DEEPFAKES AND CRIMINAL LIABILITY: CHALLENGES UNDER THE INDIAN LAW

1. INTRODUCTION

Recently, a deep fake video of India’s finance minister, Nirmala Sitharaman, allegedly  endorsing a fraudulent high-return investment scheme, went viral. The video showed  the finance minister giving an official approval for an investment scheme promising  ₹15,00,000 in monthly returns on a ₹20,000 deposit. This video misleads the viewers  by inserting fabricated subtitles and overlaying promotional graphics which misleads  the viewers. Through this, it can be seen how deepfake videos mislead society by  making believable deepfake videos. 

Deepfakes videos are used for various purposes, especially with malicious intent,  which includes identity theft, financial fraud, defamation, cyber harassment, the spread  of non-consensual intimate images and most importantly, the spread of political  misinformation. This not only threatens the right of individuals but also undermines  the credibility of digital evidence and democratic processes. With the growth of  Artificial Intelligence (AI), the creation and spread of this content has become much  easier than before, which requires immediate legal and regulatory attention. 

In India, till now, there has been no specific legislation for governing deepfakes. The  existing legal framework relies upon the scattered provisions in the Information  Technology Act, 2000, Bharatiya Nyaya Sanhita, 2023, and the Copyright Act, 1975,  and constitutional principles relating to privacy and freedom of speech. Although these  laws provide certain remedies against offences involving impersonation, obscenity,  defamation, and identity theft, they were enacted without the anticipation of  complexities related to AI-generated media. Resulting in the adequacy of the current  legal regime to address the challenges generated by deepfakes, which remains a matter  of concern. 

The emergence of deepfakes has raised questions with concerns about criminal  liability, intermediary responsibility, and admissibility of electronic evidence, and the  balance between technological innovation and fundamental rights 

This present study, it seeks to examine the legal implications of deepfakes and also  evaluate the adequacy of the existing legal framework in regulating their misuse. This  study aims to identify the flaws in the current laws and also propose suitable reforms  for developing a comprehensive regulatory mechanism that might be capable of  addressing the emerging challenges which is caused by deepfake technology in India. 

2. UNDERSTANDING DEEPFAKES AND THEIR IMPLICATIONS 2.1 MEANING AND EVOLUTION OF DEEPFAKES

A deepfake is a synthetic media that is generated through artificial intelligence  techniques, which helps create or manipulate realistic images, audio recordings, and  videos. It is a video, photo or audio recording that seems real but is actually  manipulated by AI. Deepfakes can make someone do or say something where, in fact,  the words or the actions were never done by that person. Deep fake uses different AI,  i.e., Autoencoders and Generative Adversarial Networks (GANs). GANs produce more  realistic and convincing videos because it is made up of two competing artificial neural  networks, one that produces a fake, and the other trying to detect it. 

Initially, deepfake technology was developed for research and entertainment purposes.  But the rapid advancement of AI and its availability to everyone leads to the misuse of  this technology. It can now be used for exploitation and disinformation, and could  influence elections. In today’s world, deepfakes represent one of the most significant  challenges to digital trust and information authenticity. 

2.2 TYPES OF DEEPFAKE 

Deepfakes can be classified into these categories 

1. Video deepfakes: This involves swapping of facial expressions, lip movements and body gestures, which creates realistic videos of individuals doing an act or saying something that they shouldn’t. 
2. Audio deepfakes: Voice can be cloned too by imitating the tone of a person, accent, and speaking style. These deepfakes are mostly used in fraud and impersonation schemes.
3. Image deepfakes: AI tools can generate realistic images of a person or modify an  existing photo, and this is often done without the consent of the person 
4. Text-based deepfakes: AI systems can also produce human-like text that might be used as a means to spread misinformation or impersonate an individual in digital communications. 

2.3 POSITIVE APPLICATION OF DEEPFAKE TECHNOLOGY 

Although this technology is potentially used for misconduct, deepfakes also possess  many legitimate purposes. In the entertainment industry, it can be used for visual  effects, dubbing and content localisation. The education industry also uses synthetic  media to make learning more interactive. This technology also assists persons with  speech by using voice synthesis and contributes to preserving cultural heritage by  reconstruction 

Therefore, deepfakes itself is not unlawful; the legal concerns arise from the malicious  use of this technology. 

2.4 MISUSE AND EMERGING THREATS 

1. Violation of privacy: It involves the unauthorised use of an individual’s image, voice or personal data. Misuses like these infringe upon the right to privacy and personal autonomy under Article 21 of the Constitution of India. 
2. Identity theft and financial fraud: Voice cloning and face-swapping technologies enable cybercriminals to impersonate a person and deceive victims into transferring money or revealing confidential information. These amount to identity  theft.
3. Defamation and reputational harm: It may falsely show individuals engaging in  illegal or immoral conduct, which causes irreparable damage to that persons image  and social standing. Such acts raise concerns regarding defamation. 
4. Deepfake pornography: One of the most alarming applications is the creation of non-consensual explicit content. The victims, who are particularly women, suffer severe emotional, psychological, and physical harm to their reputation due to the  circulation of the media. 
5. Political misinformation and threats to democracy: Deepfakes have the potential to influence public opinion and the electoral process through the spread of fabricated speeches and misleading information. The misuse of these deepfakes threatens  democratic institutions and undermines public confidence in the media. 
6. Challenges to digital evidence: The increase in the authenticity and reliability of digital evidence raises a serious concern. Courts and investigative agencies faces difficulties in differentiating the genuine digital evidence from manipulated  content, thereby complicating the whole administrative justice system 

2.5 NEED FOR LEGAL REGULATION 

Widespread misuse of deepfakes demonstrates that the traditional legal framework is  not designed to address the complexities of synthetic media. The absence of legislation  specifically for deepfakes in India creates uncertainty regarding criminal liability,  intermediary responsibilities, and evidentiary standards. Therefore, there is an urgent  need for a comprehensive legal framework dedicated to balancing technological  innovation with the protection of privacy, dignity, and public trust. 

3. EXISTING LEGAL FRAMEWORK IN INDIA

3.1 INFORMATION TECHNOLOGY ACT, 2000 

India at present does not have a dedicated statute for deepfakes. Consequently,  offences arising from the misconduct of these media are addressed through various  provisions of the ITA,2000 

1. Section 66C (Identity theft): This section criminalises the fraudulent or dishonest use of an individual’s electronic signature, passwords or other unique identification features. 
2. Section 66D (Cheating by personation): This section punishes cheating through impersonation by means of computer resources or communicative devices, by generating videos and voice cloning to deceive individuals into disclosing  sensitive information or illegal transfer of money
3. Sections 67 and 67A (Publication of obscene and sexually explicit material): These  sections prohibit the publication and spread of obscene and sexually explicit  content in electronic form. Deepfake pornography may fall within the scope of  these provisions. 

Limitation of the IT Act 

The IT Act,2000, although it provides certain remedies, was made long before the  emergence of deep fakes. Therefore, it does not specifically define deepfakes, regulate  synthetic media or make mechanisms for the identification and removal of those  contents. 

3.2 BHARATIYA NYAYA SANHITA, 2023 

The Bharatiya Nyaya Sanhita, 2023, replace the Indian Penal Code, and this contains  several provisions that may be invoked against the deepfake offences. 

1. Cheating and impersonation: Deepfakes, which are used for deception, fraud or impersonation, may come under the provisions related to cheating and impersonation 
2. Defamation: The spread of fabricated videos or audio recordings that are capable of damaging a person’s reputation is defamation.
3. Forgery and fabrication: The creation of manipulated digital content with the  intention to deceive comes under forgery or fabrication of false evidence. 
4. Sexual offences: The use of deepfakes to create illicit content may amount to offences against women, depending upon the circumstances and judicial proceedings 

3.3 CONSTITUTIONAL PROTECTION 

1. Article 21 (Right to privacy): In the case of Justice K.S. Puttaswamy v. Union of India (2017), the Supreme Court recognised privacy as a fundamental right. Deepfake, which involves the unauthorised use of a person, infringes the right to  privacy, dignity, and bodily autonomy. 
2. Article 19(1)(A) (Freedom of speech and expression): While freedom of speech confirms the right to communicate and express ideas, such freedom is subject to reasonable restrictions under Article 19(2). The regulations of deepfakes must  therefore strike a balance between protecting free expression and preventing harm  arising from misconduct leading to defamation and privacy violations. 

3.4 COPYRIGHT AND PERSONALITY RIGHT 

The Copyright Act, 1957, protects original literary, artistic, musical and cinematic  works. Deepfakes that is created through the unauthorised use of these materials may  constitute copyright infringement 

3.5 BHARATIYA SAKSHYA ADHINIYAM, 2023 AND EVIDENTIARY CONCERNS 

Courts rely upon digital records, videos and audio recordings during the investigation  and trials; however, the deepfake content may undermine the credibility of the  evidence, which will complicate the whole judicial process 

The Bharatiya Sakshya Adhiniyam, 2023, recognises electronic records as admissible  evidence, but the increase of deepfake technology brings the necessity to make the  forensic mechanism stronger to distinguish genuine evidence from fabricated material 

3.6 NEED FOR SPECIFIC LEGISLATION 

The existing legal framework addresses the issue only indirectly through fragmented  provisions scattered across many statutes. These laws do not directly address the issue  relating to deepfakes, accountability, mandatory labelling, consent and intermediary  responsibility. 

Therefore, due to the rapid growth of this AI technology, there is a need for  comprehensive legislation designed for this matter 

4. COMPARATIVE ANALYSIS OF DEEPFAKE REGULATIONS

4.1 REGULATIONS OF DEEPFAKES IN THE UNITED STATES

The United States does not have a comprehensive law specifically for deepfakes.  Instead, regulations are carried out through state legislations and existing laws relating  to fraud, defamation, copyright and privacy. California and Texas have enacted laws  prohibiting the use of deepfakes California’s Assembly Bill 602 provided civil remedies for the victims of sexually  explicit Deepfake content, while Assembly Bill 730 prohibits the spread of this content  to influence elections. 

Despite these initiatives, the state still faces challenges in effectively addressing the  misuse of synthetic media 

4.2 EUROPEAN UNION AND THE AI ACT 

The European Union has taken a proactive approach towards artificial intelligence  regulations through the Artificial Intelligence Act. This act classifies AI systems based  on risk and imposes transparency obligations on developers and users One of the most important features of this act is that the AI-generated content must be  disclosed appropriately unless used for legitimate purposes. Through this approach, it  seeks to balance technological innovation with the protection of fundamental rights,  privacy and democratic values 

4.3 CHINA’S APPROACH TO SYNTHETIC MEDIA 

China has emerged as one of the first countries to adopt specific regulations regarding  deepfakes and synthetic media. The provision on the Administration of Deep Synthesis  Internet Information Services came into force in 2023, aiming to require service  providers to clearly label AI-generated content before creating or spreading the media 

4.4 LESSONS FOR INDIA 

A comparative study of international approaches reveals that many countries have  recognised the unique challenges posed by deepfakes and have adopted legal  mechanisms. In contrast, India currently relies on scattered provisions under many  different statutes. 

Unlike the EU and China, India lacks legal requirements relating to content labelling,  platform accountability and disclosure of AI-generated media. Furthermore, there are  no specific laws governing deepfakes and synthetic media Therefore, India should learn valuable lessons from the international approach towards  deepfakes by adopting a specific legal structure made only for deepfakes, which will  take into account the transparency obligations, enhanced intermediary obligations,  effective remedies for the victim and mechanisms for authenticating digital evidence.  Reforms like this would enable the law to keep pace with the ongoing growth of  technology while ensuring the protection of the rights of an individual. 

5. CHALLENGES IN REGULATING DEEPFAKES

5.1 ABSENCE OF SPECIFIC LEGISLATION 

India currently does not have legislation specifically regarding deepfakes. It uses  existing provisions under the Information Technology Act, 2000, and the Bharatiya  Nyaya Sanhita, 2023 to address crimes related to cheating and defamation and  obscenity, but it does not have a law specifically dealing with AI-generated synthetic  media 

5.2 JURISDICTIONAL CHALLENGES 

These generated media can be created and spread across multiple countries, making  investigations and prosecution difficult. The transactional nature of the crime makes it  difficult and complicated to enforce, and international cooperation 

5.3 INTERMEDIARY LIABILITY 

Social media platforms play a crucial role in the dissemination of deepfake content.  Deciding the extent of their responsibility while preserving freedom of speech still  remains an active challenge 

5.4 EVIDENTIARY CHALLENGES 

These technologies undermine the authenticity of the digital evidence, making it hard  for the judiciary to pass a judgment 

5.5 BALANCING REGULATION AND FREE SPEECH 

Any attempt to regulate deepfakes must balance the protection of privacy and  reputation with the constitutional guarantee of freedom of speech and expression 

5.6 NEED FOR COMPREHENSIVE REGULATION 

The fragmented legal framework in India is not sufficient to address the risk posed by  deepfakes. There is an immediate need for a comprehensive legal framework and a  stronger platform for accountability to combat the issue of AI-enabled crimes. 

6. RECOMMENDATION AND WAY FORWARD

6.1 ENACTMENT OF A DEDICATED DEEPFAKE LAW: 

India should adopt a comprehensive legal framework specifically addressing the  creation and the spread of AI-generated malicious deepfake content. Such legislation  should define deepfakes and suggest appropriate civil and criminal penalties 

6.1 MANDATORY DISCLOSURE AND WATERMARKING 

AI-generated content should be labelled by using labels or digital watermarks to ensure  complete transparency and facilitate the identification of the owner of the manipulated  media 

6.3 STRENGTHENING PLATFORM ACCOUNTABILITY 

Social media platforms and intermediaries should be required to implement an  effective mechanism for detecting and removing harmful content while respecting  freedom of expression 

6.4 ENHANCING DIGITAL FORENSICS 

The government should strengthen its cyber forensic capabilities and equip  investigative agencies with advanced technology to check electronic evidence and  detect deepfake media 

6.5 PROMOTING PUBLIC AWARENESS 

There should be digital literacy programs and awareness campaigns to educate the  citizens about the risks associated with deepfakes and the legal remedies 

6.6 INTERNATIONAL COOPERATION 

Given the transactional nature of Deepfake-related offences, India should cooperate  wth international organisations and foreign jurisdictions to develop uniform standards  and facilitate effective enforcement 

7. CONCLUSION

The rapid growth and advancement of Artificial Intelligence have significantly  increased the creation and misuse of deepfake technology, resulting in it posing serious  threats to privacy, reputation, democratic institutions, and the authenticity of digital  evidence. While deepfakes have legitimate applications, the use of deepfakes for  malicious purposes like identity theft, fraud, defamation, and non-consensual intimate  content presents dangerous challenges to existing legal frameworks. 

India still relies on fragmented legal provisions from various statutes, including the  Information Technology Act, 2000, and the Bharatiya Nyaya Sanhita, 2023 and other  related laws to address deepfake-related offences. However, these laws are not  designed specifically for deepfake-related offences, which therefore makes them  inadequate in dealing with the evolving nature of such crimes. A comparative analysis  of different jurisdictions, like the US, EU, and China, shows how the growing global  recognition needs targeted regulation. 

Therefore, there is an urgent need for a comprehensive legal framework that clearly  defines deepfakes, strengthens platform accountability, enhances digital forensic  mechanisms, and protects individual rights without stripping the freedom of speech  and technological innovation. Effective regulation, along wth public awareness and  international cooperation, is essential to ensure that the benefits of AI are realised  while minimising the risk associated with its misuse and misconduct 

REFERENCE(S):

Cases

1. Justice K.S. Puttaswamy v Union of India (2017) 10 SCC 1.
2. Shreya Singhal v Union of India (2015) 5 SCC 1.
3. Anvar PV v PK Basheer (2014) 10 SCC 473.
4. Arjun Panditrao Khotkar v Kailash Kushanrao Gorantyal (2020) 7 SCC 1.

Statutes

5. Information Technology Act, 2000.
6. Bharatiya Nyaya Sanhita, 2023.
7. Bharatiya Sakshya Adhiniyam, 2023.

Articles

8. Robert Chesney & Danielle Citron, Deep Fakes: A Looming Challenge for Privacy, Democracy, and National Security (2019) 107 California Law Review 1753.
9. Yisroel Mirsky & Wenke Lee, The Creation and Detection of Deepfakes: A Survey (2021) 54 ACM Computing Surveys 1.
10. Piyush Jha & Simran Jain, Detecting and Regulating Deepfakes in India: A Legal and Technological Conundrum(2021) SSRN.