Introduction
We are all fond of some actors whether they are from Bollywood or any other country. People love them for their various characteristics acting, style, voice, catchphrases, image, personality, etc., which make them different from others and become an identifiable characteristic for that actor. Personality rights are the rights of famous personalities and celebrities whose name, voice, signature, or any other personality trait has commercial value and can mobilize and influence the public at large. Celebrities are known to people based on this very trait of them. It even becomes exclusive to them so that no other person can gain any benefit from it by using it without their permission. This gave them a unique right to control their image, deciding when it could be photographed and protecting against its misuse for commercial gain. Their personality rights encompass both privacy and publicity rights. However, recent advancements in AI and the rise in deepfakes have made it very easy to exploit these very exclusive rights of famous personalities. For instance, in April, Michael Schumacher’s family warned of potential legal action against a German magazine for using the voice AI generator platform Character.AI to fabricate a fake “exclusive” interview with the former Formula 1 champion.[1] This is violative, that someone is benefiting by using someone else’s personality rights and gaining benefits out of it without their consent. This concern has been very popular these days not just in Hollywood but in India also. Recently, the court in Anil Kapoor’s case has highlighted the tension between advancements in AI technology and individual rights. The court ruled in favor of the actor.[2] Courts up until now have ruled based on the right to privacy, the Trademark Act of 1999, and the Copyright Act of 1957[3]. In this article, we will first understand how these recent advancements in AI a cause of concern is. Some of the recent cases have raised voices against it. What is the current legislation and whether it is sufficient or not? Then we will look at some suggestions that can be used to tackle this growing problem.
Recent advancements in AI can be a threat to personality rights
The use of celebrities’ images without their consent isn’t new. Many local shops do have images of famous personalities to attract customers. But with the advent of this new AI era where copying someone’s voice, image, etc. in such a manner that it will look like a real one and can easily deceive is something new. By using AI, they can make an image or video that can act according to the need of the user will help them in getting benefits which is not justiciable. Also, it gave rise to deepfakes which is a digitally altered media piece that convincingly swaps one person’s likeness with another’s. A well-known instance is the “Deep Tom Cruise” videos by Metaphysic AI co-founder Chris Ume, which went viral on social media last year.[4] Additionally, AI deepfake websites have been used to fabricate interviews with elusive figures. So, how do these deepfakes make such a realistic visual that can violate someone’s personality rights? It uses a deep neural network which uses a face swapping technique. The videos can be completely unrelated; the target might be a clip from any movie, poster or a video, for example, and the videos of the person you want to insert in the film might be random clips downloaded from any social media platform. The program guesses what a person looks like from multiple angles and conditions, then maps that person onto the other person in the target video by finding common features[5]. This makes them look realistic that the celebrity is the one acting and not someone else. This is of course unfair to that person whose image, voice, or any other trait has been used to create benefits for others. Sometimes their traits have been used to diminish the image of that celebrity. For instance, Alia Bhatt’s case where a girl dressed in a blue floral co-Ord set with Alia’s face and making obscene gestures to the camera. Although netizens were able to identify that it is not an actual actor and a deepfake, this is a concern and not always one able to differentiate between the real and a deepfake.[6]
This creates a reasonable threat or violation to celebrities’ personality rights which comes under the Right to Privacy in terms of monetary loss and moral one. Where deepfakes are so easy to create using AI tools that anyone can create them acting as per the user commands and bringing benefits to them while creating loss to the person whose image has been used. This AI-generated image is different from traditional copyright infringement which usually involves copy-pasting an actor’s picture onto others which is quite detectable but with the use of deep fake, it can create a magical visual that is so real to identify. As we know, one of the main mediums through which it spreads is social media, yet its content moderation policies seem insufficient for managing this issue. In response, Facebook, in collaboration with Microsoft and Amazon, launched the ‘Deepfake Detection Challenge,’ which developed a technology to identify deepfakes. However, this technology only achieved an accuracy rate of 65.18%. This makes it ineffective in detecting deepfakes with 100% surety. Addressing the dangers of artificial intelligence, Union Minister of Electronics and Information Technology Ashwini Vaishnaw said, “a new crisis is emerging due to deepfakes” and that “there is a very big section of society which does not have a parallel verification system” to tackle this issue.[7] The problem highlighted by the minister is a cause of concern firstly, identifying deepfakes isn’t easy and secondly, we lack sufficient systems to detect them. Also, the current regulations of social media are not that stringent against them. Identifying the source of the content is also somewhat difficult to find when it spreads so quickly with reuploading by others.
Status of Legislation on Personality Rights and Its Insufficiency Against AI and Deepfakes
Recently in some of the cases, the court has widened the earlier regulations in terms of this new challenge and created precedence for it. For instance, the case of Anil Kapoor and Jacky Shroff highlighted the very same thing how it is unethical to use their traits without their consent with the use of AI which can affect the image and reduce the demand of actors as AI can do the same work. “The court can’t turn a blind eye to such misuse of a personality’s name and other elements, and dilution [and] tarnishment are all actionable torts which Kapoor would have to be protected against,” Justice Prathiba Singh reportedly said in the ruling, adding that the creation of ringtone and GIF images for commercial gains is a misuse.[8] However, this ruling of the court was not able to deter the illegal use of celebrities’ attributes for financial gains. The exception of fair dealing in the Indian Copyright Act somehow limits the actors’ right to exercise their personality. Just because they are public personalities doesn’t give others the right to use their traits which are created with the help of AI without their consent. It might be beneficial for the user, but it will also be deceptive for the audience to see their deepfakes acting rather than themselves. It doesn’t only involve using someone else’s face but sometimes their voices, talents, etc., which are so exclusive to them, and imitating them for monetary gain is unjustifiable even though it is in favor of society. The real challenge is taking concrete steps to identify violators, notify them, and ensure they comply with regulations. Jackie Shroff’s case is notable as it represents the first instance of a court imposing restrictions on an unlicensed chatbot. In this case, the court affirmed safeguarding the personality and publicity rights of film actors, restraining several social media accounts, AI Chatbots, and e-commerce websites from exploiting the actor’s name, voice, or image for commercial purposes without his consent.[9] This gave a ray of hope for the remedy against the misuse by the court. But still, proper legislation for this issue is the need for the hour which can deal with AI-generated Deepfakes.
The court has dealt with this issue until now with its incorporation with Article 21 of the Indian Constitution, which is the right to life, the Indian Copyright Act of 1957, and the Trademark Act of 1999.[10] But is it sufficient to deal with AI technological harms that are created with the deepfakes? This technology is evolving and advancing at a very fast pace which makes earlier legislation ineffective against changing needs. Current detection tools are not foolproof, often failing to identify deepfakes accurately and consistently which makes the action against them slow. Sometimes, Automated systems may generate false positives (flagging genuine content as deepfakes) and false negatives (failing to detect actual deepfakes). Both outcomes can undermine the credibility and effectiveness of the content moderation process.[11] Also, it is not necessary that the infringed content was created within the country and there is a possibility of it being produced internationally. The variance in different regions’ laws makes it hard to act against this. For instance, the US ‘Doctrine of Fair Use’, which is too broad as against the Indian one, grants protection to various forms of deepfakes including the ones created with malicious intent.[12] This highlights the need for common legislation to counter deepfakes of actors.
Conclusion
The increasing prevalence of deepfakes poses a significant threat to the personality rights of celebrities and public figures. Recent incidents involving prominent individuals such as Michael Schumacher, Alia Bhatt, and Anil Kapoor highlight the ethical and legal challenges associated with the misuse of AI technology to create realistic but false representations. Current legal frameworks, though addressing some aspects of this issue, fall short in handling the rapid advancements in AI and the complexity of deepfake technology. This gap necessitates the creation of more comprehensive and effective legislation to protect individuals’ personality rights. To effectively tackle deepfakes, a dedicated legal framework specifically addressing AI-generated content and deepfakes should be established. This framework should include aspects of personality rights, privacy, and intellectual property, with clear definitions and penalties for violations. Additionally, strict regulations should be enforced, requiring explicit consent from individuals before their likeness, voice, or any other personal attributes can be used in AI-generated content. This consent should be documented and verifiable to prevent unauthorized use. Investment in and the mandatory use of advanced deepfake detection technologies by social media platforms and other intermediaries are crucial. These technologies need to be regularly updated and improved to keep pace with AI advancements. Efficient and streamlined takedown procedures for infringing content should also be implemented, requiring intermediaries to act promptly upon receiving notifications of deepfake content to ensure swift removal and minimize harm. Promoting international collaboration to harmonize laws and enforcement mechanisms across jurisdictions is essential. This would help address deepfakes originating from different regions and ensure consistent protection for personality rights globally. Public awareness campaigns should also be launched to educate individuals about the existence and risks of deepfakes, enhancing media literacy to help people identify and mitigate the impact of deepfakes. By adopting these measures, legislators can create a robust legal environment that not only protects the personality rights of individuals but also addresses the unique challenges posed by AI and deepfake technologies. This comprehensive approach is vital for safeguarding the rights and reputations of celebrities and public figures in the digital age.
Author(s) Name: Nida Ghaffar (Jindal Global Law School, Op Jindal Global University)
References:
1 Jason Nelson, ‘‘Not Only Me’: Actor Anil Kapoor Wins AI Deepfake Court Case’ Decrypt (New York, 29 September 2023) <https://decrypt.co/198256/not-only-me-actor-anil-kapoor-wins-ai-deepfake-court-case>accessed 12 June 2024
[2] Jason Nelson, ‘‘Not Only Me’: Actor Anil Kapoor Wins AI Deepfake Court Case’ Decrypt (New York, 29 September 2023) <https://decrypt.co/198256/not-only-me-actor-anil-kapoor-wins-ai-deepfake-court-case>accessed 12 June 2024
[3] Khushi Saraf and Akshay Sriram, ‘The Dilemma of Deepfakes: Expanding the Ambit of Right to Personality to Regulate Deepfakes in India’ (2024) LSPR <https://lawschoolpolicyreview.com/2024/05/04/the-dilemma-of-deepfakes-expanding-the-ambit-of-right-to-personality-to-regulate-deepfakes-in-india/> accessed on 13 June 2024
[4] Jason Nelson, ‘‘Not Only Me’: Actor Anil Kapoor Wins AI Deepfake Court Case’ Decrypt (New York, 29 September 2023) <https://decrypt.co/198256/not-only-me-actor-anil-kapoor-wins-ai-deepfake-court-case>accessed 12 June 2024
[5] Dave Johnson and Alexander Johnson, ‘What Are Deepfakes? How Fake AI-Powered Audio and Video Warps Our Perception of Reality’ Business Insider (New York City, 15 June 2023) <https://www.businessinsider.com/guides/tech/what-is-deepfake?IR=T>accessed 12 June 2024
6 Editorial, ‘Alia Bhatt Falls Prey to Deepfake, Obscene Video Sparks Concerns over the Use of AI’ The Times of India (Delhi, 27 November 2023) <https://timesofindia.indiatimes.com/entertainment/hindi/bollywood/news/alia-bhatt-falls-prey-to-deepfake-obscene-video-sparks-concerns-over-the-use-of-ai/articleshow/105519162.cms> accessed 12 June 2024
[7] Ritika Jain, ‘India has a Deepfake Problem. But Is It A Crime’ BOOM (Mumbai, 23 November 2023) <https://www.boomlive.in/law/elections-pornography-laws-on-technology-deepfake-artificial-intelligence-23656> accessed 12 June 2024
8 Jason Nelson, ‘‘Not Only Me’: Actor Anil Kapoor Wins AI Deepfake Court Case’ Decrypt (New York, 29 September 2023) <https://decrypt.co/198256/not-only-me-actor-anil-kapoor-wins-ai-deepfake-court-case>accessed 12 June 2024
[9] Bharti Mishra Nath, ‘Why Celebrities Are Seeking Legal Protection for Their Personality Rights’ NDTV (New Delhi, 20 May 2024) <https://www.ndtv.com/opinion/why-celebrities-are-seeking-legal-protection-for-their-personality-rights-5705927> accessed 13 June 2024
[10] Neeraja Seshadri and Sindhu A, ‘Are Copyright Laws Adequate to Deal with Deepfakes?: A Comparative Analysis of Positions in the United States, India and United Kingdom’ (2020) KSLR <https://blogs.kcl.ac.uk/kslrcommerciallawblog/2020/12/17/are-copyright-laws-adequate-to-deal-with-deepfakes-a-comparative-analysis-of-positions-in-the-united-states-india-and-united-kingdom/> accessed 13 June 2024
[11] Neeraja Seshadri and Sindhu A, ‘Are Copyright Laws Adequate to Deal with Deepfakes?: A Comparative Analysis of Positions in the United States, India and United Kingdom’ (2020) KSLR <https://blogs.kcl.ac.uk/kslrcommerciallawblog/2020/12/17/are-copyright-laws-adequate-to-deal-with-deepfakes-a-comparative-analysis-of-positions-in-the-united-states-india-and-united-kingdom/> accessed 13 June 2024
[12] Neeraja Seshadri and Sindhu A, ‘Are Copyright Laws Adequate to Deal with Deepfakes?: A Comparative Analysis of Positions in the United States, India and United Kingdom’ (2020) KSLR <https://blogs.kcl.ac.uk/kslrcommerciallawblog/2020/12/17/are-copyright-laws-adequate-to-deal-with-deepfakes-a-comparative-analysis-of-positions-in-the-united-states-india-and-united-kingdom/> accessed 13 June 2024
