Navigating Actors’ Publicity Rights in the Age of Deepfakes and A.I.
In November 2020, voice actress Bev Standing received numerous messages from friends and family linking to TikTok videos narrated by Standing herself.[1] Within days, Bev Standing became the face—or more accurately, the voice—of the video-sharing app’s text-to-speech feature.[2] The problem? Standing never worked with TikTok nor had she sold them the rights to her voice.[3]
Standing has since settled with TikTok and her voice is no longer used for the speech-to-text feature,[4] but the war between actors and rapidly advancing AI-driven tools like speech-to-text, C.G.I., and deepfakes is only ramping up.[5]
The Right of Publicity
Actors, like all people, have a property-like “right of publicity,” which is really a patchwork of statutes and common law that bars the unauthorized use of a person’s name, likeness, voice or other indicia of identity without permission.[6] This right varies from state-to-state, and is usually limited to commercial uses of a person’s identity.[7] Further, the right of publicity is constrained by the First Amendment.[8] Creators of deepfakes and other A.I. generated works typically cannot be prohibited from creating disingenuous works merely because the speech is false, there must be also be some additional violation, like defamation or trademark infringement.[9]
In many states, the right of publicity is transferable and licensable, just like copyright and patents.[10] As such, individuals may sign over the ownership of the right to their own indicia by way of contract or license, as well as posthumously to their estate or beneficiaries.[11] Those beneficiaries are then free to transfer the rights of deceased celebrities to whomever they please.[12] Paul Walker was digitally resurrected with CGI to return to the Fast and Furious franchise,[13] Audrey Hepburn advertised chocolates,[14] and a hologram of Michael Jackson performed live.[15] Some film and television contracts even include simulation rights to a performer’s likeness, buried in boilerplate legalese along with traditional contractual provisions.[16] In an industry riddled with “desperate actors, confusing contracts, and multivarious laws,” how can performers protect their right to publicity?[17]
Contractual Protection?
At least one actor, Keanu Reeves, includes a clause in each of his contracts to prevent digital manipulation of his performances.[18] While seasoned actors like Reeves may rely on their bargaining power to strike favorable contract terms, industry experts fear that studios will take advantage of new actors by coercing them to sign away their publicity rights in order to land their breakout role.[19] Moreover, even the most sophisticated actors are not safe against the rise of inexpensive A.I. tools that allow anyone to create unauthorized, realistic deepfake videos of their favorite celebrity.
Last October, a deep fake made by Georgian firm Deepcake featured recently retired actor Bruce Willis tied to a bomb in an advertisement for a Russian telecoms company.[20] Numerous publications reported that Willis authorized the deepfake, claiming he was “the first Hollywood star to sell his rights to allow a ‘digital twin’ of himself to be created for use on screen,” but a representative of Willis later denied any partnership or agreement between the actor and Deepcake.[21] Unauthorized deepfakes are being used for more insidious purposes than just false endorsements—actress Emma Watson unwittingly became the “face” of a deepfake app that allows users to create deepfake pornography featuring Watson and other actors.[22] Contract law does little to prevent illicit digital manipulations, so victims of these deepfakes instead must turn to internet service providers to take down the videos and images.[23]
Internet Governance?
Social media platforms have begun cracking down on deepfakes to some degree.[24] YouTube’s misinformation policy bans videos that are doctored in a misleading way, and all deepfake videos must include a disclaimer that they are indeed fake.[25] TikTok prohibits digital forgeries “that mislead users by distorting the truth of events and cause significant harm to the subject of the video, other persons or society.”[26] Meta and Twitch ban manipulated videos that “deceive users.”[27] Meta is also working on tools to help identify deepfakes.[28] Without comprehensive federal regulation of deepfakes, experts fear that the technology will advance to the point where realistic deepfakes will escape detection by social media platforms.[29]
Rethinking the Right of Publicity
The Bruce Willis controversy has led some to wonder why our legal system allows individuals to transfer away their publicity rights to third parties.[30] The principle of transferability has its origins in other forms of intellectual property as a way for patent and copy rightsholders to voluntarily or involuntarily grant their exclusive rights to others.[31] Some argue that the ability to transfer ownership of inventions and artistic expressions should not be extended to ownership of one’s own identity.[32] Early advocates for transferability of publicity rights were, perhaps unsurprisingly, lawyers at movie studios seeking to prevent actors from working with competing studios or making endorsements and appearances without their approval.[33] SAG-AFTRA, the largest labor union for actors and performers, later advocated for transferable publicity rights so that the heirs of deceased celebrities (“delebs”) can receive income from their loved ones’ image posthumously.[34] Living or dead, individuals in many states may now transfer, license, or sell ownership of their publicity rights.[35]
While the transfer of publicity rights benefits movie studios, actors’ estates, and their descendants in a commercial sense, it often has the effect of exploiting actors’ autonomy and personhood rights.[36] Twenty-four years ago, Alice Haemmerli predicted the issue of an unauthorized computer-generated “cyberactor” based on Tom Cruise to illustrate how actors lack any meaningful publicity rights protection in state or intellectual property law.[37] Writing in 1999, before the realities of A.I. and digital manipulation, Alice Haemmerli sought to justify an autonomy-based right of publicity grounded in human freedom, as opposed to the current economy-based approach of treating publicity rights akin to copyrights and patents.[38] Haemmerli argues that the doctrine surrounding publicity-type rights has evolved in such a way that there exists a “schism between ’personal’ privacy and ‘commercial’ publicity rights.”[39] As such, she seeks a legislative reframing of the publicity right at the federal level to center around individual dignity over economic interest.[40] In 1999, many scholars saw CGI technology as constrained by the expense and massive amount of data needed, but—just as Haemmerli predicted—that is no longer a constraint given today’s inexpensive and massive capabilities for data and memory storage.[41]
Now, twenty-four years later, Haemmerli’s “cyberactor” is a reality. Her call for comprehensive federal legislation passed with morality and autonomy in mind could be exactly the protection actors need against unauthorized deepfakes and other digital manipulation. Until that happens, actors and performers like Bev Standing, Bruce Willis, and Emma Watson must rely on the confusing patchwork of intellectual property, contract, and constitutional law that comprises the right of publicity.
Footnotes
