability is not automatically imposed if a defendant has used someone else’s identity (e.g. their name, likeness, voice, or famous catchphrase) and profited from doing so.10 Instead, lower courts have experimented with a variety of multifactor balancing tests to weigh the two parties’ competing interests of property rights and free speech.11 Two main camps for balancing the First Amendment against the proprietary interest in persona have emerged: jurisdictions that have adopted and apply a copyright-inspired test, and those that apply a trademark-inspired one. If a defendant can prove their work is shielded under whichever version of the test a given jurisdiction recognizes as a matter of law, their work is protected, and the tort claim is dismissed.12
This Contribution argues that courts should ground their First Amendment analysis firmly in trademark principles and abandon the copyright analogies. Clarifying courts’ focus in this way not only more efficiently targets the core harm the right of publicity seeks to prevent—unjust enrichment—but it also mitigates the risk of overextending the right to infringe on protected uses like social commentary and parody.13 By focusing on how the average consumer would perceive a work’s source rather than scrutinizing its aesthetic qualities, judges would also have greater flexibility to grapple with the real-world challenges posed by generative AI.
The “transformative use” test, the majority approach first articulated in California, borrows heavily from copyright’s fair use doctrine—a move that has proven misguided in practice.14 When applying the transformative use test, courts ask whether a defendant has added significant creative elements to convert the work into something more than a mere imitation of the original persona.15 While the origin case, Comedy III Prods., Inc. v. Gary Saderup, Inc., involved a realistic drawing of the “Three Stooges” printed on shirts, the test has since been invoked in cases involving everything from television characters to greeting cards.16 Relevant factors considered by courts to determine whether sufficient creative elements have been added to the work include: the qualitative differences between the celebrity’s actual likeness and the defendant’s use; the percentage of the work attributable to the celebrity’s likeness versus the author’s creativity; and the extent to which the marketability of the work can be attributed to the celebrity’s fame.17 If the claim survives a motion to dismiss or summary judgment under this initial transformative test, a trier of fact will then resolve the question of whether the work is sufficiently transformative.18
The Rogers test, originating in the famous trademark case of Rogers v. Grimaldi, has since been applied in a minority of right of publicity cases as well.19 Under this test, a use of persona is presumptively protected by the First Amendment unless it lacks any artistic relevance to the work or misleads consumers into believing the person endorsed or was involved with the work.20 Certain courts, such as the Third Circuit, disregard this test as “a blunt instrument,” and in so doing mistake its clarity for oversimplicity.21 This Contribution seeks to correct this misconception perpetuated by a majority of circuits.
Conceptually, the right of publicity protects against unauthorized use of a valuable asset, much like property law protects against trespass.22 The doctrine’s logical foundation, thus, aligns quite naturally with trademark. Trademark law protects words, symbols, or designs that identify and distinguish the source of goods or services, granting the trademark holder the exclusive right to use the mark in commerce and to bring infringement actions against others whose use is likely to cause consumer confusion.23 A celebrity’s name or likeness, like a trademark, conveys a set of associations developed over time. Unauthorized use can both divert economic value to a company without the individual’s consent and deceive the public.24 Thus, the accumulation of goodwill, reputational capital, and the communicative function of symbols in the marketplace are critical to both these forms of intellectual property.
Copyright law and the transformative use test, by contrast, provide an awkward analog to the right of publicity. The emphasis on expressive originality does not neatly map onto the appropriation of one’s identity. A persona, after all, does not have to be entirely new to be identifiable. Unlike the right of publicity, copyright’s justification is less individualistic, grounded instead in the utilitarian goal of “promot[ing] the Progress of Science and useful Arts” by incentivizing the creation of additional works for the public’s benefit.25 However, unlike the inherent benefit in more written work protected by copyright law, the goal of the right of publicity is not to promote as many celebrities as possible and overcrowd the market. This theoretical mismatch plays out in practice as well. Despite a majority of jurisdictions embracing the transformative use test, its application has proven uneven and ill-suited to emerging digital forms of identity appropriation.
Courts’ inconsistent efforts to resolve a surge of “name, image, likeness” cases regarding athlete avatars in video games using the transformative use test underscores its volatility and inconsistency.26 Take, for example, when Electronic Arts Inc. was sued by college athletes in two different jurisdictions over its game “NCAA Football,” and the district courts reached opposite conclusions.27 Though both courts purported to apply the transformative use test to the exact same facts, the District Court of New Jersey found that the game deserved First Amendment protection while the District Court of Northern California did not.28 The New Jersey court emphasized that allowing video game players to customize the weight, height, and accessories of the avatars rendered the game sufficiently transformative—a detail the California court considered immaterial and unworthy of consideration.29 Instead, the California court’s analysis hinged on whether the video game avatar performed the same actions that garnered the celebrity athlete’s fame: playing collegiate football on a standard field.30
While the Third and Ninth Circuit resolved this split on appeal, both siding with the plaintiffs, the initial divergence foreshadows the difficulty of applying a test that originated in the context of a T-shirt design to the complexity of digital media.31 With the rise of AI-powered deepfakes that generate images or replicate voices with unprecedented precision, these line-drawing challenges will only be magnified.32 Courts will be forced to consider: Should customizability and interactivity matter? How can the original likeness predominate in a system made of millions of lines of code? How important is it that the likeness be placed in a whimsical setting separate from the one with which the celebrity is actually associated? If the replica looks nothing like the celebrity yet moves and talks identically to them, can they bring a viable right of publicity claim?
Consider how both tests could be applied to the following hypothetical: An aspiring songwriter uses AI to generate a track in which Beyoncé’s voice performs his original lyrics in a musical style radically different from her usual R&B repertoire. The track is a huge commercial success, sending the songwriter’s previously obscure Spotify page to a top page with fifty million streams in one month. Fans flood Beyoncé’s social media accounts, praising her bold new single. She subsequently sues the songwriter for appropriating her voice without permission.
The two leading voice appropriation cases—Midler v. Ford Motor Co. and Waits v. Frito-Lay, Inc—would provide the court limited guidance here as both involved impersonators mimicking not just plaintiffs’ voices but also replicating their most well-known songs.33 Turning instead to the transformative use framework, the hypothetical songwriter might argue that the AI-generated track is not a reproduction of Beyoncé’s existing recordings, but instead features original lyrics, new instrumental backing, and reimagines her voice model in a dramatically different musical context. After all, this conception of the transformative use test has protected works such as the portrayal of musicians Johnny and Edgar Winter as half-human, half-worm comic book characters, as well as an artistic tableau of Tiger Woods that reflected the painter’s interpretation of the golfer’s historic Masters victory.34 By the same logic, a court might reasonably conclude that Beyoncé’s persona here has been “so transformed that it has become primarily the defendant’s own expression rather than the celebrity’s likeness.”35 A court may thus hold under this test that the percentage of elements attributable to the defendant—the lyrics, the melody, and the backtrack—outweigh the sole element Beyoncé has contributed—her recognizable vocal quality.
Equity and common sense, however, point in the opposite direction. Strong evidence suggests that the AI song in this hypothetical gained traction not because of the songwriter’s own creativity, but because listeners were misled into believing Beyoncé was actually involved in the musical production. This is precisely the kind of unjust enrichment the right of publicity is meant to prevent. Importantly, this approach does not discard expressive analysis entirely. After all, the first prong of the Rogers test still requires an examination of the artistic message of the work.36 Rather, the second prong also adds a critical new layer that the copyright test lacks by considering how consumers will actually respond to the work. In the hypothetical, evidence of actual consumer confusion, the defendant’s bad faith, and the likely overlap of these two products on the market would weigh decisively in Beyoncé’s favor, allowing the court to decidedly and uniformly stamp out this exploitative use of AI more readily. Even if the defendant could successfully argue that Beyoncé’s voice was crucial to his artistic vision, he would struggle to overcome the second prong of the Rogers test considering the concrete evidence of consumer confusion. Applying the Rogers framework, therefore, would allow courts to resolve difficult issues like this one without requiring judges to speculatively quantify aesthetic qualities or to stretch the transformative use test to reach a fair result.
Importantly, a shift toward a trademark-based framework would not automatically favor plaintiffs or suppress constitutionally privileged uses of AI technologies. On the contrary, it would better safeguard subtler forms of social commentary that might nevertheless fall through the cracks of the transformative use test.37 As legal scholars have correctly cautioned, “[F]or portrayals that seek accuracy rather than distortion, [or] that seek to pay tribute rather than to criticize, the transformative use test will rarely provide a defense against a right of publicity claim.”38
An inverted version of the earlier hypothetical’s facts demonstrates the sufficiency of these safeguards. Imagine a stand-up comedian uses generative AI to produce a fake press conference video in which Mark Zuckerberg monotonously declares, “At Meta, our mission is clear: every birthday party will be attended only by holograms, every grandmother replaced with a chatbot, and every human friendship monetized at $9.99 a month.” Nothing about Zuckerberg’s appearance has been altered, and the footage resembles real-life news coverage of the CEO. The comedian plays the clip at his show, and it gets a good laugh.
Under this hypothetical, the transformative use test might not provide this comedian with an effective affirmative defense. Since the parody video directly recreates key features of Zuckerberg’s identity, like his appearance and voice, a court could find the work appropriates the “sum and substance” of his persona.39 The “new expression” here comes almost entirely from the script, not from how the celebrity’s persona is visually or stylistically altered.40 A court would therefore likely follow the decision in No Doubt v. Activision, in which the band No Doubt sued over how their avatars were used in the video game “Band Hero.”41 Although video game players could use the cartoon replicas to play any song in the game, not just those by No Doubt, in unrealistic settings like outer space, the use was found not sufficiently transformative.42 The court’s emphasis that the “avatars perform rock songs, the same activity by which the band achieved and maintains its fame” could equally apply to the hypothetical parody, which portrays a clip of the CEO of Meta carrying out his typical job function as the company’s spokesperson.43 Under the transformative use test, this satirical commentary and others like it may lose First Amendment protections.
Applying the Rogers test to the comedian hypothetical would once again rectify the situation and ensure that parodies are entitled to appropriate First Amendment protection, even when they opt for a more realistic style. As the Supreme Court of California articulated, “Once the celebrity thrusts himself or herself forward into the limelight, the First Amendment dictates that the right to comment on, parody, lampoon, and make other expressive uses of the celebrity image must be given broad scope.”44
Various trademark cases involving the toy company Mattel demonstrate how the Rogers test’s form of First Amendment analysis shields parodies that largely replicate the marks or trade dress they seek to lampoon. In one, Mattel sought to distance itself from the sexualized lyrics of Aqua’s “Barbie Girl,” worried that the public might attribute the song to the family-friendly brand.45 Mattel likewise took issue with a series of images by photographer Thomas Forsythe depicting nude Barbie dolls being attacked by household appliances.46 In both instances, however, the Ninth Circuit found these tongue-in-cheek works qualified for First Amendment protection under the Rogers test even though the word mark “Barbie” and the actual Barbie doll product were directly used.47 The contexts in which Mattel’s intellectual property was placed made it so consumer confusion was implausible—much like in the Zuckerberg hypothetical. Just as no reasonable consumer would assume Mattel had produced an innuendo-filled song or photographs showcasing the destruction of their dolls, no reasonable viewer would conclude that Meta is literally replacing grandmothers with chatbots, for instance.
As the Supreme Court has recognized, a parody must communicate two simultaneous messages: It must “conjure up enough of [an] original to make the object of its critical wit recognizable,” but it also must “create contrasts, so that its message of ridicule or pointed humor comes clear.”48 When both are successfully conveyed, “a parody is not often likely to create confusion.”49 Zuckerberg might not enjoy this mocking of his company and its ethos, but that is not what the right of publicity is designed to protect. In true misappropriation of likeness cases, the harm lies not in unflattering depictions, but in freeloading off of a celebrity’s goodwill.50 Therefore, the Rogers framework offers a more principled balance between the right to publicity and expression. In particular, the test provides a clearer measure of the defendant’s intent and impact than the transformative use test, which strains to capture both indirectly through superficial aesthetic observations.
Uniform adoption of the Rogers framework would not require doctrinal rupture either. The outcomes of many historical right of publicity cases which have been decided either under the transformative use test or prior to the invention of the modern tests would remain intact, but the rationale would be less prone to subjectivity: Liability would attach only where the plaintiff’s accumulated goodwill was appropriated, not whenever a celebrity and their persona is commented on by the public.
For instance, in White v. Samsung Electronics America, Inc., the defendant company ran a print advertisement in 1988 depicting a robot, dressed in a blonde wig and evening gown, turning letters on a futuristic game board, a clear homage to Vanna White, the hostess of “Wheel of Fortune.”51 The Ninth Circuit sided with White, holding that the right of publicity extends beyond literal likeness to the broader appropriation of identity, therefore refusing to grant First Amendment protection to Samsung as a matter of law.52 Importantly, applying the Rogers test would have yielded the same outcome. By featuring White’s distinctive persona on their advertisement, Samsung created an implied endorsement, using White’s goodwill and fame to endear consumers to the electronics being sold. Buyers would reasonably believe that White had approved the ad, which is precisely the kind of misattribution of source that undermines the value of a celebrity’s identity. Thus, even under a trademark-style analysis, Samsung’s motion to dismiss would not have been granted, supporting the doctrinal consistency a uniform adoption of the Rogers test would preserve.53
The same would also hold true in past cases where works were ultimately found to merit First Amendment protection. In Kirby v. Sega of America, Inc., the California Court of Appeals considered a claim by pop singer Kierin Kirby, known for her hit song “Groove Is in the Heart” and her catchphrase “ooh la la.”54 Kirby alleged that Sega misappropriated her identity through an animated video game character named “Ulala,” who resembled her retro-funk stage persona.55 The court, however, concluded that Sega’s use was sufficiently transformative, emphasizing the character’s distinct creative features as a space journalist with unique dance moves based on Japanese anime.56 Under the Rogers test, the outcome would not change. The average reasonable consumer would not have believed that Kirby had endorsed or collaborated on the game, and thus no misappropriation of goodwill occurred. Not only is the visual connection between Kirby and the character attenuated, but “Ulala” was one of the raw materials of a narrative-driven video game rather than as a commercial mascot explicitly promoting a product.57
Reorienting the right of publicity toward trademark principles thus provides both doctrinal coherence and practical clarity. It anchors the right in its true purpose—preserving the economic and reputational value of identity—while simultaneously better safeguarding expressive culture. As technology continues to create new tools for the replication of persona, the need for a clearer, more defensible test will only become more pressing.58 For the immediate future, it seems the judiciary will remain the primary arena for resolving conflicts between deepfake technology and the First Amendment. To ease these growing pains caused by generative AI, courts should uniformly apply the Rogers test to strike the fairest balance between expressive freedom and financial interests in the cultivation of persona.
* Samantha O’Connell is a J.D. Candidate (2026) at New York University School of Law. This Contribution is a commentary on the problem at the 2025 Cardozo FAME Center Intellectual Property Law Moot Court Competition, hosted by Cardozo School of Law. One of the questions presented involved the balance between the First Amendment and the right of publicity. This Contribution presents an argument based on the broader doctrinal issues the problem conveyed, but the views expressed herein do not necessarily represent the author’s views.
1.Compare Pavesich v. New England Life Ins. Co., 122 Ga. 190, 194 (1905) (recognizing, for the first time, a right of privacy based on defendant’s nonconsensual publishing of plaintiff’s photograph in a newspaper), and N.Y. Civ. Rights Law § 51 (McKinney 2024) (recognizing, in 1903, a statutory right of privacy after court in Roberson v. Rochester Folding Box Co., 171 N.Y. 538 (1902) refused to recognize it in common law), with Zacchini v. Scripps-Howard Broad. Co., 433 U.S. 562, 576 (1977) (“The rationale for [protecting the right of publicity] is the straightforward one of preventing unjust enrichment by the theft of good will. No social purpose is served by having the defendant get free some aspect of the plaintiff that would have market value . . . .” (emphasis added) (quoting Harry Kalven Jr., Privacy in Tort Law – Were Warren and Brandeis Wrong?, 31 Law & Contemp. Prob. 326, 331 (1966))).
2. Restatement (Third) of Unfair Competition § 46 (Am. L. Inst. 1995).
3. See Right of Publicity Statutes & Interactive Map, Right of Publicity (last visited Sept. 18, 2025), https://rightofpublicity.com/statutes.
4. Bobby Allyn, Scarlet Johansson Says She is ‘Shocked, Angered’ Over New ChatGPT Voice, NPR, May 20, 2024, https://www.npr.org/2024/05/20/1252495087/openai-pulls-ai-voice-that-was-compared-to-scarlett-johansson-in-the-movie-her.
5. See Memorandum from Quinn Emanuel Trial Lawyers, The Right of Publicity in the AI Age, at 4 (Oct. 2, 2023), https://www.quinnemanuel.com/the-firm/publications/the-right-of-publicity-in-the-ai-age/.
6. See Cardtoons, L.C. v. Major League Baseball Players Ass’n, 95 F.3d 959, 970 (10th Cir. 1996) (“[C]ommercial speech is best understood as speech that merely advertises a product or service for business purposes. As such, commercial speech may receive something less than the strict review afforded other types of speech.”) (internal citations omitted).
7. Id.
8. Sarver v. Chartier, 813 F.3d 891, 905 (9th Cir. 2016).
9. See Zacchini, 433 U.S. at 576.
10. See e.g., Hart v. Elec. Arts, Inc., 717 F.3d 141, 170 (3d Cir. 2013) (sustaining college football player’s claim against NCAA video game for including avatar with the same height, weight, number, hometown, and college team as him); Hirsch v. S.C. Johnson & Son, Inc., 90 Wis.2d 379, 402 (1979) (allowing athlete’s claim against shaving gel company for using his nickname “Crazylegs” to proceed to trial); Carson v. Here’s Johnny Portable Toilets, Inc., 698 F.2d 831, 836 (6th Cir. 1983) (sustaining television host’s claim against portable toilet company for using his famous saying, “Here’s Johnny”).
11. While not incompatible, the First Amendment versus intellectual property doctrine derived from sources like the Copyright, Trademark, and Commerce Clauses protect conflicting values that courts have to reconcile. Intellectual property law grants exclusive rights or limited monopolies that restrict others’ expression or use of information, while the First Amendment protects free expression and the free flow of ideas amongst the public. U.S. Const. amend. I; art. I, § 8, cl 8, cl 3; Eldred v. Ashcroft, 537 U.S. 186, 219 (2003) (explaining how First Amendment safeguards are built into intellectual property regimes to ensure the right of IP holders to monopolize certain speech remains limited in time and scope).
12. See Hilton v. Hallmark Cards, 599 F.3d 894, 910 (9th Cir. 2010) (holding that defendant was “only entitled to the defense as a matter of law if no trier of fact could reasonably conclude that the [work] was not transformative.” (alteration in original)).
13. See Zacchini, 433 U.S. at 576 (emphasizing “unjust enrichment by the theft of goodwill” as the ill the right of publicity tries to prevent).
14. See Comedy III Prods., Inc. v. Gary Saderup, Inc., 25 Cal. 4th 387, 404 (2001) (originating the test); Hart, 717 F.3d at 158 (endorsing the transformative use test as most useful for balancing the interests at stake in right of publicity cases); Pellegrino v. Epic Games, Inc., 451 F. Supp. 3d 373, 380 (E.D. Pa. 2020) (adopting the test within the Third Circuit).
15. Comedy III, 25 Cal. 4th at 404.
16. De Havilland v. FX Networks, LLC, 21 Cal. App. 5th 845, 850 (2018); Hilton, 599 F.3d at 910.
17. Comedy III, 25 Cal. 4th at 405–06.
18. Hilton, 599 F.3d at 910.
19. Parks v. LaFace Records, 329 F.3d 437, 448 (6th Cir. 2003); Seale v. Gramercy Pictures, 949 F. Supp. 331, 336–37 (E.D. Pa. 1996).
20. Rogers v. Grimaldi, 875 F.2d 994, 1000 (2d Cir. 1989).
21. Hart, 717 F.3d at 157.
22. Haelan Labs., Inc. v. Topps Chewing Gum, Inc., 202 F.2d 866, 868 (2d Cir. 1953) (“[A] man has a right in the publicity value of his photograph, i.e., the right to grant the exclusive privilege of publishing his picture . . .”) (emphasis added); Parks, 329 F.3d at 459 (“‘[A] celebrity has an interest that may be protected from the unauthorized commercial exploitation of that identity.’ As such, the common law right of publicity forms a species of property right.”) (quoting Carson, 698 F.2d at 835).
23. 15 U.S.C. § 1114(1).
24. See Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763, 767–68 (1992) (“The Lanham Act was intended to make ‘actionable the deceptive and misleading use of marks’ and ‘to protect persons engaged in commerce against unfair competition.’”) (quoting § 45, 15 U.S.C. § 1127).
25. U.S. Const. art. I, § 8, cl. 8; Fox Film Corp. v. Doyal, 286 U.S. 123, 127 (1932) (“The sole interest of the United States and the primary object in conferring the monopoly [of copyright] lie in the general benefits derived by the public from the labors of authors.”).
26. See, e.g., Hart v. Elec. Arts, Inc., 808 F. Supp. 2d 757, 760 (D.N.J. 2011) (holding eventually reversed by 3rd Circuit); Keller v. Elec. Arts, Inc., No. C 09-1967 CW, 2010 U.S. Dist. LEXIS 10719, at *1 (N.D. Cal. Feb. 8, 2010).
27. See Hart, 808 F. Supp. 2d at 760; Keller, 2010 U.S. Dist. LEXIS at *1.
28. Hart, 808 F. Supp. 2d at 760; Keller, 2010 U.S. Dist. LEXIS at *1.
29. Hart, 808 F. Supp. 2d at 760; Keller, 2010 U.S. Dist. LEXIS at *1.
30. Keller, 2010 U.S. Dist. LEXIS at *1.
31. See Hart, 717 F.3d at 158 (reversing and remanding the lower court decision); Keller v. Elec. Arts Inc. (In re NCAA Student-Athlete Name & Likeness Licensing Litig.), 724 F.3d 1268, 1284 (9th Cir. 2013) (affirming the lower court decision).
32. Robert C. Post & Jennifer E. Rothman, The First Amendment and the Right(s) of Publicity, 130 Yale L.J. 86, 89 (“Judicial analysis of this conflict is notoriously incoherent and inconsistent.”); Sara Asher, A Federal Right of Publicity to Navigate the Wild West of Generative AI Content, 36 St. Thomas L. Rev. 105, 119 (2024) (highlighting the blurred lines and confusion that deep fakes can create).
33. See Midler v. Ford Motor Co., 849 F.2d 460, 463 (9th Cir. 1988) (involving an impersonator covering Midler’s famous song in a Ford commercial after she explicitly turned down the endorsement opportunity); Waits v. Frito-Lay, Inc., 978 F.2d 1093, 1097–98 (9th Cir. 1992) (involving an impersonator singing a song meant to invoke Waits’s experimental jazz style and rasp in a chip commercial after the singer explicitly turned down the brand deal).
34. Winter v. DC Comics, 30 Cal. 4th 881, 890 (Cal. 2003); ETW Corp. v. Jireh Publ’g, Inc., 332 F.3d 915, 937 (6th Cir. 2003).
35. Comedy III, 25 Cal. 4th at 406.
36. See, e.g., Rogers, 875 F.2d at 1001 (analyzing film’s overall satirization of the “gaudiness and banality of contemporary television” when deciding whether its title “Ginger and Fred” was artistically relevant).
37. F. Jay Dougherty, All the World’s Not a Stooge: The “Transformativeness” Test for Analyzing a First Amendment Defense to a Right of Publicity Claim Against Distribution of a Work of Art, 27 Colum. J.L. & Arts 1, 32 (2004) (calling the transformative use test “a slender, effervescent wall to protect freedom of expression.”).
38. Stacey L. Dogan & Mark A. Lemley, What the Right of Publicity Can Learn from Trademark Law, 58 Stan. L. Rev. 1161, 1178 (2006).
39. Comedy III, 25 Cal. 4th at 406.
40. Id. at 404 (applying the “new expression, meaning, or message” factor from copyright’s fair use doctrine to a right of publicity case).
41. No Doubt v. Activision Publ’g, Inc., 192 Cal. App. 4th 1018, 1033 (2011).
42. Id. at 1023, 1033.
43. Id. at 1034.
44. Comedy III, 25 Cal. 4th at 403.
45. Mattel, Inc. v. MCA Records, 296 F.3d 894, 901 (9th Cir. 2002).
46. Mattel, Inc. v. Walking Mountain Prods., 353 F.3d 792, 796 (9th Cir. 2003).
47. Id. at 808; MCA Records, 296 F.3d at 902.
48. Jack Daniel’s Props. v. VIP Prods. LLC, 599 U.S. 140, 161 (2023) (internal citation omitted).
49. Id.
50. See Cardtoons, 95 F.3d at 974 (refusing to grant celebrities control over parodic use of their likeness as it would “only allow them to shield themselves from ridicule and criticism.”).
51. White v. Samsung Elecs. Am., Inc., 971 F.2d 1395, 1396 (9th Cir. 1992).
52. Id. at 1399–1401.
53. This trend holds for a wide variety of other right of publicity cases where the court refused to acknowledge the defendant’s First Amendment affirmative defense. See, e.g., Hilton v. Hallmark Cards, 599 F.3d 894, 912 (9th Cir. 2010) (rejecting the transformative use test for a greeting card featuring Hilton as a waitress, referencing her role in “The Simple Life,” with her famous phrase “That’s hot”); Comedy III, 25 Cal. 4th at 393 (siding against defendant artist who sold T-shirts bearing a realistic charcoal drawing of “The Three Stooges”).
54. Kirby v. Sega of Am., Inc., 144 Cal. App. 4th 47, 50–1 (2006).
55. Id. at 52.
56. Id. at 59.
57. The same outcome would apply for a variety of cases involving sufficiently transformative works. See, e.g., Cardtoons, 95 F.3d at 976 (protecting maker of parodic baseball cards that lampooned players’ unflattering traits); Hamilton v. Speight, 827 F. App’x 238, 239–40 (3d Cir. 2020) (siding with defendant game-maker for a character that resembled wrestler/football player but hunted reptoid monsters with futuristic guns).
58. Legislation has yet to provide a clear solution either. The NO FAKES Act, which proposes a uniform federal standard for the unauthorized use of someone’s likeness via AI, could have radically reshape the right of publicity landscape. NO FAKES Act of 2024, S. 4875, 118th Cong. (2024); NO FAKES Act of 2025, S. 1367, 119th Cong. (2025). However, the bill has stalled since its 2024 introduction, and Biden’s executive order calling for the regulation of AI has since been revoked. Exec. Order No. 14,179, 90 Fed. Reg. 8741 (Jan. 23, 2025).