Right of Publicity

Hart v. Electronic Arts: The Transformative-Use Test and an Athlete's Likeness in a Video Game

The Third Circuit held that EA's photorealistic use of a college quarterback's avatar in NCAA Football was not transformative enough to defeat his right of publicity — adopting the transformative-use test as the circuit's framework for likeness-in-media disputes.

January 27, 2026
A football player's silhouette rendered as a digital avatar on a screen
The court found that the game's value depended on the realism of the athletes it depicted, not on any creative transformation of them. Shutterstock
Educational content, not legal advice. This article explains general legal concepts. It does not create an attorney–client relationship. For your specific situation, consult a licensed attorney.

Hart v. Electronic Arts, Inc., No. 11-3750, 717 F.3d 141 (3d Cir. May 21, 2013), is the decision that imported California’s transformative-use test into the federal court of appeals’ treatment of the right of publicity and applied it to interactive entertainment. Ryan Hart, a quarterback who wore number 13 for Rutgers University from 2002 through 2005, sued Electronic Arts under New Jersey law for using his likeness and biographical attributes in the NCAA Football franchise (the 2004, 2005, and 2006 editions). The District of New Jersey granted EA summary judgment on First Amendment grounds. A divided Third Circuit panel — Judge Greenaway writing for the majority, joined by Judge Tashima (sitting by designation), with Judge Ambro dissenting — reversed and remanded, holding that EA’s use of Hart’s unaltered identity was not sufficiently transformative to claim constitutional protection. The opinion remains the leading circuit-level articulation of how name, image, and likeness rights collide with expressive works.

At a glance

  • Case: Hart v. Electronic Arts, Inc., No. 11-3750, 717 F.3d 141 (3d Cir. 2013)
  • Court / panel: U.S. Court of Appeals for the Third Circuit; Greenaway, J. (majority), Tashima, J. (by designation), Ambro, J. (dissenting)
  • Decided: May 21, 2013
  • Governing law: New Jersey common-law right of publicity
  • Holding: The First Amendment does not bar Hart’s right-of-publicity claim because EA’s NCAA Football did not sufficiently transform his identity; summary judgment for EA reversed and remanded
  • Test adopted: The transformative-use test, drawn from Comedy III Productions v. Gary Saderup, Inc., 25 Cal. 4th 387 (2001)
  • Significance: First federal appellate adoption of the transformative-use test for video games; the Ninth Circuit reached the same result two months later in In re NCAA Student-Athlete Name & Likeness Licensing Litigation (Keller v. EA), 724 F.3d 1268 (9th Cir. 2013)

The factual record was unusually clean, which is part of why the case became a teaching vehicle. EA built its avatars to match real players: Hart’s avatar shared his jersey number, height, weight, home state, and on-field statistics, and was placed on a virtual Rutgers squad playing the sport through which Hart had earned public recognition. EA did not license those attributes and did not pay the players, relying instead on the proposition that the games were protected expression. The dispute therefore did not turn on whether EA had appropriated Hart’s identity — it plainly had — but on whether the First Amendment immunized that appropriation.

Name, image, and likeness under New Jersey law

The right of publicity, as the panel framed it, protects a person’s proprietary interest in the commercial value of his or her identity. New Jersey recognizes the tort as a species of the broader privacy framework, reaching the unauthorized commercial exploitation of a person’s name, likeness, or other indicia of identity. The interest is economic as much as dignitary: it secures to the individual the fruits of the recognition he has cultivated, and it prevents others from free-riding on that recognition without consent.

For an athlete, the “likeness” at stake is not merely a facial photograph. It is the cluster of recognizable attributes — number, position, physical build, team affiliation, and statistical record — that together identify the person to fans. The Hart court took this composite view seriously. Because EA assembled precisely those attributes and tied them to the activity for which Hart was known, the appropriation reached the core of his publicity interest rather than some incidental detail. That observation matters for the modern NIL economy: the value being captured is the identity-in-context, and a defendant cannot defeat a claim simply by altering one variable while preserving the recognizable whole.

The court was careful to distinguish the publicity tort from defamation or false endorsement. Hart did not need to prove that consumers believed he had sponsored the game, nor that anything false had been said about him. The wrong was the unconsented commercial use of his identity itself. That framing keeps the right of publicity analytically distinct from the Lanham Act’s confusion-based regime — a distinction that recurs whenever a likeness appears in a product that is also, in some sense, speech.

The transformative-use test for likeness in video games

The doctrinal heart of Hart is the panel’s choice of test. The court began by acknowledging that video games are fully protected expression — a point the Supreme Court had settled in Brown v. Entertainment Merchants Ass’n, 564 U.S. 786 (2011). Protected speech, however, does not categorically defeat a right-of-publicity claim; some accommodation between the two interests is required. The panel surveyed the competing frameworks: the Rogers v. Grimaldi artistic-relevance test borrowed from trademark law, the “predominant use” test that asks whether the work’s commercial purpose outweighs its expressive one, and the transformative-use test derived from copyright’s fair-use inquiry.

The court rejected Rogers as ill-suited to publicity claims, reasoning that a test designed to gauge consumer confusion about source does not measure the distinct harm of misappropriated identity. It rejected the predominant-use test as standardless and unduly subjective, requiring judges to act as art critics weighing commercial against expressive value. It settled instead on the transformative-use test from Comedy III: the dispositive question is whether the celebrity’s likeness is “one of the raw materials from which an original work is synthesized,” such that the product has become primarily the defendant’s own expression, or whether the depiction is “the very sum and substance of the work in question.”

Applied to NCAA Football, the test produced a defeat for EA. The avatar did Hart’s real-world activity — playing college quarterback — in a digitally recreated version of his real-world setting. The realism was the point: EA’s commercial proposition was that players could control authentic-seeming versions of actual athletes and teams. The majority held that this literal, context-preserving depiction was not transformed into something primarily EA’s own expression. Notably, the court rejected EA’s argument that the game’s user-customization features supplied the necessary transformation. That players could alter an avatar’s appearance did not matter, because Hart’s “unaltered likeness is central to the core of the game experience”; the default presentation, not a hypothetical edited one, defined the use. The court also declined to credit creative elements located elsewhere in the game — the graphics engine, the virtual stadiums, the interactive mechanics — reasoning that transformation must inhere in the depiction of the plaintiff, not in the surrounding software.

The right of publicity versus the First Amendment

Hart is best understood as an exercise in line-drawing between two values the Constitution and the common law both protect. The majority did not hold that athletes may suppress all unlicensed depictions; it held that a depiction designed to be as faithful to reality as possible, and marketed on the strength of that fidelity, sits at the far end of the spectrum from protected creative reimagining. A biography, a documentary, a satirical cartoon, or a stylized parody would add the author’s own expressive contribution; a near-exact reproduction placed in its natural setting adds little but appropriated value.

Judge Ambro’s dissent accepted the transformative-use test as the correct framework but parted ways on its application. He would have found the game as a whole sufficiently transformative, emphasizing the interactive and creative architecture surrounding the avatars and warning that the majority’s focus on the isolated likeness risked swallowing expressive works that necessarily incorporate real people. The disagreement previews a durable tension in the doctrine: whether transformation is assessed at the level of the individual depiction or the work in its entirety. That question — narrow lens versus wide lens — frequently decides cases under the transformative-use test, and Hart did not resolve it so much as choose a side.

The decision also gains weight from its near-simultaneous twin. Two months later, in the Keller appeal, the Ninth Circuit applied the same transformative-use analysis to the same EA franchise and reached the same conclusion. The convergence of two circuits on identical facts gave the holding outsized practical force and contributed to EA’s decision to settle the consolidated publicity litigation and to discontinue the NCAA Football series for years. The cases did not bind every jurisdiction, but together they signaled that photorealistic, unlicensed use of identifiable athletes was legally untenable.

Open questions

  • Is transformation measured by the discrete depiction of the plaintiff or by the expressive work as a whole? Hart chose the former; the dissent and other courts pull toward the latter, and the answer often dictates the outcome.
  • How does the transformative-use test interact with the Supreme Court’s only right-of-publicity decision, Zacchini v. Scripps-Howard Broadcasting Co., 433 U.S. 562 (1977), which permitted a publicity claim even for newsworthy content?
  • After Jack Daniel’s Properties v. VIP Products, 599 U.S. 140 (2023), narrowed Rogers in the trademark setting, do courts have any remaining reason to prefer Rogers over the transformative-use test for likeness claims?
  • Does generative AI — which can synthesize a “new” but recognizable likeness — fit the transformative-use rubric at all, or does it demand a different analytic vocabulary?

Implications

  • Realism is a liability, not a shield. The more faithfully a product reproduces an identifiable person in his natural context, the weaker its transformative-use defense. Fidelity that drives commercial value cuts against the defendant.
  • Customization features do not launder appropriation. A defendant cannot rely on a user’s hypothetical ability to alter a likeness; courts evaluate the default, as-shipped depiction.
  • The transformative-use test is now the circuit default for likeness-in-media. Practitioners in the Third and Ninth Circuits should brief publicity disputes around Comedy III transformation rather than Rogers artistic relevance.
  • NIL value is composite. Number, position, physique, team, and statistics together constitute the protected identity; altering one element while preserving the recognizable whole will not defeat a claim.
  • The holding helped reshape an industry. The convergence of Hart and Keller drove settlement and the suspension of the NCAA Football franchise, illustrating how publicity doctrine can reorder licensing markets.

Frequently asked questions

Did Hart have to prove that consumers thought he endorsed the game? No. The right of publicity protects against unconsented commercial use of identity itself, independent of any consumer confusion about sponsorship or endorsement. That is what distinguishes it from a false-endorsement claim under the Lanham Act, and it is why Hart could prevail without showing that anyone believed he had authorized NCAA Football.

What exactly is the transformative-use test? Borrowed from Comedy III Productions v. Gary Saderup and the fair-use tradition, it asks whether the celebrity’s likeness is merely raw material that the defendant has synthesized into a new, primarily original expression, or whether the likeness is the very substance of the work. If the depiction is essentially a literal, unaltered reproduction trading on the person’s recognizability, it fails the test and the First Amendment provides no defense.

Is Hart binding nationwide? No. It is binding precedent only in the Third Circuit. Its influence, however, is amplified by the Ninth Circuit’s contemporaneous Keller decision reaching the same result on materially identical facts. Together they are the most frequently cited authorities on likeness rights in video games, but other circuits and state courts remain free to adopt different frameworks.

Authorities and sources