AI-Generated Voicing And Posthumous Rights - A Critical Analysis

Author: LawSphereX

Category: General Law

Published: 2026-08-23T17:46:18.085837+00:00

Updated: 2026-08-26T17:12:21.605+00:00

AI-GENERATED VOICING AND POSTHUMOUS RIGHTS: A

CRITICAL ANALYSIS

Siva Raman K R, LL.M (Intellectual Property and Trade Law), Christ (Deemed To Be University), Bengaluru, Karnataka

ABSTRACT

Artificial Intelligence (AI) technologies, especially voice cloning, present unique challenges to India's frameworks governing personality rights, especially posthumous personality rights. Voice cloning creates a computergenerated replica of a person's voice, often using AI algorithms. Traditional protections aim to safeguard the identity and voice of individuals after death. However, the ability of AI to replicate a deceased individual's voice with high accuracy brings complex ethical and legal difficulties to the fore. This paper delves into the interplay between AI-generated voice cloning and posthumous personality rights within the Indian legal framework, underscoring the current legal system's shortcomings in managing the unauthorized use of cloned voices across media and digital domains.

With the advent of voice cloning technologies, India's legal infrastructure, which does not explicitly safeguard posthumous voice rights, faces new challenges. This exploration looks into the ethical dimensions surrounding consent, ownership, and control over a deceased person's voice. This becomes particularly pertinent in a cultural context where the voices of iconic figures from cinema and politics are deeply ingrained. The analysis critically reviews existing laws about intellectual property, privacy, and personality rights, pinpointing the legislative voids that render deceased individuals susceptible to exploitation via AI technologies. The paper advocates for immediate law reforms in India to protect individuals' rights posthumously in the AI era.

It suggests the formation of definitive legal frameworks and regulatory bodies tailored to address voice cloning intricacies, thereby preserving the dignity and legacy of the deceased. Through the lens of case studies, legal precedents, and international best practices, this discussion contributes toward shaping firm legal safeguards for posthumous personality rights in India, focusing on the ethical and legal challenges posed by AIinduced voice cloning. There is a need for a balanced approach that respects the memory and legacy of individuals while embracing technological advancements in a manner that is ethical, legal, and culturally informed. Keywords: Artificial intelligence, voice cloning, post-humous rights, personality rights, right to privacy

STATEMENT OF PROBLEM

The urgency of the problem lies in the Indian legal system's current inability to adequately address the complexities introduced by AI in replicating a deceased individual's voice with high accuracy. This deficiency raises pressing questions about consent, ownership, and control over one's voice after death, creating immediate vulnerabilities for exploitation in digital and media platforms. The lack of clear legislative guidelines to protect posthumous personality rights in the AI era is a matter of immediate concern, as it may lead to the misuse of these technologies, potentially tarnishing the legacy of individuals and infringing upon their posthumous dignity.

EVOLUTION OF AI-BASED VOICE CLONING

The Oxford English dictionary defines Al as "the capacity of computers or other machines to exhibit or simulate intelligent behavior. ”1 Artificial intelligence (AI) is a broad term encompassing machines capable of making decisions. Essentially, a machine's "artificial intelligence" refers to its decision-making ability derived from raw data provided by humans. 2Humans have an innate ability to understand and, thereby, learn things, but such ability is absent or not fully developed as such ability is not an inbuilt characteristic of artificial intelligence. But now we have come close where an artificial intelligence can learn and understand autonomously using neural network designed based on Human brain. 3 With regard to voice, AI-generated voice models rely on techniques like deep learning and textto-speech software to produce artificial voices.

However, before exploring AI-generated voices, it is important to grasp the concepts of machine learning and deep learning. Machine learning, a branch of AI, involves 'instilling intelligence into a system or machine without the need for explicit programming. ' Machine learning is currently applied across various services, including medical diagnoses, transaction fraud detection, and social media filters. Machine learning advanced significantly through 'deep learning,' a major step forward in replicating human intelligence using deep neural networks. A neural network is a series of algorithms that attempt to recognize underlying relationships in a set of data through a process that mimics the

1 Artificial Intelligence, OXFORD ENG. DICTIONARY,

https://www.oed.com/viewdictionaryentry/Entry/271625

2 Jojo John Moolayil, A Layman's Guide to Deep Neural Networks, TOWARDS DATA SCI. (July 24, 2019),

3 Generating Rembrandt, supra note 1, at 675; Deep Learning vs. Machine Learning: Beginner's Guide,

COURSERA (Aug. 11, 2022) [hereinafter "Coursera"], https://www.coursera. org/articles/ai-vs-deep-learningvs-machine-learning-beginners-

way the human brain operates. The term 'deep' refers to the multiple layers of computing units, known as 'neurons,' that are integrated into a single neural network. 4 This essentially breaks the impediment as it makes the machine learn from its mistakes and then adapt to changes without an external interaction based on a multi-layered neural web that works similarly to how a human brain works. Artificial voice is a term that does not have a proper and definite meaning. 5 AI-generated voice cloning is a newly developed technology that replicates a person's voice, resulting in what is referred to as a 'deepfake. ' A deepfake is a type of AI-generated content that uses deep learning to create convincing fake images, audio, and video. Before the advent of deep learning, voice cloning was a challenging task, constrained to two main approaches within text-to-speech (TTS) technology.

These approaches allowed a computer to read a user-written prompt and generate an audio recording of that text being 'spoken. '6 AI-generated voices, like any powerful technology, can be likened to a double-edged sword, with both beneficial applications and significant risks. While these voices have beneficial applications in areas such as translation, e-books, advertising, e-learning, augmented and virtual reality, brand mascots, patient management, and video game characters, they also pose significant risks. These voices can be misused for various harmful activities, including phishing scams, blackmail, fraud, and the spread of misinformation. This potential for misuse underscores the need for caution and robust legal safeguards. 7 AI- has been used to replicate the voices of celebrities and other prominent and established personalities worldwide. A few such examples are John F.

Kennedy ("JFK"), a former President of the United States, was assassinated in

1963 on his way to deliver a speech at the Dallas Trade Market. To honor JFK's legacy,

the creative agency Rothco collaborated with CereProc to artificially recreate his voice and deliver the speech he could not. The project took eight weeks and involved

4 Supra note 3

5 Matt Muldoon, Voice Artificial Intelligence: What Is It?, READSPEAKER (Sept. 29, 2020),

https://www.readspeaker.ai/blog/voice-artificial-intelligence [https://perma.cc/GB96-9K54].

6 Kim Martin, What Is Voice Cloning?, ID R&D, https://www.idmd.ai/what-is-voice-cloning

[https://perma.cc/NV7P-5E8T] (last visited Feb. 3, 2023). 7 Sarah Kuta, You Can Listen to Audiobooks Narrated by Artificial Intelligence, SMITHSONIAN MAG. (Jan. 9, 2023), https://www. smithsonianmag.com/smart-news/you-can-now-listen-to-audiobooksnarrated-by-artificialintelligencebut-is-that-a-good-thing-180981402/; Tomasz Jadczyk et al., Artificial Intelligence Can Improve Patient Management at the Time of a Pandemic: The Role of Voice Technology, J. MED. INTERNET RSCH. (May 25, 2021), https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8153030/ [https://perma.cc/GJ9Q-3VDV];

analyzing over 831 analog recordings to produce the 22-minute speech. They studied JFK's speech patterns using deep neural networks and generated an output that closely mimicked his voice. 8 Kim Kwang-seok, a beloved singer in South Korea, tragically took his own life in 1996 at the age of thirty-one, leaving his fans heartbroken. In 2021, Supertone, a South Korean startup, artificially recreated Kwang-seok's voice to perform a song on a show for the broadcaster SBS. Using singing voice synthesis (SVS), Supertone replicated Kwang-seok's voice, incorporating human emotions to make it sound more authentic. SVS technology learns voices by analyzing multiple songs and their corresponding notes and lyrics, allowing it to capture the late singer's unique style and pronunciation. The AI-generated voice was so realistic that it moved his fans to tears.

Unlike in the case of Anthony Bourdain, Kwang-seok's family consented to reproducing his voice. SBS paid the family a one-time fee for using his voice, and while parts of the show will be published on YouTube, Supertone has decided not to publicly release the song performed in the show. 9 These were examples of AI-generated voice models with predetermined outputs; on the other hand, Artificial voices can also be generated autonomously with limited human instruction. An example of such generated AI voice cloning is that of Steve Jobs, best known as the co-founder of Apple Computers, who passed away in 2011 due to cancer. 10 Podcast. ai, a podcast that generates topics and scripts entirely through AI, featured an episode where it created an impressively realistic version of Jobs' voice. The podcast utilizes Play.

ht, a tool that employs text-to-speech (TTS) technology to produce lifelike voices. The transcripts for the podcast are developed using "fine-tuned language models. " 11 8 Rothco Recreates JFK's Voice for Dallas Speech, MARKETING. IE, https://marketing. ie/rothco-recreates-jfksvoice-for-dallas-speech/ (last visited Jan. 4, 2024) 9 Gawon Bae, South Korea Has Used AI to Bring a Dead Superstar's Voice Back to the Stage, but Ethical Concerns Abound, CNN (Jan. 25, 2021), https ://www. cnn. com/2021/01/25/asia/south-korea-kim-kwang-seokai-dst-hnkintl/index. htmil [https://perma. cc/XVN7-E73M].

10 See Steve Jobs, BIOGRAPHY (Apr. 27, 2017), https://www.biography.com/businessfigure/steve-jobs

[https://perma.cc/DAU6-GN8U].

11 PODCAST.AI, https://podcast.ai/about [https://perma.cc/4NLU-9QLU] [hereinafter Podcast.ai]

HISTORICAL AND LEGAL FOUNDATIONS OF POSTHUMOUS PERSONALITY

RIGHTS

Posthumous personality rights are rooted in historical and legal traditions that recognize an individual's persona as a protected interest that survives beyond death. This notion extends from early legal doctrines that considered defamation of the deceased harm to their family and heirs to modern interpretations that see the continued commercial and personal value in an individual's likeness, name, and personality traits. Historically, the legal framework has evolved to balance the rights of the deceased with free expression and commercial interests, leading to a patchwork of laws and judicial decisions across different jurisdictions. Efforts to establish a cohesive body of law around posthumous personality rights reflect a broader understanding of the enduring impact of an individual's legacy and the moral and economic considerations in controlling and benefiting from that legacy.

This dynamic area of law continues to adapt, addressing challenges posed by technological advancements and the increasing commercialization of personal identities. The concept of posthumous personality rights from an Indian perspective intertwines with both traditional values and modern legal principles. India, with its rich cultural heritage, places significant importance on the reputation and honor of individuals, echoing these sentiments even after death. However, the legal framework in India does not explicitly recognize posthumous personality rights, leaving the concept in a somewhat nebulous state amidst the country's common law traditions. India's legal system is built on a combination of common law inherited from the British colonial era, statutory laws enacted by the Parliament, and traditional practices.

Regarding personal and property rights, the Indian Constitution does not specifically address the rights of individuals posthumously in the context of controlling or benefiting from one's legacy. However, the right to reputation is considered a part of the right to life under Article 21 of the Constitution, and this has been interpreted to extend some protections to the deceased, primarily through defamation laws, which can be invoked by the heirs or family members. As amended, the Copyright Act of 1957 is another area that indirectly affects posthumous personality rights by protecting the works of authors, composers, and artists after their death, thereby contributing to their legacy. The Act grants protection to the works of deceased creators

for a certain period (currently 60 years from the year of death of the author), allowing their estates or heirs to benefit financially. However, this does not cover the full spectrum of personality rights outside of their works, such as image, likeness, or name. Recent developments in India have shown an increasing awareness and discourse around the need for explicit legal recognition of personality rights, both in life and posthumously, against the backdrop of technological advancements and the commercial use of personal images and information. The courts have occasionally stepped in to protect the legacy of iconic figures, but a coherent legal doctrine is yet to emerge.

While the Indian legal and cultural frameworks recognize the importance of an individual's legacy, the specific concept of posthumous personality rights is still evolving, with much of the discourse and potential legal developments yet to unfold. This reflects a broader global trend of legal systems trying to keep pace with the implications of new technologies and changing societal values on legacy rights. Posthumous personality rights are rooted in historical and legal traditions that recognize an individual's persona as a protected interest that survives beyond death. This notion extends from early legal doctrines that considered defamation of the deceased harm to their family and heirs to modern interpretations that see the continued commercial and personal value in an individual's likeness, name, and personality traits.

Historically, the legal framework has evolved to balance the rights of the deceased with free expression and commercial interests, leading to a patchwork of laws and judicial decisions across different jurisdictions. Efforts to establish a cohesive body of law around posthumous personality rights reflect a broader understanding of the enduring impact of an individual's legacy and the moral and economic considerations in controlling and benefiting from that legacy. This dynamic area of law continues to adapt, addressing challenges posed by technological advancements and the increasing commercialization of personal identities.

EVOLUTION OF POSTHUMOUS PERSONALITY RIGHTS GLOBALLY

Posthumous personality rights originate in the broader notion of personality rights, designed to protect an individual's name, image, likeness, and voice. Initially, these rights were recognized primarily for living individuals, focusing on safeguarding personal dignity and preventing unauthorized commercial exploitation.12 However, as mass media expanded in the late 19th

12 Prosser, W. L. (1960). "Privacy." *California Law Review*, 48(3), 383-423.

and early 20th centuries, it became increasingly apparent that the identities of deceased individuals, particularly celebrities, required similar protection. The theoretical foundations for posthumous personality rights emerged, emphasizing that an individual’s identity and reputation should be safeguarded even after death, especially against commercial misuse that could distort their legacy. These foundations have been influenced by the evolving understanding of privacy, dignity, and property rights as they relate to human identity. 13 The United States has pioneered recognizing and developing posthumous personality rights, primarily through state-level legislation. 14 Known as the "right of publicity," these rights enable the estates of deceased individuals to control the commercial use of their persona, including their name, image, and voice.

States such as California and Tennessee, home to many entertainment celebrities, have enacted strong statutory protections extending posthumous rights for decades, allowing heirs to license or restrict the use of a deceased person's likeness. This legal recognition was largely driven by high-profile cases involving the estates of famous personalities, such as Elvis Presley and Marilyn Monroe, where courts confirmed that the right of publicity does not necessarily terminate with death. 15 These developments have also set a precedent for protecting posthumous personality rights in other jurisdictions. In civil law countries like France and Germany, posthumous personality rights are approached from a perspective deeply rooted in protecting human dignity.

16 These jurisdictions often view personality rights as inalienable, continuing to protect an individual's identity and reputation even after death. For instance, French law provides strong posthumous rights, particularly for famous figures, ensuring that their image and persona cannot be exploited commercially without proper authorization from their heirs. German law similarly extends protection to the deceased, viewing the unauthorized commercial use of a person's image or likeness as a violation of human dignity, which remains protected even posthumously. 17 These legal frameworks reflect a more comprehensive approach to personality rights, prioritizing the integrity and dignity of individuals beyond their lifetime.

13 McCarthy, J. Thomas. (2018). *The Rights of Publicity and Privacy*. Thomson Reuters.

14 Nimmer, Melville B. (1954). "The Right of Publicity." *Law and Contemporary Problems*, 19(2), 203-223. 15 Kwall, Roberta Rosenthal. (1997). "The Right of Publicity vs. The First Amendment: A Property and Liability Rule Analysis." *Stanford Law Review*, 70(1), 47-76. 16 Pollaud-Dulian, Frédéric. (2009). "Le droit à l'image des morts." *Revue Internationale de Droit Comparé*, 61(4), 831-850.

17 Von Lewinski, Silke. (2008). *International Copyright Law and Policy*. Oxford University Press

THE INDIAN LEGAL LANDSCAPE

A copyright confers an exclusive right on the owner, for a finite period, to prevent others from

exploiting its subject matter, an invention, a design, or a literary or other artistic work. 18 Thus, copyright may exist in, among other things, text, images, music, and video files transmitted via the internet, which would come within the ambit of the term ‘work’ for the purposes of copyright protection. 19 The Indian Copyright Act of 1957 confers performer's rights to artists in India under Sections 38A and 38B. However, these rights do not adequately protect individuals whose voices are replicated through artificial intelligence (AI). According to the Act, a 'performer' refers to a person who delivers a performance, and their rights only cover the visual or acoustic presentation they have performed. Thus, only personality rights can be invoked when voices are cloned using AI.

Personality rights protect individuals' identity, including their voice, likeness, name, image, and other distinctive identifying characteristics, from commercial exploitation. There are three main justifications for considering the protection of AI-replicated voices from the perspective of personality rights: first, the right to privacy; second, the prevention of unauthorized commercial use of the artist's identity; and third, the ethical imperative to restrict such uses. First, the right to privacy, which has been regarded as an integral part of Article 21 of the Indian Constitution20, is fundamental, as individuals have a vested interest in controlling how their voices, an intimate aspect of their identity, are used, even after their death. Unauthorized replication can infringe upon this privacy, leading to potential misuse.

Second, preventing unauthorized commercial use is crucial to safeguard against exploitation. the prevention of unauthorized commercial use aligns with the tort of "passing off," which is recognized under Indian law and has been extended to cover personality rights. 21 An artist’s voice is not just a personal attribute but also a valuable asset that can be commercially exploited. Without proper controls, there is a risk of individuals or companies profiting from an artist's voice without consent or compensation to their estate. Lastly, the ethical imperative to restrict such uses

18 Rodneyd Ryder, Intellectual Property and the Internet (Lexis Nexis Butterworths 2002) 64

19 Sumreen Siddiqui and Sonali Mathur, ‘Liability for Copyright Infringement on the Internet’ (2006) 5 Law Review GLC 147, 148.

20 Justice K.S. Puttaswamy (Retd.) v. Union of India (2017)

21 ICC Development (International) Ltd. v. Arvee Enterprises (2003)

underscores the importance of respecting the dignity and legacy of the deceased. Replicating a voice without consent may lead to scenarios that misrepresent the individual, potentially tarnishing their reputation or legacy. Thus, these three justifications collectively support the argument for protecting AI-replicated voices within the framework of personality rights.

COPYRIGHT OVER A PERSON’S VOICE AND AI-GENERATED WORK:

This issue in hand shall be dealt with twofold: firstly, the copyrightability of voice and secondly, the Copyrightability of AI-generated work. Under the current framework, Voice is not a subject matter of copyright. Under Section 13 of the Indian Copyright Act of 1957, the law protects only original works of authorship fixed in a tangible form. In the case of Midler v. Ford Motor Co. , the Ninth Circuit determined that since vocal sounds are not "fixed," a voice cannot be protected by copyright. 22 Secondly, concerning the copyrightability of AI-generated works, there are three essential criteria for granting copyright protection: “originality,” “work of authorship,” and “fixed in a tangible form. " For a work to be considered original, it must be created independently by the author without the contributions of others.

It should reflect the author's creativity without being copied from someone else's work. The original work does not necessarily have to be new or innovative, but it must go beyond a mere mechanical or routine processes that do not require creativity. In the case of AI-generated music under discussion, only the vocals are generated mechanically, and the user must carry out all other essential steps, such as writing lyrics, editing audio, mixing, and mastering. As the originality requirement is minimal, most works, including AI-generated music, easily fulfill this criterion. 23 The third requirement is that the generated work must be "sufficiently stable to be perceived, reproduced, or communicated for a period longer than transitory duration. " If music created with AI vocals is fixed in a tangible form, it satisfies the copyright requirement.

A work is fixed in a tangible medium if it is "sufficiently permanent or stable to permit it to be perceived, reproduced, or otherwise communicated for more than transitory duration”. The work can be fixed with the aid of a machine. 24 The tangible medium requirement is interpreted broadly. Examples of tangible mediums include paper, film,

22 Midler v. Ford Motor Co., 849 F.2d 460, 462 (9th Cir. 1988); 17 U.S.C. § 102(a).

23 McJohn, Stephen M. 2015. Examples & Explanations: Intellectual Property. 5th ed. Philadelphia, PA: Wolters Kluwer. 24 See Jane C. Ginsburg & Luke Ali Budiardjo, Authors and Machines, 34 BERKELEY TECH. L.J. 343, 410-11 (2019).

read-only memory (in the case of computer programs), computer files, and physical recordings. Another obstacle to AI music receiving copyright protection is the issue of authorship. Copyright protection is only applicable to works with human authors. Even though this requirement is not explicitly stated in the Copyright Act of India or the Copyright Law of the United States, the courts have declined to protect works created by non-human entities. The court in the well-known Naruto v. Slater case, also called the "monkey selfie copyright dispute," determined that animals did not have the legal right to file a claim under the Copyright Act. 25 A human photographer, David Slater, left his camera unattended in a wildlife reserve. A monkey named Naruto took a selfie with the camera, which Slater later included in a book.

Slater explained that he had left the equipment to attract the monkeys for a close-up photo26. He also stated that he had adjusted the camera settings to capture a clearer photo. People for the Ethical Treatment of Animals (PETA) sued Slater on behalf of Naruto, arguing that the photo resulted from a series of intentional and voluntary actions by Naruto, constituting original works of authorship by Naruto, not Slater. 27 The court ruled that, even if the photo was taken entirely autonomously without Slater's assistance, animals do not have the legal standing to bring a lawsuit for copyright infringement. Many argued that, based on this ruling, animals cannot be considered authors because only humans can assert authorship. Consequently, despite playing a major role in the voice cloning process, AI cannot be recognized as an author for copyright purposes.

28 However, voices have been protected as an integral part of personality rights. In a significant case, Amitabh Bachchan v. Rajat Nagi and Others ("Amitabh Bachchan Case"),29 the Delhi High Court made significant advancements in recognizing personality rights within India. The court issued a provisional in rem injunction prohibiting the unauthorized commercial exploitation of Amitabh Bachchan's distinctive personal characteristics, including his voice and likeness. Applying this order to both identified and unidentified defendants marked the first occasion in India where a general John Doe order was issued to safeguard an individual's

25 Naruto, 888 F.3d at 422.

26 Andres Guadamuz, Can the Monkey Selfie Case Teach Us Anything about Copyright Law?, WIPO MAG.

(Feb. 2018), https://www.wipo.int/wipo_magazine/en/2018/01/article_0007.html [https://perma.cc/GS9Q- KEYV].

27 Id

28 Zack Naqvi, Artificial Intelligence, Copyright, and Copyright Infringement, 24 MARQ. INTELL. PROP. L. REv. 15, 23 (2020).

29 2022 SCC OnLine Del 4110.

personality rights.

SUGGESTIONS:

Voice cloning or AI-generated voices are here to stay. It is time to make space for them within intellectual property law. This section delves into why the common law right of publicity, which is currently the only available method for controlling AI-generated voice models, is inadequate in addressing the abovementioned concerns and also puts forth suggestions to overcome this. The common law right of publicity is ill-prepared to tackle the challenges of crafting an artificial voice. Initially, the common law right of publicity pertains to "the inherent right of every human being to control the commercial use of his or her identity,"30 offering protection only against commercial uses. Commercial use refers to the exploitation of an individual's identity for the purpose of selling products or services.

Consequently, individuals cannot depend on their right of publicity to hinder the creation and usage of AI voice for noncommercial purposes. Secondly, even in territories that recognize a posthumous right of publicity, protection is constrained to a fixed period after one's demise. 31Hence, in those states, protection would still be too restricted to grant them or their estate sufficient control over the creation and utilization of their artificial voice. Remember the following text: "If users disclose that an artificial voice is being used, it might help avoid a common law right of publicity claim as the audience would then understand that it is not the actual person speaking. 32 Recently, actress Scarlett Johansson sued an AI app developer for using her voice and likeness in an advertisement. 33 The ad included a statement, "Images produced by Lisa AI.

It has nothing to do with this person. 34" The court might decide that the disclaimer shields the developer from liability. Additionally, applying the right of publicity to AI voices could discourage the development and use of AI technology. Suppose users fear being sued for violating the common 30 Mark Roesler & Garrett Hutchinson, What's in a Name, Likeness, and Image? The Case for a Federal Right of Publicity Law, AM. BAR ASS'N (Sept. 16, 2020), https://www. americanbar. org/groups/intellectual_propertylaw/publications/landslide/20 20-21/septemberoctober/what-s-in-a-name-likeness-image-case-for-federal-right-ofpublicity-law/ (quoting 1 J. THOMAS MCCARTHY & ROGER E SCHECHTER, THE RIGHTS OF PUBLICITY AND PRIVACY § 1:3 (2d ed. 2019)) (last visited Jan. 5, 2024). 31 Publicity Rights Under State Laws, JUSTIA, https ://www. justia.

com/entertainment-law/publicityrights/#:- :text=Generally%2C%20the%20righto20ofo20publicity,greatero20obstacles %20in%20establishing%20misappropriation [https://perma. cc/6TZL-6SWJ] (last visited Jan. 5, 2024). 32 Midler v. Ford Motor Co. , 849 F. 2d 460, 463 (9th Cir. 1988); Waits v. FritoLay, Inc. , 978 F. 2d 1093, 1102 (9th Cir. 1992).

33 Emma Roth, Scarlett Johansson Hits AI App with Legal Action for Cloning Her Voice In an Ad, VERGE

(Nov. 1, 2023), https://www.theverge.com/2023/11/1/23942557/scarlett-j 34 Scarlett Johansson Takes Legal Action Against Lisa AI App - Open AI Sea. https://openaisea.com/scarlettjohansson-takes-legal-action-against-lisa-ai-app/

law right of publicity using an artificial voice. In that case, they may not have the motivation to invest the time, money, and effort required to create an artificial voice. " It is argued in this paper to extend copyright law protection to AI-generated voice models and includes elements of consent and issuing disclaimers as requirements for copyrightability. As AI-generated voices become more advanced, conflicts arise between public, personal, and commercial interests. Consider the case of Kim Kwang-seok, as mentioned earlier. 35 The public wanted to hear more of his voice even after his death. His family sought compensation and wanted to protect his reputation. The company that developed his artificial voice aimed to profit from it. Kim Kwang-seok's family agreed to the use of his voice and received compensation.

However, what happens if the individual or their family does not want their voice artificially replicated or the company cannot profit from replicating it? The solution discussed in this part would help balance the public interest, right to privacy and commercial interest surrounding this issue. In Sushant Singh Rajput's case, the current Indian legal system states that rights to personality and likeness end upon death. Nevertheless, the financial advantages of commercializing an individual's persona and the ethical issues surrounding unrestricted usage create opportunities for these rights to be passed on to heirs. In the case of a deceased individual, this also safeguards the honor associated with that person and their family, assuming that the heirs will act in the best interests of the deceased.

Furthermore, the court ruling in the Sushant Singh Rajput case did not address the issue of replicating someone's characteristics through AI, thus limiting its relevance in today's context. While not widespread, some countries worldwide acknowledge celebrities' rights after their death. For example, California in the US has regulations in the California Civil Code, Section 3344. 1, that protect personality rights for 70 years after the individual's death. The Indiana Code Title 32, Section 36-1 grants publicity rights for the duration of a person's life and an additional century after their passing. Several court rulings in the US have also affirmed these rights.

35 Supra note 11