Sedlik v. Von Drachenberg

FOR PUBLICATION UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT JEFFREY B. SEDLIK, an individual, No. 24-3367 D.C. No. Plaintiff - Appellant, 2:21-cv-01102DSF-MRW v. KATHERINE VON DRACHENBERG, an individual; OPINION AKA Kat Von D; KAT VON D, INC., a California corporation; HIGH VOLTAGE TATTOO, INC., a California corporation, Defendants - Appellees. Appeal from the United States District Court for the Central District of California Dale S. Fischer, District Judge, Presiding Argued and Submitted July 14, 2025 Pasadena, California Filed January 2, 2026 Before: Kim McLane Wardlaw, Salvador Mendoza, Jr., and Anthony D. Johnstone, Circuit Judges. 2 SEDLIK V. VON DRACHENBERG Per Curiam Opinion; Concurrence by Judge Wardlaw; Concurrence by Judge Johnstone SUMMARY * Copyright The panel affirmed the district court’s judgment after a jury trial in favor of Katherine Von Drachenberg and her tattoo parlor, High Voltage Tattoo, in an action brought under the Copyright Act by Jeffrey Sedlik, alleging infringement of his copyright in a photograph of Miles Davis. The jury found that six allegedly infringing works—a tattoo, a sketch, and four social media posts—were not substantially similar to the photograph. Von Drachenberg stipulated that four additional social media posts, referred to as the “Process Images,” which depicted her in the process of inking the tattoo, were substantially similar to the photograph because they contained a reproduction of the photograph, but the jury found that those works were not infringing because they were a fair use of the photograph. The panel held that the district court’s denial of Sedlik’s motion for summary judgment was not reviewable on appeal because it did not involve a purely legal question independent of disputed facts. * This summary constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader. SEDLIK V. VON DRACHENBERG 3 The panel held that the district court did not err in denying Sedlik’s Fed. R. Civ. P. 50(b) motion for judgment as a matter of law because he did not show that the jury’s verdict was contrary to the only reasonable conclusion permitted by the evidence for both the intrinsic and extrinsic tests for substantial similarity of the parties’ works. The extrinsic test assesses the objective similarities of the works, while the intrinsic test considers similarity of expression from the standpoint of the ordinary reasonable observer. The intrinsic test is reserved for the finder of fact, while the extrinsic test may, when appropriate, be determined as a matter of law at an earlier stage of litigation. The panel declined to disturb the jury’s findings as they pertained to the intrinsic test. The panel concluded that because the jury found that the six works were not intrinsically similar to the photograph, it need not reach the extrinsic test. Concurring in the judgment, Judge Wardlaw wrote that she fully agreed with Judge Johnstone’s concurrence and with his analysis of the court’s flawed extrinsic-intrinsic test for substantial similarity. Judge Wardlaw suggested that the intrinsic test for substantial similarity has fundamental flaws, and that the court should consider dispensing with it altogether. Concurring, Judge Johnstone, joined by Judge Wardlaw, wrote that he concurred in the per curiam opinion because it applied the current state of the court’s case law. Judge Johnstone also joined Judge Wardlaw’s call to realign the court’s copyright doctrine with the principles established by the Copyright Clause, the Copyright Act, and the Supreme Court’s copyright case law. Judge Johnstone added that the court’s doctrine has also drifted from its own origins. He wrote that the court developed the intrinsic test to protect a 4 SEDLIK V. VON DRACHENBERG work’s full expression, but over time it has lost its legal content, and now the standardless intrinsic test invites juries to reach copyright verdicts unconstrained by copyright law. The panel addressed additional issues in a concurrentlyfiled memorandum disposition. COUNSEL William F. Patry (argued), Quinn Emanuel Urquhart & Sullivan LLP, New York, New York; Moon H. Lee, Quinn Emanuel Urquhart & Sullivan LLP, Los Angeles, California; Robert E. Allen, Jason C. Linger, and Lara A. Petersen, Glaser Weil Fink Howard Jordan & Shapiro LLP, Los Angeles, California; for Plaintiff-Appellant. Allen B. Grodsky (argued), Grodsky Olecki & Puritsky LLP, Los Angeles, California, for Defendants-Appellees. Nancy E. Wolff, Cowan DeBaets Abrahams & Sheppard LLP, New York, New York, for Amicus Curiae the Copyright Alliance. Thomas B. Maddrey, American Society of Media Photographers, San Francisco, California; Stephen M. Doniger, Doniger Burroughs PC, Venice, California; Mickey H. Osterreicher and Alicia W. Calzada, National Press Photographers Association, Athens, Georgia; for Amici Curiae American Society of Media Photographers, National Press Photographers Association, American Photographic Artists, American Society for Collective Rights Licensing, Digital Media Licensing Association, Professional Photographers of America, and North American Nature Photography Association. SEDLIK V. VON DRACHENBERG 5 Matthew Hersh, Mestaz Law, Phoenix, Arizona; Sandra Aistars, Arts & Entertainment Advocacy Clinic, George Mason University, Antonin Scalia Law School, Arlington, Virginia; for Amici Curiae Tattoo Artists Ross C. Berg, Jonny Gomez, and Maxime Plesciabuchi. Andrew Grimm, Digital Justice Foundation, Omaha, Nebraska; Gregory Keenan, Digital Justice Foundation, Floral Park, New York; for Amici Curiae Lynn Goldsmith & Digital Justice Foundation. Christopher J. Sprigman, Engelberg Center for Innovation Law and Policy, New York University School of Law, New York, New York; Molly Van Houweling, Berkeley Center for Law & Technology, University of California Berkeley School of Law, Berkeley, California; for Amici Curiae Sprigman and Van Houweling. Erik Stallman, Samuelson Law Technology & Public Policy Clinic, University of California Berkeley School of Law, Berkeley, California, for Amicus Curiae Professor Pamela Samuelson. Rebecca Tushnet, Harvard Law School, Cambridge, Massachusetts, for Amici Curiae Copyright Law Professors. Christopher Bavitz, Cyberlaw Clinic, Harvard Law School, Cambridge, Massachusetts, for Amicus Curiae Authors Alliance. Corynne McSherry and Kit Walsh, Electronic Frontier Foundation, San Francisco, California, for Amicus Curiae Electronic Frontier Foundation. 6 SEDLIK V. VON DRACHENBERG OPINION PER CURIAM: Jeffrey Sedlik appeals the district court’s denial of his motions for summary judgment and for judgment as a matter of law following a jury verdict finding that Katherine Von Drachenberg and her tattoo parlor, High Voltage Tattoo (“HVT”), were not liable for copyright infringement of Sedlik’s photograph (“the Photograph”). 1 This case arose because Von Drachenberg used Sedlik’s photograph of Miles Davis as the basis of a tattoo of Davis’s likeness for her client (“the Tattoo”). Von Drachenberg also drew a sketch of the Photograph (“the Sketch”) and made several social media posts 2 related to the Tattoo and the Photograph. The jury found that six allegedly infringing works—the Tattoo, the Sketch, the Messy Progress Post, the Final Tattoo Post, the Instagram Story, and the Light Box Post—were not substantially similar to the Photograph. While Von Drachenberg stipulated that the four Process Images, which depicted Von Drachenberg in the process of inking the Tattoo, were substantially similar to the Photograph because they contained a reproduction of the Photograph, the jury found that those works were not infringing because they were a fair use of the Photograph. — 1 — We address Sedlik’s appeal of the district court’s denial of his Rule 50(b) motion as to fair use and motion for a new trial in a memorandum disposition filed concurrently with this opinion. — 2 — Following the district court’s terminology, we refer to the social media posts as the “Messy Progress Post,” the “Final Tattoo Post,” the “Instagram Story,” the “Light Box Post,” and the “Process Images.” SEDLIK V. VON DRACHENBERG 7 On appeal, Sedlik argues that the district court erred in denying his motion for summary judgment and Rule 50(b) motion for judgment as a matter of law. We affirm because the district court’s denial of Sedlik’s motion for summary judgment is not reviewable on appeal, and the district court did not err in denying Sedlik’s Rule 50(b) motion because the jury’s verdict was based on an application of the intrinsic test. I. Background Sedlik is a professional photographer and a professor who has been featured in numerous publications throughout his extensive career. Von Drachenberg is a professional tattoo artist and reality television star who has appeared on shows such as “Miami Ink” and “LA Ink.” HVT was Von Drachenberg’s tattoo shop. A. The Photograph In 1989, Sedlik created an iconic photograph depicting world-renowned jazz musician Miles Davis. The Photograph depicts a dimly lit Davis making a “Shh!” gesture with his fingers over pursed lips against a black background, his brows slightly furrowed. Sedlik specifically adjusted Davis’s fingers so that the “Shh!” symbol would represent a series of musical notes. He also adjusted Davis’s hair to achieve his desired creative expression, selected Davis’s wardrobe and jewelry, and instructed Davis on his positioning and facial expressions. Lastly, Sedlik made several decisions regarding lighting, camera positioning, and lens and shutter settings to further achieve his desired result. Sedlik has offered and sold non-exclusive copyright licenses authorizing limited reproduction, distribution, display, and creation of derivative works of the Photograph 8 SEDLIK V. VON DRACHENBERG for commercial and non-commercial purposes since its creation. Sedlik has issued one tattoo license before, though it was retroactive, and he charged tattoo artist Bryan Vanegas zero dollars. The Photograph was registered with the United States Copyright Office in 1994 and remains registered to this day. B. The Tattoo Von Drachenberg opened her own tattoo shop, HVT, as a “creative space . . . for artists,” where she also privately tattooed clients of her own. Since 2012, Von Drachenberg has not charged for her tattoos but instead treats them as gifts to the individuals that she tattoos. It is not possible to book an appointment with Von Drachenberg. In 2017, Blake Farmer—a friend of Von Drachenberg — told her that he played the trumpet and expressed how important Miles Davis was to him. Von Drachenberg offered to gift Farmer a tattoo of Davis. Farmer sent Von Drachenberg’s assistant a digital file of the photo that he wanted to use as a reference. It was the Photograph. Farmer testified that he chose the Photograph as a reference image because he “liked that it was simple,” and “[m]ost other [images of Davis] had him playing the trumpet,” which he did not believe would be suitable for an arm tattoo. Over the course of two sessions in 2017, Von Drachenberg tattooed a realistic tattoo of Davis on Farmer’s shoulder. Von Drachenberg began by creating the Sketch. She placed tracing paper over a copy of the Photograph, using it to map out the spaces she would follow while tattooing. Von Drachenberg then created a stencil using a thermal-fax machine, which she applied to Farmer’s arm. This created a map of sorts for Von Drachenberg to follow as she inked the tattoo by hand. After Farmer approved of SEDLIK V. VON DRACHENBERG 9 the placement and the size of the Tattoo, Von Drachenberg sketched what was essentially a “topographical map” using a liner tattoo machine to guide her freehand shading. Ninetynine percent of the Tattoo consisted of freehand shading. Von Drachenberg posted on social media several Process Images featuring KVD in the process of tattooing Farmer’s arm. These posts show Sedlik’s full photo in the background. The posts all featured the following image: The final Tattoo appears on the right, and the Photograph on the left: 10 SEDLIK V. VON DRACHENBERG C. The Action After discovering the Tattoo and Defendants’ social media posts, Sedlik contacted Von Drachenberg’s agent to discuss the matter, but received no response. Sedlik proceeded to file an action for copyright infringement in the Central District of California. Sedlik moved for summary judgment on his copyright infringement claims, but the district court denied the motion. The district court found that, with respect to the Tattoo, there were triable issues as to substantial similarity under both the extrinsic and intrinsic tests. The Tattoo was the only challenged work analyzed at this stage. The case proceeded to trial in January 2024. At trial, the jury had the opportunity to examine the Tattoo in person, on Farmer’s arm. The jury also heard testimony from Sedlik, Von Drachenberg, one of Von Drachenberg’s employees, and Bryan Vanegas. Both Sedlik and Defendants moved for judgment as a matter of law before the case was submitted to the jury. See SEDLIK V. VON DRACHENBERG 11 Fed. R. Civ. Proc. 50(a). After resting their defense, Defendants stipulated that the four Process Images were substantially similar because they depicted the Photograph itself. The jury, however, found that the Process Images were a fair use of the Photograph. 3 The jury found that the other works—the Tattoo, the Sketch, the Messy Progress Post, the Final Tattoo Post, the Instagram Story, and the Light Box Post—were not substantially similar. After trial, Sedlik moved for judgment as a matter of law. Sedlik requested that the court displace the jury’s verdict on substantial similarity. The district court declined to do so for several reasons. The court stated that the intrinsic test for substantial similarity is “uniquely suited for determination by the trier of fact because of its focus on the lay person, and so a court must be reluctant to reverse a jury’s finding that two works are intrinsically similar.” The district court rejected Sedlik’s argument that he had satisfied the intrinsic test because Defendants failed to present any evidence showing that the works have a different total concept and feel, noting that “the entire point of the intrinsic test is that it is from the perspective of the ordinary person without expert assistance.” It also rejected Sedlik’s arguments on the extrinsic test, stating that it “must draw the reasonable inference that the jury found that it was the unprotected elements of the [Photograph] that were copied.” — 3 — The jury mistakenly determined that eight posts that were found to be not substantially similar were a fair use. When a jury gives “superfluous answers in violation of a trial court’s express instructions contained on the special verdict form,” those answers “do not constitute legitimate or viable findings of fact” and “[t]he trial court must therefore dismiss them as surplusage, as a matter of law.” Floyd v. Laws, 929 F.2d 1390, 1397 (9th Cir. 1991). 12 SEDLIK V. VON DRACHENBERG II. Standard of Review We review de novo a district court’s denial of a motion for summary judgment, Banuelos v. Constr. Laborers’ Tr. Funds for S. Cal., 382 F.3d 897, 902 (9th Cir. 2004), and a renewed motion for judgment as a matter of law under Rule 50(b), Josephs v. Pac. Bell, 443 F.3d 1050, 1062 (9th Cir. 2006). III. Discussion A copyright plaintiff must prove (1) ownership of a valid copyright, and (2) the defendant copied protected aspects of that work’s expression. Rentmeester v. Nike, Inc., 883 F.3d 1111 , 1116–17 (9th Cir. 2018), overruled on other grounds by Skidmore as Tr. for Randy Craig Wolfe Tr. v. Led Zeppelin, 952 F.3d 1051 (9th Cir. 2020) (en banc). The second element has two separate components: “copying” and “unlawful appropriation.” Id. at 1117 ; 4 Melville B. Nimmer & David Nimmer, Nimmer on Copyright § 13.01[B] (2017). Only unlawful appropriation is at issue in this appeal. Unlawful appropriation—that is, illicit copying— requires a showing that two works are substantially similar. Skidmore, 952 F.3d at 1064 . We use a two-part test to determine whether the defendant’s work is substantially similar to the plaintiff’s copyrighted work. Id. The first part, the extrinsic test, “assesses the objective similarities of the two works, focusing only on the protectable elements of the plaintiff’s expression.” Rentmeester, 883 F.3d at 1118 . In making this assessment, the court must filter out any unprotectable elements of the plaintiff’s works. Id. The second part, the intrinsic test, “test[s] for similarity of expression from the standpoint of the ordinary reasonable observer, with no expert assistance.” Jada Toys, Inc. v. SEDLIK V. VON DRACHENBERG 13 Mattel, Inc., 518 F.3d 628, 637 (9th Cir. 2008) (citation omitted). The intrinsic test is reserved for the finder of fact, while the extrinsic test may, when appropriate, be determined as a matter of law at an earlier stage of litigation. Funky Films, Inc. v. Time Warner Ent. Co., 462 F.3d 1072 , 1076–77 (9th Cir. 2006), overruled on other grounds by Skidmore, 952 F.3d 1051 . Both tests must be satisfied for the works to be deemed substantially similar. Id. A. Motion for summary judgment Sedlik first seeks review of the district court’s denial of his motion for summary judgment, contending that the district court erred in its application of the extrinsic test for substantial similarity “due to the overwhelming and obvious similarities” between the Photograph and the challenged works. Generally, the denial of a motion for summary judgment is not reviewable on an appeal from the final judgment after a full trial on the merits, because “it would be . . . unjust to deprive a party of a jury verdict after the evidence was fully presented, on the basis of an appellate court’s review of whether the pleadings and affidavits at the time of the summary judgment motion demonstrated the need for a trial.” Locricchio v. Legal Servs. Corp., 833 F.2d 1352, 1359 (9th Cir. 1987); see also, e.g., Ortiz v. Jordan, 562 U.S. 180 , 183–84 (2011) (“May a party . . . appeal an order denying summary judgment after a full trial on the merits? Our answer is no.”); Dupree v. Younger, 598 U.S. 729 , 735–36 (2023) (“[A]n appellate court’s review of factual challenges after a trial is rooted in the complete trial record, which means that a district court’s factual rulings based on the obsolete summary-judgment record are useless”). 14 SEDLIK V. VON DRACHENBERG We recognize a narrow exception to this rule exists when summary judgment involves “a purely legal question . . . whose answer is independent of disputed facts.” Jensen v. EXC, Inc., 82 F.4th 835, 857 (9th Cir. 2023) (quoting Dupree, 598 U.S. at 737); see also Pavon v. Swift Transp. Co., 192 F.3d 902, 906 (9th Cir. 1999). This exception does not apply here because Sedlik’s arguments “hardly present ‘purely legal’ issues capable of resolution ‘with reference only to undisputed facts.’” Williams v. Gaye, 895 F.3d 1106, 1122 (9th Cir. 2018) (quoting Ortiz, 562 U.S. at 189 ). Rather, the district court’s orders denying Sedlik’s motion for summary judgment and motion for reconsideration were based on the existence of multiple triable issues of fact as to substantial similarity and fair use. Thus, the district court’s ruling does not involve the sort of “legal issue” that this court typically reviews in a narrow subset of cases. Id.; see also, e.g., Pavon, 192 F.3d at 906 (reviewing the district court’s summary judgment ruling on claim preclusion); Escriba v. Foster Poultry Farms, Inc., 743 F.3d 1236, 1243 (9th Cir. 2014) (reviewing whether “the district court erred as a matter of law by entertaining [defendant’s] ‘legally impossible’ theory of the case that [plaintiff] affirmatively declined to take FMLA leave”). We therefore decline to review the district court’s denial of summary judgment. B. Motion for judgment as a matter of law Sedlik also appeals the district court’s denial of his renewed motion for judgment as a matter of law. In reviewing the district court’s denial, we must determine “whether the evidence permits only one reasonable conclusion, and that reasonable conclusion is contrary to the jury’s verdict.” See Josephs, 443 F.3d at 1062 . Because Sedlik brought a Rule 50(a) motion on substantial similarity, SEDLIK V. VON DRACHENBERG 15 his Rule 50(b) motion was properly made. Thus, Sedlik must show that the jury’s verdict was contrary to the only reasonable conclusion permitted by the evidence for both the intrinsic and extrinsic tests, since both tests are required for substantial similarity. Funky Films, 462 F.3d at 107. Sedlik argues that no reasonable juror could find that the six works were not substantially similar to the Photograph. As for intrinsic similarity, he argues that “all of the works have a ‘mood and sentiment’ of melancholy, moodiness, and movement,” and thus the intrinsic test is met as a matter of law. We have recognized, however, that “the intrinsic test for substantial similarity is ‘uniquely suited for determination by the trier of fact’” because of its focus on the layperson. Gray v. Hudson, 28 F.4th 87, 97 (9th Cir. 2022) (quoting Sid & Marty Krofft Television Prods., Inc. v. McDonald’s Corp., 562 F.2d 1157 , 1166 (9th Cir. 1977)). For that reason, we are reluctant to reverse a jury’s application of the intrinsic test. We therefore decline to disturb the jury’s implicit findings as they pertain to the intrinsic test for substantial similarity. See Three Boys Music Corp. v. Bolton, 212 F.3d 477, 485 (9th Cir. 2000), overruled on other grounds by Skidmore, 952 F.3d 1051 . Sedlik alternatively argues that the “intrinsic test is not needed” at all because the objective similarities between the six works and the Photograph are “overwhelming.” He bases this argument on our decision in Unicolors, Inc. v. Urban Outfitters, Inc., 853 F.3d 980 (9th Cir. 2017). Unicolors held that in “exceptional cases,” a district court may grant summary judgment to a plaintiff on the basis that the “extrinsic similarity is so strong” that “no reasonable jury could find that the works are not substantially similar in their overall concept and feel.” Id. at 986–87. In such cases, the court “need not delve into a complex subjective analysis of 16 SEDLIK V. VON DRACHENBERG the works to assess substantial similarity and does not risk supplanting the jury’s subjective interpretation with its own.” Id. Here, on summary judgment, the district court concluded to the contrary—that the application of the extrinsic test was a triable issue for the jury. And following a jury verdict finding no substantial similarity based on the evidentiary record, the different procedural posture means that reversing the jury verdict would be tantamount to “supplanting the jury’s subjective interpretation with [our] own,” which we cannot do. Id. We therefore affirm the jury’s verdict based on the intrinsic test. And because the jury found that the six works were not intrinsically similar to the Photograph, we need not reach the extrinsic test for substantial similarity. See Cavalier v. Random House, Inc., 297 F.3d 815, 822 (9th Cir. 2002) (both intrinsic and extrinsic tests must be met to find that the allegedly infringing work was substantially similar to the protected work). IV. Conclusion First, the district court’s denial of Sedlik’s motion for summary judgment is not reviewable on appeal. Although a narrow exception applies when summary judgment involves a purely legal question, the district court’s order denying summary judgment was based on several triable issues of fact as to substantial similarity. Second, the district court did not err in denying Sedlik’s Rule 50(b) motion for judgment as a matter of law. Because the jury’s verdict was based on an application of the intrinsic test, we will not second-guess it. AFFIRMED. SEDLIK V. VON DRACHENBERG 17 WARDLAW, Circuit Judge, with whom JOHNSTONE, Circuit Judge, joins, concurring in the judgment: I fully agree with Judge Johnstone’s excellent concurrence and with his analysis of our flawed extrinsicintrinsic test for substantial similarity. I write separately to suggest that the “intrinsic test” for substantial similarity has fundamental flaws, and we should consider dispensing with it altogether. Consider the dispositive role the intrinsic test played here in resolving a case with complex legal questions. Photographer Jeffrey Sedlik claimed that tattoo artist Katherine Von Drachenberg infringed his copyright in his iconic photograph of the trumpet player Miles Davis. The case involved complex issues of copyright law, including “distinguishing between the protected and unprotected” components of a work depicting a human face. Photographs of people are notoriously tricky subjects of copyright analysis, given that many components of a photograph, such as a person’s facial features and pose, are not protected by copyright when “viewed in isolation,” and “[w]hat is protected by copyright is the photographer’s selection and arrangement of the photo’s otherwise unprotected elements.” Rentmeester v. Nike, Inc., 883 F.3d 1111, 1119 (9th Cir. 2018). The case presented the additional wrinkle that the allegedly infringing work was in a different medium—tattooed on the human body. The outcome, however, did not come down to these difficult questions of copyright law. Rather, it came down to the “intrinsic test,” a test created by our court and which the Supreme Court has never blessed. The intrinsic test is “subjective and asks whether the ordinary, reasonable person would find the total concept and feel of the works to be 18 SEDLIK V. VON DRACHENBERG substantially similar.” Three Boys Music Corp. v. Bolton, 212 F.3d 477, 485 (9th Cir. 2000) (citation and internal quotation marks omitted). Because the intrinsic test considers the perspective of the “lay observer,” “analytic dissection and expert testimony are not appropriate.” Sid & Marty Krofft Television Prods., Inc. v. McDonald’s Corp., 562 F.2d 1157 , 1164–66 (9th Cir. 1977); see also Funky Films, Inc. v. Time Warner Ent. Co., L.P., 462 F.3d 1072, 1077 (9th Cir. 2006) (intrinsic test is “exclusively the province of the jury”). The intrinsic test is “virtually devoid of analysis” and is “a mere subjective judgment as to whether two literary works are or are not similar.” Shaw v. Lindheim, 919 F.2d 1353, 1357 (9th Cir. 1990). In line with our precedent, the jury here was instructed that “[t]he intrinsic test is a holistic comparison that focuses on whether the works are substantially similar in the total concept and feel of the works.” Then, with no more guidance, 1 the jury was left to consider a “holistic comparison [of] . . . the total concept and feel” of the works. Id. The jury found that the works did not have a substantially similar “total concept and feel,” and thus failed the intrinsic test. Because a copyright plaintiff must meet both the extrinsic and intrinsic tests for substantial similarity, the jury’s finding on the intrinsic test was fatal to Sedlik’s claim. This test and outcome distort copyright law. The intrinsic test’s instruction to consider the “total concept and feel of the works” contradicts the Copyright Act’s statement — 1 — The jury had previously received the instruction that “[o]nly the particular expression of an idea can be copyrighted” and copyright does not protect “the underlying ideas contained in the work, such as any procedures, processes, systems, methods of operation, concepts, principles or discoveries.” SEDLIK V. VON DRACHENBERG 19 that “[i]n no case does copyright protection . . . extend to any idea [or . . .] concept.” 17 U.S.C. § 102 (b); see Pamela Samuelson, A Fresh Look at Tests for Nonliteral Copyright Infringement, 107 Nw. U.L. Rev. 1821, 1832 (2013) (“Strangely enough, no court applying the total concept test has noticed that [it] might run afoul of the strictures of § 102(b).”). As Professor Nimmer explains, the “total concept and feel” test “subvert[s] the very essence of copyright, namely the protection of original expression.” 4 Melville B. Nimmer & David Nimmer, Nimmer on Copyright § 13D.31[C] (2025). Consider the two parts of the standard, “concept” and “[f]eel”: “Concepts are statutorily ineligible for copyright protection,” and “feel hardly fares better; it is hard to imagine a more amorphous referent.” Id. (internal quotation marks omitted). And, of course, a work’s “feel” may arise from unprotected elements of a work. See Christopher Jon Sprigman & Samantha Fink Hedrick, The Filtration Problem in Copyright’s “Substantial Similarity” Infringement Test, 23 Lewis & Clark L. Rev. 571 , 580 (2019). The intrinsic test is a creation of our court; the Supreme Court has never said that the ordinary observer’s spontaneous impression of the “total concept and feel of the works,” without any expert guidance, should be a dispositive factor in copyright infringement. Indeed, Supreme Court precedent suggests the opposite: that, to accord with the Copyright Act, a court should focus on carefully filtering out concepts and ideas. While the Supreme Court has not spelled out the exact inquiry for substantial similarity, Feist Publications, Inc. v. Rural Telephone Service Co., Inc., 499 U.S. 340 (1991), is instructive on this point. Feist dealt with a copyright infringement suit brought by Rural Telephone Service Company (“Rural”) against Feist Publications 20 SEDLIK V. VON DRACHENBERG (“Feist”). Rural published a telephone directory, which it refused to license to Feist. Id. at 343 . Feist, then, extracted listings from Rural’s directory, and published its own telephone directory. Id. at 343–44. At issue was whether Feist’s use of Rural’s white pages constituted copyright infringement. Id. at 344 . In finding that Feist had not infringed Rural’s copyright, the Supreme Court explained that “[t]he mere fact that a work is copyrighted does not mean that every element of the work may be protected. Originality remains the sine qua non of copyright; accordingly, copyright protection may extend only to those components of a work that are original to the author.” Id. at 348 . Thus, the Supreme Court carefully parsed the original, protectible components of Rural’s directory from the unoriginal, unprotectable components. The Supreme Court concluded that a directory’s “choices as to selection and arrangement [of the compilations], so long as they are made independently by the compiler and entail a minimal degree of creativity,” could be “sufficiently original” to be protected. Id. at 348 . The “raw facts,” however—the names and phone numbers themselves—were unoriginal and “may be copied at will.” Id. at 350 . The Supreme Court concluded that because Rural’s white pages were “limited to basic subscriber information and arranged alphabetically,” they were “selected, coordinated, and arranged in a way that utterly lacks originality,” and “Feist’s use of the listings cannot constitute infringement.” Id. at 363–64. Crucially, Feist never suggested that the ordinary observer’s impression of the “concept and feel” of the works was essential to infringement. As Nimmer describes, Feist did not consider the “total concept and feel” of the works, but rather defined infringing conduct as the “copying of SEDLIK V. VON DRACHENBERG 21 constituent elements of the work that are original.” 4 Nimmer § 13D.15 (quoting Feist, 499 U.S. at 361 ). “Indeed, to focus on the precise scope of defendant’s admitted copying and to detail how that copying was limited to unprotected expression, it would seem that—to the exact contrary of the audience test—critical analysis is essential. The Court’s approach in Feist, on the facts there presented, is therefore inhospitable to an unadorned audience test.” 4 Nimmer § 13D.15. After all, copyright law “is intended to protect writers from the theft of the fruits of their labor, not to protect against the general public’s ‘spontaneous and immediate’ impression that the fruits have been stolen.” Id. at § 13D.16[B][1]. To be sure, a jury’s observation that two works have a similar “concept and feel” may “be important evidence” of infringement. Id. But the jury’s impression that no such similarity exists—as occurred here—does not necessarily mean there was no infringement, as there can be “theft without an immediate and spontaneous detection by the ordinary observer.” Id. This is particularly true where, as here, the works exist in different media: a photograph and a tattoo. An ordinary observer, untrained in both media, “may fail to note similarities that, if analyzed and dissected, would be only too apparent.” Id. at § 13D.16[B][2]. Or, worse, an ordinary observer may find, in practice, that works in different media almost never have the same “total concept and feel.” Thus, the intrinsic test leaves the artist’s work vulnerable to “clever thieves.” Id. at § 13D.16[B][3]. Thus, I suggest that we consider dispensing with the intrinsic test. As the Eleventh Circuit has recognized, “the ability to separate protectable expression from nonprotectable expression is, in reality, a question of law or, at the very least, a mixed question of law and fact.” Intervest 22 SEDLIK V. VON DRACHENBERG Const., Inc. v. Canterbury Estate Homes, Inc., 554 F.3d 914, 920 (11th Cir. 2008). Thus, “[i]t is difficult for a juror, even properly instructed, to conclude, after looking at two works, that there is no infringement where, say, 90% of one is a copy of the other, but only 15% of the work is protectable expression that has not been copied.” Id. Allowing the jury to render the final decision as to infringement under the guise of the work’s “total concept and feel,” immune from an appellate court’s review, impermissibly gives this complex legal question—a question of “unlawful appropriation”—to the factfinder. See Rentmeester, 883 F.3d at 1117 (emphasis added). A better solution, in my view, would be to focus the test on the filtration of protectable and unprotectable elements, and remove any inquiry into a work’s “total concept and feel.” Therefore, I agree with Judge Johnstone’s assessment that had we been able to do so in this case, we would have concluded that Von Drachenberg’s tattoo infringed Sedlik’s copyright in his photograph of Miles Davis. JOHNSTONE, Circuit Judge, with whom WARDLAW, Circuit Judge, joins, concurring: Tattoo artist Kat Von Drachenberg inked photographer Jeffrey Sedlik’s photograph of Miles Davis on a friend’s arm. There is no dispute that Sedlik held a valid copyright in the photograph or that Von Drachenberg copied from it to create the tattoo. She traced Sedlik’s photograph on a light box to create a stencil, which she transferred onto her friend’s arm. Then, with Sedlik’s photograph hanging at her side as a reference, she tattooed its image onto her friend’s skin. Von Drachenberg posted photos of the process and SEDLIK V. VON DRACHENBERG 23 resulting tattoo on social media to market her artistry and tattoo shop. The tattoo resembles the protected aspects of the photograph. “To the[] trained eyes” of amici tattoo artists, “there is no difference between the two works that is not explained by the medium in which [Sedlik’s photograph] was replicated into [Von Drachenberg’s tattoo].” It shares all its major creative elements: Davis’s distinctive finger pose (arranged by Sedlik); his facial expression (encouraged by Sedlik); his dark wavy hair surrounding his face (styled at Sedlik’s direction); the shadows and highlights on his face and hands (lit by Sedlik); and the specific angle, position, and focus of the viewer’s perspective (chosen by Sedlik). Von Drachenberg admitted at trial that these features were “very similar” to the photograph. Indeed, Von Drachenberg’s friend had asked that the tattoo match Sedlik’s photograph, Von Drachenberg followed a process designed to do so, and her shop concluded that she succeeded—it touted the tattoo on social media as “100% exactly the same as the reference.” But when Sedlik sued Von Drachenberg for copyright infringement, a jury found that the tattoo was not “substantially similar” to the photograph and returned a verdict for Von Drachenberg. At trial, Von Drachenberg testified that she made minor changes from Sedlik’s photograph because of the inherent challenges in replicating a photograph on human skin. She “added shading, highlights, and texture to several parts of Miles Davis’s face”; altered waves and highlights in his hair to “creat[e] a kind of ‘visual noise’ that flies all around the tattoo and gives the sense of movement,” as well as leaving room for additional tattoos; and left out certain details of the photograph, like Sedlik’s reflection in Davis’s eyes, part of 24 SEDLIK V. VON DRACHENBERG Davis’s wrist and his clothing, and the photograph’s black background. In determining substantial similarity, however, these differences matter little. See Newton v. Diamond, 388 F.3d 1189, 1195 (9th Cir. 2004) (“[N]o plagiarist can excuse the wrong by showing how much of his work he did not pirate.” (quoting Sheldon v. Metro–Goldwyn Pictures Corp., 81 F.2d 49, 56 (2d Cir. 1936))). Instead, what our cases required the jury to focus on— and what mattered most to the verdict—was the jury’s subjective perception of “the total concept and feel” of the works. See Doc. 219, Jury Instructions, Page ID 4180 (describing the intrinsic test as “a holistic comparison that focuses on whether the works are substantially similar in the total concept and feel of the works”). As Von Drachenberg correctly argues, our “intrinsic test” for substantial similarity focuses solely on “an ordinary person’s holistic and subjective impressions of the similarities in the works” based on “total concept and feel.” And here the jury felt that Von Drachenberg’s tattoo was not “substantially similar” to Sedlik’s photograph. But a jury’s job is to find facts, not feelings. Because a jury’s subjective impression of a work’s “total concept and feel” is all but unreviewable, we have never reversed a jury’s no-infringement verdict under the intrinsic test. The per curiam opinion applies the current state of our case law in affirming the district court, so I concur. I also join Judge Wardlaw’s call to realign our copyright doctrine with the principles established by the Copyright Clause, the Copyright Act, and the Supreme Court’s copyright case law. I write to add that our doctrine has also drifted from its own origins. A half-century ago, we developed the intrinsic test to protect a work’s full expression. But over time it has lost its legal content. Now the standardless intrinsic test invites SEDLIK V. VON DRACHENBERG 25 juries to reach copyright verdicts unconstrained by copyright law. In practice, its holistic protection has become an empty promise. I. We developed the intrinsic test to protect the whole of an expression. Federal copyright law protects “pictorial . . . works,” but “[i]n no case does copyright protection . . . extend to any . . . concept.” 17 U.S.C. § 102 (a)(5), (b). So how did we get from a statute that protects against the copying of pictures, not concepts, to a doctrine that turns on similarity in “total concept and feel”? The roots of the intrinsic test lie in the basic rule that copyright protects only an original expression of an idea—not the idea itself, nor the use of unoriginal scènes à faire that are necessary to express it. See Baker v. Selden, 101 U.S. 99 , 101–03 (1879); 12 Melville B. Nimmer & David Nimmer, Nimmer on Copyright §§ 313.3(A), 313.4(I) (2025). As Judge Learned Hand lamented almost a century ago, distinguishing protected expression from unprotected ideas and stock elements is no easy task. Nichols v. Universal Pictures Corp., 45 F.2d 119, 121 (2d. Cir 1930) (“Nobody has ever been able to fix that boundary, and nobody ever can.”). It was in undertaking that challenge in Roth Greeting Cards v. United Card Co., 429 F.2d 1106 (9th Cir. 1970), that we first looked to “total concept and feel.” Roth concerned a defendant’s allegedly unlawful copying of the plaintiff’s greeting cards. Id. at 1107–08. Like the works at issue, the defendant’s cards “b[ore] a remarkable resemblance” to plaintiff’s copyrighted work, with many similarities: “the characters depicted in the art work,” the specific phrases and lettering, and “the arrangement of the words on the greeting card[s].” Id. at 26 SEDLIK V. VON DRACHENBERG 1107, 1110. Yet the district court concluded that the defendant did not copy any protected expression because most of these similarities—like the common phrases used in the cards—consisted of unprotected ideas or stock elements. Id. at 1109 . On appeal, we agreed that many of the cards’ shared elements were unprotected in isolation. Id. But that did not end the inquiry: “proper analysis of the problem require[d] that all elements of each card, including text, arrangement of text, art work, and association between art work and text, be considered as a whole.” Id. Taken together, these elements created an expression that was sufficiently “original” to merit copyright protection. Id. at 1109–10. And we held that, despite some differences in specific elements, the defendant’s cards were substantially similar to the plaintiff’s protected combination because “in total concept and feel the cards . . . [we]re the same.” Id. at 1110 . Roth used “total concept and feel” to refer to what our more recent cases describe as the “protectable . . . selection and arrangement of . . . otherwise unprotected elements.” Tangle, Inc. v. Aritzia, Inc., 125 F.4th 991 , 997 (9th Cir. 2025) (internal quotations omitted) (quoting Hanagami v. Epic Games, Inc., 85 F.4th 931, 943 (9th Cir. 2023)). We incorporated Roth’s meaning of “total concept and feel” into our first articulation of the intrinsic test in Sid & Marty Krofft Television Prods., Inc. v. McDonald’s Corp., 562 F.2d 1157 (9th Cir. 1977) (“Krofft”). There, we addressed a claim that “Mayor McCheese,” a character in a McDonald’s advertising campaign, infringed the plaintiffs’ copyright in H.R. Pufnstuf, a character in a children’s television show. Id. at 1161–62. “The real task in a copyright infringement action,” we reasoned, “is to determine whether there has been copying of the expression of an idea rather SEDLIK V. VON DRACHENBERG 27 than just the idea itself.” Id. at 1163. Because there “must be substantial similarity not only of the general ideas but of the expressions of those ideas,” we set forth our now-familiar two-part test for substantial similarity. Id. at 1164. In part one of the analysis, the “extrinsic test,” we ask “whether there is substantial similarity in ideas” underlying the two works. Id. at 1164 (emphasis added). We do so by examining “specific criteria which can be listed and analyzed,” some of which might be individually unprotectable, like “the type of artwork involved, the materials used, the subject matter, and the setting for the subject.” Id. While still a factual inquiry, this “analytic dissection” could “often be decided as a matter of law” by breaking down the works into their constituent elements and comparing those elements for proof of copying. Id. If this extrinsic comparison shows a substantial similarity between these elements, then the trier of fact assesses the intrinsic similarity. In part two, the “intrinsic test,” we ask “whether there is substantial similarity in expressions” of the shared idea based on the “response of the ordinary reasonable person.” Id. (emphasis added). This inquiry turns on the overall “impact of the respective works upon the minds” of lay observers, making it “uniquely suited for determination by the trier of fact.” Id. at 1166. Thus, in Krofft, we held that the existence of some dissimilarities in “the constituent parts” of the two characters at issue—like their “clothing, colors, features, and mannerisms”—did not preclude a finding of substantial similarity under the intrinsic test. Id. at 1165–67. Instead, as in Roth, what mattered was that Mayor McCheese “captured the ‘total concept and feel’ of the Pufnstuf show” when the arrangement of all the elements was considered as a whole. Id. at 1167 (quoting Roth, 429 F.2d at 1110 ). Under 28 SEDLIK V. VON DRACHENBERG this broader, holistic view of substantial similarity, we affirmed the jury’s infringement verdict. Id. In the decade after Krofft, we continued applying the extrinsic test to assess the similarity of ideas based on the works’ individual elements, then used the intrinsic test to discern the similarity of the works’ expression of those shared ideas based on the overall effect—that is, “total concept and feel”—of the combination of those elements. See, e.g., See v. Durang, 711 F.2d 141 , 143–44 (9th Cir. 1983); Litchfield v. Spielberg, 736 F.2d 1352, 1356 (9th Cir. 1984); Berkic v. Crichton, 761 F.2d 1289, 1292 (9th Cir. 1985); McCulloch v. Albert E. Price, Inc., 823 F.2d 316, 319 (9th Cir. 1987); Narell v. Freeman, 872 F.2d 907 , 912–13 (9th Cir. 1989). Over time, however, we gradually expanded the types of similarities considered in the extrinsic analysis. Although Krofft’s extrinsic test had looked only to discrete elements like “the type of artwork” or “subject matter” of two works, 562 F.2d at 1164, later cases also compared broader similarities of “theme” and “mood,” see, e.g., Litchfield, 736 F.2d at 1356–57; Berkic, 761 F.2d at 1292 ; Olson v. Nat’l Broad. Co., 855 F.2d 1446 , 1450–51 (9th Cir. 1988), which Krofft had examined as part of the intrinsic test, 562 F.2d at 1166. We recognized the increasingly “lengthy list of concrete elements” considered “under the extrinsic test” in Shaw v. Lindheim, 919 F.2d 1353, 1356 (9th Cir. 1990). And we reasoned that the shift required a rethinking of the extrinsic test: “Now that it include[d] virtually every element that may be considered concrete,” the test could “no longer be seen as a test for mere similarity of ideas.” Id. at 1357 . Shaw thus recast the extrinsic test as an “objective” analysis of the similarity of works’ expression, not just of the unprotected ideas underlying that expression. Id. But if the extrinsic test SEDLIK V. VON DRACHENBERG 29 examined “all objective manifestations of creativity,” including the cumulative impact of works’ combination of discrete elements, where did that leave the intrinsic test? In Shaw’s words, it “ha[d] become a mere subjective judgment as to whether two . . . works are or are not similar,” which was “now virtually devoid of analysis.” Id. Put another way, the extrinsic test “ha[d] expanded at the expense of the now almost atavistic” intrinsic test. Douglas Y’Barbo, The Origin of the Contemporary Standard for Copyright Infringement, 6 J. Intell. Prop. L. 285 , 311 (1999). Shaw’s reframing of the extrinsic and intrinsic test “as objective and subjective analyses of expression” has remained the law of our circuit for thirty-five years. 919 F.2d at 1357 . While once we limited our extrinsic analysis to discrete elements, now we consider all objective similarities, including those arising from works’ overall “selection and arrangement” of elements at the first step. Woodland v. Hill, 136 F.4th 1199 , 1210–11 (9th Cir. 2025) (quoting Rentmeester v. Nike, Inc., 883 F.3d 1111, 1119 (9th Cir. 2018)). And where our old intrinsic test involved at least some analysis of how works combine different elements to create an overall “concept and feel,” the modern test involves a purely “subjective comparison of the works” in the often decisive second step. Rentmeester, 883 F.3d at 1118 . What began as a factual test for determining substantial similarity of an idea’s overall expression has become a subjective test based on the jury’s own impression. II. The intrinsic test leads to unpredictable and asymmetrical results. As it now operates, the intrinsic test has distorted copyright law in our circuit. Because we have stripped away any objective analysis from the intrinsic test, courts cannot 30 SEDLIK V. VON DRACHENBERG decide it as a matter of law, precluding summary judgment for almost all copyright plaintiffs. The lack of objective considerations also leaves juries without guidance at trial, leading to unpredictable verdicts based on instructions to decide cases based on “total concept and feel.” And while defendants may appeal infringement verdicts under the extrinsic test, plaintiffs have no meaningful way to seek review of unfavorable judgments. The result has been to increase unpredictability of outcomes in copyright cases and to stack the procedural deck against plaintiffs on questions of substantial similarity. To start, the intrinsic test makes it almost impossible for copyright plaintiffs to win their cases at summary judgment. Because courts may decide the extrinsic test as a matter of law, defendants often dispose of cases at summary judgment by showing a lack of extrinsic similarity. See, e.g., Folkens v. Wyland Worldwide, LLC, 882 F.3d 768, 776 (9th Cir. 2018); Brown Bag Software v. Symantec Corp., 960 F.2d 1465 , 1475–77 (9th Cir. 1992). Plaintiffs, on the other hand, must satisfy both the extrinsic and intrinsic tests to succeed in their claims. Rentmeester, 883 F.3d at 1118 (citing Funky Films, Inc. v. Time Warner Ent. Co., 462 F3d 1072 , 1077 (9th Cir. 2006). And we have held that “it is improper . . . to base a grant of summary judgment on [the] purely subjective determination of similarity” that now guides the intrinsic test. Shaw, 919 F.2d at 1359 ; see also L.A. Printex Indus., Inc. v. Aeropostale, Inc., 676 F.3d 841, 852 (9th Cir. 2012). We have strayed from that rule in only one published opinion, affirming summary judgment under the intrinsic test in the rare instance where two works were “overwhelmingly identical” and in the same medium. Unicolors, Inc. v. Urban Outfitters, Inc., 853 F.3d 980, 987 (9th Cir. 2017) (quoting Twentieth Century–Fox Film Corp. SEDLIK V. VON DRACHENBERG 31 v. MCA, Inc., 715 F.2d 1327, 1330 (9th Cir. 1983)). Functionally precluding summary judgment for copyright plaintiffs favors defendants. See Clark D. Asay, An Empirical Study of Copyright’s Substantial Similarity Test, 13 U.C. Irvine L. Rev. 35, 88 (2022). It also “substantially dilute[s]” the effect of Rule 56. Unicolors, 853 F.3d at 987 . And as amicus the Copyright Alliance explains, requiring that all plaintiffs proceed to trial in clear-cut cases “runs the high risk of denying [them] the ability to enforce their copyrights by making it prohibitively expensive to do so.” Even when plaintiffs with strong cases reach trial, their chances of prevailing are still uncertain. The intrinsic test “ostensibly assesses qualities of the works themselves, ones that exist independently of the observer.” Bruce E. Boyden, The Grapes of Roth, 99 Wash. L. Rev. 1093 , 1127 (2024). But substantial similarity in “total concept and feel” depends on an observer’s impressions, which may vary dramatically from person to person. Id. Because the test is “devoid of analysis,” Shaw, 919 F.2d at 1357 , “[t]he jury is left to use its subjective judgment, often . . . without instructions designed to steer that judgment.” Christopher Jon Sprigman & Samantha Fink Hedrick, The Filtration Problem in Copyright’s “Substantial Similarity” Infringement Test, 23 Lewis & Clark L. Rev. 571 , 580 (2019). So “when substantial similarity cases go to the jury,” as they must for most plaintiffs to prevail, “a panel of non-experts is asked to apply an empty standard.” Boyden, supra, at 1098. The intrinsic test’s vagueness is made worse because the parties cannot use expert testimony to focus the inquiry. Substantial similarity, after all, “is to be judged by the ordinary observer.” Mark A. Lemley, Our Bizarre System for Proving Copyright Infringement, 57 J. Copyright Soc’y U.S.A. 719, 738–39 (2010); see Krofft, 562 F.2d at 1164. So 32 SEDLIK V. VON DRACHENBERG while plaintiffs must proceed to trial to succeed on their claims, once there, they cannot use experts to assist the jury in “dissect[ing] the protectable from the unprotectable” and “guide the drawing of inferences from different sorts of similarities.” Lemley, supra, at 736. The subjective nature of the intrinsic test and the parties’ inability to present expert testimony make the outcomes of cases unpredictable and unprincipled—different juries or judges might reach opposite conclusions on the same record. See id.; Pamela Samuelson, A Fresh Look at Tests for Nonliteral Copyright Infringement, 107 Nw. U. L. Rev. 1821 , 1832 (2013). Little wonder, then, that copyright scholars have criticized our intrinsic test for producing inconsistent and unpredictable results. See, e.g., Boyden, supra, at 1096; Samuelson, supra, at 1832; Amy B. Cohen, Masking Copyright Decisionmaking: The Meaninglessness of Substantial Similarity, 20 U.C. Davis L. Rev. 719 , 757 (1987). Ultimately, our vague intrinsic test favors defendants. To be sure, the intrinsic test may provide the occasional windfall to plaintiffs by freeing the jury to assess substantial similarity in “total concept and feel,” no matter if the similarity stems from elements that copyright law protects. See Boyden, supra, at 1098; Samuelson, supra, at 1832; Lemley, supra, at 739. But without the opportunity to dissect the protectable and unprotectable elements of the work, or to hear expert testimony doing so, “juries are more likely to find infringement in dubious circumstances.” Lemley, supra, at 739. Or, as apparently happened here, the intrinsic test can divert the jury’s attention away from objectively substantial copying of protected expression toward trivial differences, resulting in an inscrutable no-infringement verdict. See Nicole Doc, Do We Even Need a Test?: A Reevaluation of Assessing Substantial Similarity in a SEDLIK V. VON DRACHENBERG 33 Copyright Infringement Case, 15 J. L. & Pol’y 1375, 1400– 01 (2009) (arguing that, in many cases, lay observers are “simply not capable of detecting very real appropriation” where an infringing work “consists merely of clever juxtaposition and alteration of unessential details in the plaintiff’s work” (quoting Nimmer on Copyright § 13.03[E][2])); Y’Barbo, supra, at 293 n.29. While the intrinsic test may produce erroneous verdicts for both plaintiffs and defendants, only defendants can correct such verdicts. When a jury errs by finding intrinsic similarity based on works’ unprotected elements, defendants can challenge the resulting infringement verdict under the extrinsic test, which allows us to filter those unprotected elements out of the analysis. We have not hesitated to reverse infringement verdicts in such cases. See, e.g., Gray v. Hudson, 28 F.4th 87 , 96–97, 103 (9th Cir. 2022); Corbello v. Valli, 974 F.3d 965 , 973–75 (9th Cir. 2020). But when a jury errs by finding a lack of intrinsic similarity despite substantial copying of protected elements, plaintiffs have no recourse. Because the intrinsic test is a purely “subjective assessment of the ‘concept and feel’ of two works,” we have held that it is “a task no more suitable for a judge than a jury.” Shaw, 919 F.2d at 1360 . As a result, we cannot “second-guess the jury’s application of the intrinsic test.” Three Boys Music Corp. v. Bolton, 212 F.3d 477, 485 (9th Cir. 2000). Indeed, we have never reversed a jury’s noinfringement verdict under the test: “the absence of legal analysis . . . frustrate[s] appellate review.” Shaw, 919 F.2d at 1357 ; see also Lemley, supra, at 739–40. And after a full trial on the merits, as in this case, we cannot review a denial of a plaintiff’s motion for summary judgment even when there was no genuine dispute about intrinsic similarity. See 34 SEDLIK V. VON DRACHENBERG Locricchio v. Legal Services Corp., 833 F.2d 1352, 1359 (9th Cir. 1987). The cumulative effect of all this has been to undermine copyright protections and their purpose of “promot[ing] the Progress of Science and useful Arts.” U.S. Const. art. I, § 8, cl. 8. Few argue that this is the result that copyright law contemplates. “[T]he almost universal opinion among copyright scholars [is] that ‘total concept and feel’ is a horrible test for infringement.” Boyden, supra, at 1095; see also id. at 1095–96 (collecting criticisms of the test as “threaten[ing] to subvert the very essence of copyright,” “almost perversely calculated to miss the point,” “internally inconsistent,” “vacuous,” providing “unconstrained space for the impressions of the judge or jury,” and “lead[ing] to unpredictable, unreviewable similarity determinations”). III. Conclusion By all accounts, the tattoo at issue here is a tour de force. As amici tattoo artists note, photorealistic images are nearly impossible to replicate on human skin. But Von Drachenberg did it: skillfully tracing Sedlik’s photograph of Miles Davis, transferring the resulting stencil to her friend’s arm, and carefully reproducing every significant detail that Sedlik had crafted in his original photograph. It is an understatement to say that the tattoo is substantially similar to Sedlik’s photograph —it is, as Von Drachenberg’s tattoo shop put it, “100% exactly the same.” Our caselaw required the jury to apply a subjective “total concept and feel” test without meaningful guidance. That led the jury to deny the substantial similarity between the photograph and the tattoo. And no matter how many protectable elements of the photograph show up in the tattoo, we cannot meaningfully SEDLIK V. VON DRACHENBERG 35 review the jury’s finding of no similarity. In my view, a test that produces such a result cannot be right. We should defer to juries’ assessments of substantial similarity. But for juries to make those calls consistently, they need more than an empty standard premised on “total concept and feel.” And despite the best efforts of the district court, juries sometimes misapply the law. When they do, courts must be able to correct course to ensure that copyright serves its purpose of encouraging and protecting creative expression. See Andy Warhol Found. for the Visual Arts, Inc. v. Goldsmith, 598 U.S. 508 , 526 (2023); U.S. Const. art. I, § 8, cl. 8. Because our intrinsic test leaves us no choice but to rubber-stamp the no-infringement verdict, I concur in the majority opinion. But I agree with Judge Wardlaw that it is time to discard the intrinsic test and replace it with a legal rule that offers more consistent and even-handed protection to copyrighted works.