Right to Reputation in the Age of Artificial Intelligence: Navigating the Challenges of Synthetic Media

I. Introduction

“A man’s reputation a part of himself, as his body and limbs are, and reputation is a sort of fight to enjoy the goods of opinion of others, and it is capable of growth and real existence, as an arm or leg. Reputation is, therefore, a personal fight, and the right to reputation is put among those absolute personal fights equal in dignity and importance to security from violence..”²- Justice V.R. Krishna Iyer.

The observation made by Justice Iyer in Kiran Bedi and ors v. Committee of Inquiry, aptly highlights the significance of reputation as an integral aspect of human life. A person’s reputation is not merely a superficial attribute, rather, it is an integral part of their character and integrity, which is essential to living a life of honour and respect in the society. It is, however, a delicate entity, susceptible to external influence and beyond our direct control. This emphasises how crucial it is to protect one’s reputation because it has a significant influence on one’s status and place in society. It underlines the necessity to safeguard one’s reputation, as it can gravely impact one’s standing and position in the society.

The right to reputation is based on the principle of “Actio Injuriarum”, which is a component of the Roman Dutch Law of Delict, provides a solution for the intentional violations of personal rights concerning dignity and reputation (animus injuriandi).³ Although there is no specific definition for right to reputation but Defamation has existed as a wrong in the society since a very long time. Reputation, although beyond our control, is inter-twined with the element of privacy and dignity. The European Convention on Human Rights, while providing protection for freedom of expression, acknowledges that freedom comes with attendant responsibilities, and such freedoms may be limited to uphold various social interests, like the recognition of reputation as a vital facet of individual rights.⁴

The delicate balance between individual liberties and the right to reputation, has been enshrined under prominent International documents. Article 17 and 19 of the ICCPR affirms that no individual should face arbitrary and unlawful interference with their privacy, home or correspondence nor should they endure unjust attacks on their honour and reputation, not at the cost of other’s right to freedom of speech and expression.⁵ This balance gets more difficult in the era of larger and more complex data sets, which emphasises the necessity to handle and navigate such contemporary challenges where reputation intersects with data protection.⁶ Reputation, integral to personal and psychological integrity⁷, functions as a crucial social signal, therefore, emphasising the importance of preserving individual dignity within the scope of private life and strike a balance wherever necessary.⁸

The cornerstone of democracy lies in the principles of free speech and expression. However, this freedom also imposes a duty to uphold and safeguard individual reputation, which is integral to human dignity. Additionally, democracy requires measures to shield against harmful falsehoods that defame and tarnish one’s reputation and character in the society.⁹

II. Right to reputation in india

In the Indian society, reputation stands as a pillar of immense importance, often prized above material wealth. A noble name commands respect and admiration, while tarnished reputation face social exile. Defamation, therefore, carries significant consequences, inflicting deep wounds upon the individual’s character. The sanctity of one’s reputation is held as a sacred trust and any stain on it invites shame, trauma and relentless harassment. In this cultural ethos, reputation remains the most priceless possession, a testament to one’s integrity and honour. The Supreme Court, while deciding the case of Subramaniam Swamy v Union of India¹⁰, observed in its judgement a prominent shloka from the Subhashitratbhandagaram,- “One who possess fame alone does live….. who has no fame and negative praise is equal to one who is dead while alive..”¹¹. The court in this case heavily stressed on the role of reputation within the right to life as enshrined under Article 21¹², recognising it as a fundamental right. The necessity to balance the fundamental rights is paramount, as the court cautioned against sacrificing one’s reputation on the altar of another’s freedom of speech.¹³ The court examined defamation not only in the light of a civil wrong but also a criminal offence by elucidating that, “causing harm to the reputation of a person is the basis on which the offence is founded and mens rea is a condition precedent to constitute the same offence”.¹⁴

In a recent case of violation of right to reputation for being referred to as an “accused” to a cheating case on various Internet platforms even after being acquitted from the same. Submitting a plea, the petitioner contented, “Reputation of a person is neither metaphysical nor a property but an integral part of his sublime frame and a dent in it is a rupture of a person’s dignity, negated and infringes the fundamental values of citizenry rights..”¹⁵ Defamation is therefore a well-recognised civil as well a criminal offence in India. The Law Commission of India also submitted its Report No. 285 in February 2024, to retain defamation as a criminal offence, observing that India is a nation that successfully and joyfully coexists with many languages, ideologies, and ways of thinking because of the nature of society, people desire to enjoy both their freedom and to protect things they value the most¹⁶. Emphasising on the need to retain the criminalisation of defamation, it submitted that “reputation is something which can’t be seen and can only be earned. It’s an asset which is built in a lifetime and destroyed in seconds. The whole jurisprudence around the law on criminal defamation has the essence of protecting one’s reputation and its facets.¹⁷”

However, with contemporary times comes contemporary problems especially rapidly evolving technological landscapes bring forth novel threats to individual reputation. Addressing these emerging risks and challenges urgently demands swift action, forward thinking and interdisciplinary approach by strategically utilising such cutting edge technologies like Artificial Intelligence, which, although created by human beings has ironically begun to pose threats to the protection and safeguard of human rights itself. Thus, it becomes imperative to adapt and leverage innovative solutions to safeguard personal reputation in this digital age.

III. Artificial intelligence and reputation manipulation

Artificial Intelligence is a product of humanity’s desire for convenience and advancement.¹⁸ Throughout history, humans have sought to create tools and technologies to enhance the efficiency of the tasks, pursuit of solving complex problems and pushing the boundaries of human knowledge. While AI technologies have the potential to completely transform and revolutionise various aspects of our lives, like healthcare, transportation and communication, their rapid development also brings up serious issues with fairness, privacy and autonomy.

One of the primary challenges comes from the inherent biases in AI algorithm, which reinforces inequality and discrimination, especially when it comes to the decision making process in the criminal justice system. The increasing reliance on AI in every field of human development has raised concerns about privacy and government overreach. Especially with the spread of deep fake technology, AI driven misinformation and manipulation of the digital media jeopardises the credibility and trustworthiness of the sources of such information, which threatens individual’s reputation.

The emergence of this ‘infodemic’¹⁹, has seen a boost since the Covid 19 lockdown, when people helplessly began to rely on the digital media. The dissemination of correct and incorrect information, morphed and phony images or videos of celebrities using AI technologies like Deep Fakes,²⁰ has become immensely popular, particularly in recent years. The introduction of new AI technologies every month, presents a significant challenge, as the pace of innovation outstrips the development of strategies to effectively manage their impact. This relentless cycle not only exacerbates existing challenges but also introduces new risks that compromise the reputation and security of individuals.

A recent case involving Bollywood actor Anil Kapoor²¹, vividly illustrates the danger posed by AI technology setting a poignant example of the real world consequences of AI misuse. Kapoor found himself targeted by various AI entities that utilised his pictures, signature dialogues, and other elements of his likeness without his consent. Recognising the potential harm to his reputation and right to live with dignity, he sought legal recourse and was granted an ex parte omnibus injunction by the Bombay High Court, restraining 16 entities from exploiting his name, persona and images using AI tools.²²

Similarly, the recent circulation of deep fake videos on social media featuring Rashmika Mandanna, an Indian actress, reflects the growing prevalence of AI driven manipulation in digital content creation.²³ The ease at which individual’s identities can be exploited and misrepresented, raises concerns about its potential impact of trust, reputation and more so the lack of robust measures to tackle and safeguard individual’s rights against such deceptive content.

The dual capacity of AI to both fabricate false content as genuine and undermine trust in authentic images and videos, is a serious threat particularly within the criminal justice system, where documentary evidence plays a crucial role, especially cases involving defamation and reputation protection.²⁴The escalation of AI entities, with new and upgraded features emerging daily, aggravates these challenges. For instance, Open AI’s video generator SORA, capable of producing life like videos with minor glitches only based on text prompts, introduces a new level of complexity, making it increasingly difficult for the Government and the individuals to effectively combat the spread of false information.²⁵ The wide accessibility of such technology, could facilitate hateful, sexual and violent imagery and videos targeting both public and private individuals alike.²⁶AI technologies have therefore been wielded as potent tolls capable of inflicting significant harm, from creating lifelike videos, images and voice mimicry, AI manipulation has transcended traditional boundaries, posing profound threats to personal integrity and societal trust.

IV. The recent case of walters v openai

The discussion of AI and its complexities in the digital landscapes, necessitates the discussion of a recent case which involved the intricate intersection of AI technology and reputation management and the tarnishing of an individual’s reputation caused by a prominent AI chatbot, ChatGPT, developed by OpenAI. In the very first case of defamation and injury of the right to reputation by an Artificial Intelligence chatbot, Mark Walters v. OpenAI²⁷, Mark Walters, a citizen and resident of the State of Georgia, the United States and a radio host by profession, had filed a defamation suit before the Superior Court of the Gwinnet County, State of Georgia against a prominent AI limited liability company, OpenAI L.L.C, claiming to have produced defamatory statements against the plaintiff while a third party, Fred Riehl, a journalist used ChatGPT, an AI chatbot platform by OpenAI to inquire about a certain lawsuit. In the interaction, ChatGPT had provided false details of the lawsuit as prompted by Fred Riehl, and accused Mark Walters, the plaintiff of being engaged in serious criminal offences as misappropriation of funds and embezzlement even though the plaintiff had never been involved in any such lawsuits.²⁸

The defendant argued that the plaintiff had failed to establish the grounds for defamation and although stated that, despite conveying its limitations and risk of producing inaccurate information, chatGPT was continuously pushed and its warnings were disregarded, compelling it to produce such output.²⁹ However, the court rejected OpenAI’s plea as of the latest development in the case.³⁰

The case requires the discussion of the essentials grounds for defamation-

• The statement must be likely to tarnish the reputation of a person.³¹
• It must be reasonably inferred from the statement that it exclusively refers to the plaintiff³², and
• Such statements must be published, even if it is published to the person defamed.³³
• Presence of element of harm and injury of caused by such statement.

However, any suit of defamation, whether a libel or a slander, would be held as defamation per se on the fulfilment of a number of conditions, of which, accusing someone of committing a serious crime tops the list. Therefore, in Walters v. OpenAI, ChatGPT’s fake prompts of accusing Walters of committing crimes as serious as embezzlement fulfils all the above conditions and therefore the harm is presumed and the party might not need to prove specific damages caused by such statements.

Therefore, through a careful examination of the facts of the case, it becomes evident that there is an urgent requirement for stringent accountability measures in the realm of Artificial Intelligence particularly concerning the dissemination and generation of content and statements that might potentially damage a person’s reputation, reflecting the requirement of ethical and legal responsibilities upon the developers as well as the users of AI technologies.

V. Challenges in suing artificial intelligence for breach of reputation in india

Artificial Intelligence being a non-human entity poses substantial challenges in the field of defamation actions against it. Generally speaking, a lawsuit for defamation needs to meet a number of requirements, like- defamatory nature of the statement, specific targeting towards the aggrieved individual, evidence of harm caused etc, as already enumerated above. Additionally, Defamation law also makes distinction between defamation per quod, which required the plaintiff to establish harm or damages caused by the statement, and defamation per se¸ which do not require the petitioner to prove any harm caused per se if the following conditions are satisfied-

• Being accused of committing a serious crime, or having a venereal/infectious disease,³⁴
• To question a woman’s chastity and accuse her of loose character,³⁵
• If the statement suggests a behaviour which is inconsistent with an individual’s role, occupation,³⁶ etc.

Another essential element of defamation lawsuits is the existence of “actual malice”. To prove the presence of “actual malice” in defamation, the following two conditions has to be satisfied as enumerated in the case of New York Times Co. v Sullivan³⁷,-

i. The defendant while publishing the statement had an intention to convey the defamatory meaning of the statement³⁸.

ii. The defendant must possess a high degree of awareness of about the lack of credibility of the statement³⁹.

Therefore, the important issue that comes up while considering suing an AI platform like a chatbot or a video generator for defamation, crucial questions arise as to the accountability for such wrong. The foundation of a defamation suit lies in the existence of “actual malice”⁴⁰, but as products of Artificial Intelligence, they are devoid of emotions and consciousness, so can they be held liable for malice? Who can then be held liable when the entity in question is an AI tool? Can tools like a chatbot even be considered capable of being sued for defamation?

In India, there is no distinction between a Libel and a Slander, both are equally punishable under defamation. However, to constitute any criminal suit, the presence of mens rea is imperative. Sec 499 of the Indian Penal Code, defines Defamation as, “Whoever by words either spoken or intended to be read, or by signs or by visible representations, makes or publishes any imputation concerning any person intending to harm, or knowing or having reason to believe that such imputation will harm, the reputation of such person, is said, except in the cases hereinafter excepted, to defame that person⁴¹” which also stresses on the element of “knowledge” and “intention” behind such defamation.

Therefore, with no direct legal provision in India for suing AI entities, the focus shifts to- individuals associated with the development, utilisation and dissemination of AI tools. Nevertheless, the ambiguity persists regarding who bears responsibility- should it be the developers, or the users, or those disseminating false information generated by AI?⁴² Given that the existing laws typically require proof of actual malice and intention, the spotlight turns to the human entities connected to the AI tools. However, Sec 79 of the Information Technology Act, 2000 further complicates the matter by stating, “…an intermediary shall not be liable for any third party information, data or communication link made available or hosted by him”.⁴³ This provision further aggravates the challenge of identifying the appropriate parties to be held accountable. This not demands for a clarity in the legal approach by addressing such ambiguities, but also requires ethical reflection surrounding technological innovation, as it is essential for upholding justice and safeguarding individual’s rights in the digital age.

VI. Conclusion and suggestions

With great technological advancements comes greater responsibilities to uphold the rights and dignity of every individual. In the era of Artificial Intelligence, safeguarding one’s reputation requires a multidimensional strategy that takes social, ethical and legal factors into consideration. As demonstrated in the case of Walters v OpenAI⁴⁴, the arbitrary use of AI tools can pose significant threat to individual’s reputational rights. It is very important for the society, to recognise that AI technologies like ChatGPT, SORA etc. are creations of human intelligence and have the capacity to both enhance and violate the fundamental rights of individuals. Mindful and careful utilisation of AI tools is crucial, with users and developers alike bearing the responsibilities that these technologies are not wielded in a manner that violated the rights of others.

As the proverb goes, Prevention is better than cure, given this, preventative actions are essential to tackle the legal challenges created by AI. Unfortunately, the existing regulations of defamation under the Indian legislations like the IPC or the IT Act fall short and are insufficient to cover all dimensions of the issue. Consequently, it is necessary to adopt legislation that will specifically address the intersection of Artificial Intelligence and Human Rights, even with the legal background, it is necessary to further think about ethical policy consideration. AI developers not only should follow a list of ethical guidelines but also prevent technology from creating and distributing defamatory content. Similarly, users of such tools must also exercise caution and responsibility when using AI technologies, ensuring that their actions adhere to the ethical principles and that they respect the rights of others.

In this light, protecting the right to reputation in the age of AI, necessitates a collective effort from law makers, legal experts, technologists, and the society as a whole. By cultivating an ethical AI development culture and establishing legislations that will also protect and preserve our rights, we can transverse the difficulties of the digital era while preserving the dignity and integrity of every individual. Therefore, as we harness the power of AI, let us never forget the duty to protect the intrinsic worth and reputation of every individual our digital ecosystem.

*****

Footnotes

1. Author is a LL.M. student at Indian Institute of Legal Studies, India.

2. Kiran Bedi v. Committee of Inquiry, 1989 AIR 1618, 1989 SCR (3) 997 ↩

3. Mangala Wijesinghe, ‘Actio Injuriarum in Roman Law of Delict: Elements of Liability (2018) ↩

4. European Convention on Human Rights, available at: https://www.echr.coe.int/documents/d/echr/conve ntion_ENG (last visited on February 29, 2024) ↩

5. International covenant on Civil and Political Rights 1966, available at: https://www.ohchr.org/en/instruments-mechanisms/instruments/international-covenant-civil-and-political-rights (last visited on February 29, 2024) ↩

6. Yvonne McDermott, ‘Conceptualising the right to data protection in an era of Big Data’ 4 SAGE OPEN 211 (2017) ↩

7. Putistin v Ukraine, App no. 16882/03 (2013) ↩

8. Anne SY Cheung, Wolflang Schulz, ‘Reputation protection on online rating sites’ 21 Stan. Tech. L. Rev. 310 (2018) ↩

9. Law Commission of India, 285^th Report on The Law of Criminal Defamation (February 2024) ↩

10. Subramaniam Swamy v. Union of India, AIR 2016 SC 2728 ↩

11. ibid ↩

12. The Constitution of India, art. 21 ↩

13. id. At 9 ↩

14. ibid ↩

15. S Sakeer Hussain v Union of India & Anr (2023) ↩

16. Supra note 8 at 2 ↩

17. ibid ↩

18. Meenakshi Nadimpalli, ‘Artificial Intellligence Risks and Benefits’ 6 IJIRSET 1 (2017) ↩

19. World Health Organisation, available at: https://www.who.int/health-topics/infodemic (last visited on March 1, 2024) ↩

20. Ashish Jaiman, ‘The danger of deepfakes’, The Hindu, Jan. 01, 2023 ↩

21. Anil Kapoor v. Simply Life & ors (2023) ↩

22. Naik Naik & Company, available at: https://naiknaik.com/2023/10/17/vtubers-live-streamers-using-copyrighted-materials-its-legal implications/. (last visited on March 1, 2024) ↩

23. Shivani Mankerni, ‘The TRUTH behind Rahsmika Mandanna’s Deepfake video’, The Times of India, Nov. 8, 2023 ↩

24. Joseph Kroetsch, ‘Skepticism in era of AI Deepfakes will erode defamation claims”, Bloomberg Law, Apr. 20, 2023 ↩

25. New Scientist, available at: https://www.newscientist.com/article/2417639-realism-of-openais-sora-video-generator-raises-security-concerns/ (last visited March 1, 2024) ↩

26. ibid ↩

27. Walters v OpenAI, Ga. Super. Ct,., No. 23-A-o4o860- 2, 1/11/24 ↩

28. Id at 25 ↩

29. Ibid ↩

30. Knowing Machines, available at: https://knowingmachines.org/knowing-legal-machines/legal-explainer/cases/ walters-v-openai (last visited on Mar. 4, 2024) ↩

31. Salmond and Heuston, Law of Torts 197, (RVF Heston and RA Buckley, London, 7^th edn, 1923) ↩

32. Dr. RK Bangia, Law of Torts 154, (Allahabad Law Agency, Haryana, 23^rd edn, 2013) ↩

33. R v. Adams, 1888 ↩

34. Bloodworth Law PLLC, available at: https://lawyerfightsforyou.com/what-is-per-se-defamation-vs-per-quod-defamation/ (last visited on Mar. 4, 2024) ↩

35. ibid ↩

36. id ↩

37. New York Times v. Sullivan, 376 U.S. 254 (1964) ↩

38. American Bar Association, available at: https://www.americanbar.org/groups/communications_law/public ations/communications_lawyer/2023-summer/how-best-explain-actual-malice-juries-starters-dont-use-those-words/ (last visited on Mar. 6. 2024) ↩

39. ibid ↩

40. NOLO, available at: https://www.nolo.com/legal-encyclopedia/artificial-intelligence-defamation-and-libel-is-anyone-liable.html (last visited on Mar. 6, 2024) ↩

41. The Indian Penal Code, 1860 (Act 45 of 1860), s. 499 ↩

42. Supra note 39 at 8. ↩

43. Information Technology Act, 2000 (Act 21 of 2000) s. 79 ↩

44. Supra note 26 at 6 ↩