The Authors Alliance found this write-up by Professor Jennifer Rothman at the University of Pennsylvania useful and wanted to share it with our readers. You can find Professor Rothman’s original post on her website, Rothman’s Roadmap to the Right of Publicity, here.
On July 12th, the Senate Judiciary Committee’s Subcommittee on Intellectual Property held its second hearing about artificial intelligence (AI) and intellectual property, this one was to focus expressly on “copyright” law. Although copyright was mentioned many times during the almost two-hour session and written testimony considered whether the use of unlicensed training data was copyright infringement, a surprising amount of the hearing focused not on copyright law, but instead on the right of publicity.
Both senators and witnesses spent significant time advocating for new legislation—a federal right of publicity or a federal anti-impersonation right (what one witness dubbed the FAIR Act). Discussion of such a federal law occupied more of the hearing than predicted and significantly more time than was spent parsing either existing copyright law or suggesting changes to copyright law.
In Senator Christopher Coons’s opening remarks, he suggested that a federal right of publicity should be considered to address the threat of AI to performers. At the start of his comments, Coons played an AI-generated song about AI set to the tune of New York, New York in the vocal style of Frank Sinatra. Notably, Coons highlighted that he had sought and received permission to use both the underlying copyrighted material and Frank Sinatra’s voice.
In addition to Senator Coons, Senators Marsha Blackburn and Amy Klobuchar expressly called for adding a federal right of publicity. Blackburn, a senator from Tennessee, highlighted the importance of name and likeness rights for the recording artists, songwriters, and actors in her state and pointed to the concerns raised by the viral AI-generated song “Heart on My Sleeve”. This song was created by a prompt to produce a song simulating a song created by and sung by Drake and The Weekend. Ultimately, Universal Music Group got platforms, such as Apple Music and Spotify, to take the song down on the basis of copyright infringement claims. Universal alleged that the use infringed Drake and The Weekend’s copyrighted music and sound recordings. The creation (and popularity!) of the song sent shivers through the music industry.
It therefore is no surprise that Jeffrey Harleston, General Counsel for Universal Music Group, advocated both in his oral and written testimony for a federal right of publicity to protect against “confusion, unfair competition[,] market dilution, and damage” to the reputation and career of recording artists if their voices or vocal styles are imitated in generative AI outputs. Karla Ortiz, a concept artist and illustrator, known for her work on Marvel films, also called for a federal right of publicity in her testimony. Her concerns were tied to the use of her name as a prompt to produce outputs trained on her art in her style and that could substitute for hiring her to create new works. Law Professor Matthew Sag supported adoption of a federal right of publicity to address the “hodgepodge” of state laws in the area.
Dana Rao, the Executive Vice President and General Counsel of Adobe, expressed support for a federal anti-impersonation right, which he noted had a catchy acronym—the FAIR Act. His written testimony on behalf of Adobe highlighted its support for such a law and gave the most details of what such a right might look like. Adobe suggested that such an anti-impersonation law would “offer[] artists protection against” direct economic competition of an AI-generated replication of their style and suggested that this law “would provide a right of action to an artist against those that are intentionally and commercially impersonating their work through AI tools. This type of protection would provide a new mechanism for artists to protect their livelihood from people misusing this new technology, without having to rely solely on copyright, and should include statutory damages to alleviate the burden on artists to prove actual damages, directly addressing the unfairness of an artist’s work being used to train an AI model that then generates outputs that displace the original artist.” Adobe was also open to adoption of “a federal right of publicity . . . to help address concerns about AI being used without permission to copy likenesses for commercial benefit.”
Although some of the testimony supporting a federal right of publicity suggested that many states already extend such protection, there was a consensus that a preemptive federal right could provide greater predictability, consistency, and protection. Senator Klobuchar and Universal Music’s Harleston emphasized the value of elevating the right of publicity to a federal “intellectual property” right. Notably, this would have the bonus of clarifying an open question of whether right of publicity claims are exempted from the Communications Decency Act’s § 230 immunity provision for third-party speech conveyed over internet platforms. (See, e.g. Hepp v. Facebook.)
Importantly, Klobuchar noted the overlap between concerns over commercial impersonation and concerns over deepfakes that are used to impersonate politicians and create false and misleading videos and images that pose a grave danger to democracy.
Of course, the proof is in the pudding. No specific legislation has been floated to my knowledge and so I cannot evaluate its effectiveness or pitfalls. Although the senators and witnesses who spoke about the right of publicity were generally supportive, the details of what such a law might look like were vague.
From the right-holders’ (or identity-holders’) perspective the scope of such a right is crucial. Many open questions exist. If preemptive in nature, how would such a statute affect longstanding state law precedents and the appropriation branch of the common law privacy tort that in many states is the primary vehicle for enforcing the right of publicity? When confronted with similar concerns over adopting a new “right of publicity” to replace New York’s longstanding right of privacy statute that protected against the misappropriation of a person’s name, likeness, and voice, New York legislators astutely recognized the danger of unsettling more than 100 years of precedents that had provided (mostly) predictable protection for individuals in the state.
Another key concern is whether these rights will be transferable away from the underlying identity-holders. If they are, then a federal right of publicity will have a limited and potentially negative impact on the very people who are supposedly the central concern driving the proposed law. This very concern is central to the demands of SAG-AFTRA as part of its current strike. The actors’ union wants to limit the ability of studios and producers to record a person’s performance in one context and then use AI and visual effects to create new performances in different contexts. As I have written at length elsewhere, a right of publicity law (whether federal or otherwise) that does not limit transferability will make identity-holders more vulnerable to exploitation rather than protect them. (See, e.g., Jennifer E. Rothman, The Inalienable Right of Publicity, 100 Georgetown L.J. 185 (2012); Jennifer E. Rothman, What Happened to Brooke Shields was Awful. It Could Have Been Worse, Slate, April 2023.)
Professor Matthew Sag rightly noted the importance of allowing ordinary people—not just the famous or commercially successful—to bring claims for publicity violations. This is a position with which I wholeheartedly agree, but Sag, when pressed on remedies, suggested that there should not be statutory damages. Yet, such damages are usually the best and sometimes only way for ordinary individuals to be able to recover damages and to get legal assistance to bring such claims. In fact, what is often billed as California’s statutory right of publicity for the living (Cal. Civ. Code § 3344) was originally passed under the moniker “right of privacy” and was specifically adopted to extend statutory damages to plaintiffs who did not have external commercial value making damage recovery challenging. (See Jennifer E. Rothman, The Right of Publicity: Privacy Reimagined for a Public World (Harvard Univ. Press 2018)). Notably, Dana Rao of Adobe, recognizing this concern, specifically advocated for the adoption of statutory damages.
The free speech and First Amendment concerns raised by the creation of a federal right of publicity will turn on the specific scope and likely exceptions to such a law. Depending on the particulars, it may be that potential defendants stand more to gain by a preemptive federal law than potential plaintiffs do. If there are clear and preemptive exemptions to liability this will be a win for many repeat defendants in right of publicity cases who now have to navigate a wide variety of differing state laws. And if liability is limited to instances in which there is likely confusion as to participation or sponsorship, the right of publicity will be narrowed from its current scope in most states. (See Robert C. Post and Jennifer E. Rothman, The First Amendment and the Right(s) of Publicity, 130 Yale L.J. 86 (2020)).
In short, the focus in this hearing on “AI and Copyright” on the right of publicity instead supports my earlier take that the right of publicity may pose a significant legal roadblock for developers of AI. Separate from legal liability, AI developers should take seriously the ethical concerns of producing outputs that imitate real people in ways that confuse as to their participation in vocal or audiovisual performances, or in photographs.