Increasing The Cases Of Deepfake And Its Regulations In India
Author: LawSphereX
Category: General Law
Published:
Updated:
INCREASING THE CASES OF DEEPFAKE AND ITS
REGULATIONS IN INDIA
Saurabh Shrivastava, Junior Research Fellow in faculty of Law, University of Lucknow Prof. (Dr.) S.K. Singh, Professor at Faculty of Law, Sri JNPG College, Lucknow
ABSTRACT
As the technology is evolving and digitalisation is happening, new challenges are coming into the picture. Deep fake is one of such challenge. The problem of the deepfake is growing rapidly and it have a very serious impact upon the country. The proliferation of deepfake technology has presented unprecedented challenges to legal frameworks worldwide, raising significant concerns about privacy, defamation, intellectual property, misinformation, and national security. Prime Minister Modi also raised the concern of the deepfake and declared it as the new age problem. Internet is full of the deepfake contents, primarily pornographic videos. South Korea declared the Digital emergency due to growing concerns of deepfake. Deepfake video of renowned actress Rashmika Mandana made this issue a topic of discussion for the whole country.
This paper explores the complexities of regulating deepfakes within existing legal paradigms, focusing on the intersection of technology, ethics, and the law. It examines the inadequacies of current regulations in addressing the malicious use of deepfakes, including their potential to manipulate elections, propagate fake news, and harm individuals through non-consensual explicit content. Drawing from comparative legal analyses, this study evaluates legislative efforts in jurisdictions such as the United States, European Union, and India, highlighting the tension between ensuring freedom of expression and protecting individual rights.
By proposing a framework for adaptive legal responses, including enhanced digital literacy, improved AI detection technologies, and robust accountability mechanisms, the study aims to provide a comprehensive blueprint for policymakers to mitigate the adverse impacts of deepfake technology while fostering its positive applications.
Introduction Advancement of information technology and artificial intelligence helps the people to solve the real-time problems of the world. But challenges of these technologies are also very complicated and needs the expertise and technical solutions. Deep fake images and videos are one of them. Making the morphed videos and images of someone by superimposing the face of that person on some others body by the use of artificial intelligence as a real image or video of that person is called deep fake1. They seem real and it is nearly impossible to ascertain that whether it is real or fake.
Although video and image editing technologies like photoshop and photo-editors are known to the world but the first ever use of deepfake was heard in the 2017 when the social media site Reddit claimed that it used AI to superimposing the face of other person in the face of other in the original videos. 2 Sometimes these deep fakes are used for the fun, education, criminal forensics, humour and entertainment. But sometimes these are used for spreading misinformation, pornography, identity theft, forgery, sabotage of elections or for other cybercrimes. These are also used for denigrating and dis-reputing others. Because it simulates as real so it spreads false news and defames others as people easily believe it.
Now deepfakes can easily be created by even the non-skilled persons as there are a lot of online tools are available which does not require any special knowledge or skill for creating such content. Even a layman can morph the image and audio-video clip and spread the false information. Now the tools of making deepfakes are easily accessible in cheaper, easier and faster manner. Social media platforms are flooded with deep fake videos and images. Deep fake video of actress Rashmika Mandhana goes viral on social media, this alarmed the people and law enforcement agencies. Indian government issued advisories to the social media platforms for regulating such types of contents. Prime Minister also raised the concern about deep fakes and called it a “new age Sankat (problem)”. 3 IT Minister Ashwini Vaishnav said “a new crisis is 1 Alanazi, S. , Asif, S.
Exploring deepfake technology: creation, consequences and countermeasures. Hum. -Intell. Syst. Integr. (2024). https://doi. org/10. 1007/s42454-024-00054-8
2 Gabe Regan: AI Explained, A Brief History of Deepfakes (Jun. 1, 2024) /
https://www.realitydefender.com/blog/history-of-deepfakes 3 ANI, "Looming threat of deep fakes in digital media a big concern": PM Modi urges media to educate people on AI, ANI News (Nov.17, 2024)
emerging due to deep fakes” and that “there is a very big section of society which does not have a parallel verification system” to tackle this issue. Deepfake have a wider range of applications. Although it is developed for the good purposes but today it is used mostly for spreading fake news and to be used maliciously. Unethical and immoral use of deepfake is very harmful for the society. Mostly deep fakes are obscene, vulgar and in the form of pornographies. These contents potentially damage and denigrate a person’s reputation. They are also offence against the modesty of the women as they mostly target the female celebrities. Porn sites are full of deep fake contents of many reputed celebrities. People worldwide are victims of these things. Even digitally advanced country South Korea is facing the deep fake crisis.
There the sexually explicit deepfakes of several female graduates of Seoul National University went viral by their college mates. More than 380 people got arrested for the cases of deepfakes and out of which 80% are teenagers. 4 However, there are some good uses of deepfake technology. As it is used for regenerating the voices and images of dead artists in entertainment industry. It is also used in the medical and health industries for treatments of memory loss and others. It has widespread use in the commerce and advertisement industries as well. Meaning of deepfake Deepfake: a combination of deep learning and fake contents, is a process where a fake content is created by the swapping of faces in a video or an image with the use of machine language and artificial intelligence. It is a kind of photoshopped content but created by the use of artificial intelligence.
5 Deepfake is not defined in any Indian legislation but according to European Union's Artificial Intelligence Act [EU AI Act] it is an, “AI-generated or manipulated image, audio or video https://www. aninews. in/news/national/general-news/looming-threat-of-deep-fakes-in-digital-media-a-bigconcern-pm-modi-urges-media-to-educate-people-on-ai20231117140918/ 4 Korea JoongAng Daily, Adolescents 80% of 474 suspects arrested for deepfake porn this year, (Oct. 16, 2024) https://koreajoongangdaily. joins. com/news/2024-10-16/national/socialAffairs/Adolescents-80-of-474-suspectsarrested-for-deepfake-porn-this-year-/2156145
5 Mahmud, Bahar & Sharmin, Afsana. (2020). Deep Insights of Deepfake Technology: A Review.
content that resembles existing persons, objects, places or other entities or events and would falsely appear to a person to be authentic or truthful. ”6 According to Merriam Webster dictionary “a deepfake is an image, or a video or audio recording, that has been edited using an algorithm to replace the person in the original with someone else (especially a public figure) in a way that makes it look authentic”. Fake content is not new to the world but making of fake content with the help of artificial intelligence and algorithms makes it deepfake. Using algorithms the faces of some other person is swapped or superimposed upon the face of original video and making is as the non-morphed content falls under the deepfake. It is different from the normal fake contents as it is highly difficult to differentiate as to whether it is real or fake.
Why it is necessary to regulate Generally, deepfakes are harmless when it is used for a bona-fide purpose like entertainment medical etc. But it becomes a crime when it used for porn, spreading fake news, forgery, to defame, identity theft etc. As it is extremely difficult to differentiate as to whether it is real or morphed, people easily believe it and they frame an opinion on the basis of this. Many researches suggests that deepfakes are purposefully use for spreading false information, defaming others and for spreading hate and fear. So, absence of safeguard and regulations against misuse of deepfakes may lead to grave danger to fundamental rights of individuals and also may create a threat to the society and democracy. Soft target of the deepfake is the women as most of deepfakes are the morphed images or videos of the females.
The porn sites are full of deep fake videos of well-known actresses and personalities. One report suggests that more than 96 percent of deepfakes are in the form of the pornography and their targets are well known celebrities mostly females. 7 There was one software which make the nude image of any person in a single click, it was widely used to harass the women. Deepfakes are also widely used for harassing the world leaders, deepfake videos of world leaders in the form of pornographies are widely spread over the internet to
6 Article 3(44)(b)(l) of the European Union Artificial Intelligence Act (EU AI Act)
7 Ajder, H., Patrini, G., Cavalli, F., & Cullen, L. (2019, September). The State of Deepfakes: Landscape, Threats, and Impact.
harass them or to defame them for manipulating the elections. Therefore, it is not a local threat but a severe global problem and it is extremely necessary to regulate and to avoid its harms. Existing Legislations on deepfake Indian legal framework lacks specific measures to address this burning issue. There is no such specific legislation which particularly deals with the deep fakes. However, Bhartiya Nyaya Sanhita, Information Technology Act, Prevention of children from sexual offences act deals with the deep fakes to an extent. They explicitly and indirectly prohibit the deepfakes. As these laws prohibits the defamation, identity theft, hate speech, mala-fide practices in elections, porn, sexually explicit contents etc...
Some measures also regulate the social media platforms to ensure that sharing or spreading of misinformation, fake news, hate speech and threats to national security shall be prohibited. Section 66E of IT Act prohibits the violation of privacy of a person by capture, publication or transmission of a person’s images in media using deepfakes. 8 Section 66D of the IT Act prohibits the impersonating someone’s voice by using communication devices or computer resources and cheating by that. It provides punishment of three years imprisonment and fine up to one lakh rupees. Section 67A of IT Act deals with the punishment for publication or transmission of sexually explicit material in electronic form. In the first conviction it provides the punishment of imprisonment up to 5 years and fine up to ten lakh rupees.
However, in the subsequent conviction it provided the harsher punishment that is imprisonment up to seven years and fine up to ten lakh rupees. Section 67 of IT Act provides the punishment for the transmission and publication of obscene materials in electronic form. Punishment provided for this in the first conviction is the imprisonment up to three years and fine up to five lakh rupees, while in the second and subsequent conviction is the imprisonment for a term which may extend to five years and also with fine which may extend to Ten Lakh Rupees. The Digital Personal Data Protection Act, 2023 (DPDPA) of India indirectly addresses
8 The Information Technology Act, 2000 (Act 21 of 2000)
concerns related to deepfakes by focusing on the protection and lawful processing of personal data. Although it does not explicitly mention deepfakes, the Act provides a framework that could be leveraged to address the misuse of personal data, including the unauthorized creation and distribution of deepfakes. The Bhartiya Nyaya Sanhita, 2023 (Indian Penal Code Reforms) indirectly addresses issues related to deepfakes by incorporating provisions to combat cybercrimes, defamation, and offenses against personal privacy. While it doesn’t explicitly mention deepfake technology, several sections can be interpreted to deal with crimes involving deepfakes.
Here is how the Bhartiya Nyaya Sanhita, 2023 addresses the legal challenges posed by deepfakes: Section 74 and 77 (Voyeurism and Harassment of Women): Punishes creating or disseminating content that depicts a woman in a compromising position without her consent. This can be applied to non-consensual pornographic deepfakes. Section 356 (Defamation): Deals with false and defamatory content created using deepfakes to harm a person’s reputation. Section 294 (Obscenity): Prohibits the creation, possession, or distribution of obscene material, including deepfakes. Section 318 (Cheating): Applies to cases where deepfakes are used to mislead individuals or commit fraud. Section 196 (Promoting Disharmony): Punishes those who create or share deepfakes intending to incite enmity or disrupt public peace. Judicial Trends on deepfake • Justice Puttaswamy v.
Union of India9 : Privacy violations through deepfakes can lead to criminal charges under this provision.
9 Justice K.S. Puttaswamy (Retd) Vs Union of India (2017) 10 SCC 1
Rajat Sharma v. X Corp 10 : Rajat Sharma, Editor-in-Chief of India TV, discovered deepfake videos circulating on social media platforms, particularly Facebook, where his likeness and voice were manipulated to promote spurious medicines. The court directed Meta Platforms Inc. to promptly block or remove all deepfake videos featuring Sharma’s manipulated likeness. Meta was ordered to provide details of the users who posted the infringing content, including names, addresses, phone numbers, and email addresses. The Department of Telecommunications and the Ministry of Electronics and Information Technology were instructed to ensure that internet service providers block such content and prevent further dissemination. • Amitabh Bachchan vs.
Rajat Negi11: In this case, where the voice and image of renowned Bollywood actor Mr Amitabh Bachchan was used without his consent for commercial purposes—legal action was taken against those responsible, and takedown notices were issued to address the misuse of Mr. Bachchan's identity • Anil Kapoor v. Simply Life India & Ors12 : Renowned actor Anil Kapoor sought legal action to prevent the misuse of his name, voice, appearance, and other personality traits through deepfake technology, which was being used without his consent. The court issued an ex-parte verdict restraining the defendants from using Kapoor’s persona, including his name, voice, and likeness, without authorization. This judgment underscored the importance of safeguarding individuals’ personality rights against unauthorized exploitation through deepfake. • Chaitanya Rohilla v.
Union of India13 : Advocate Chaitanya Rohilla filed a Public Interest Litigation (PIL) highlighting the lack of regulations governing artificial intelligence and deepfake technologies in India, emphasizing potential privacy violations and the spread of misinformation. The court sought the Centre’s response on the need for regulatory frameworks to address the challenges posed by AI and deepfake technologies. In November 2024, the Ministry of Electronics and Information Technology informed the court of its decision to establish a dedicated committee to examine deepfake issues and provide recommendations for regulation. The committee
10 Rajat Sharma v. X Corp, (Formerly Twitter), 2024 SCC OnLine Del 4399
11 Amitabh Bachchan v. Rajat Nagi, 2022 SCC OnLine Del 4110
12 Mr Anil Kapoor v. Simply Life India and Ors, CS(COMM) 652/2023
13 Chaitanya Rohilla v. Union of India, W.P.(C) No.677/2021
was directed to consult with various stakeholders, including internet intermediaries, telecom service providers, victims of deepfakes, and platforms deploying deepfake technology. The committee was given three months to submit its report, with the next court hearing scheduled for March 25, 2025. Legislations of other developed countries Deepfake is a worldwide problem and even the developed countries are also suffering with a range of problems arising because of deepfakes. Even the deepfakes of Pope Francis and renowned actress Taylor Swift went viral on the social media, which causes a lot of hue and cry over the world. Whole of the world feels the necessity to regulate the deepfakes. South Korea declared the deepfake emergency due to continuous increase of the deepfake offences.
Countries around the world made the different legal mechanisms for regulating and to deal with widespread cases of deepfake. United Kingdom: There is no exclusive legislation in UK dealing with deepfakes, but now it has covered under the Online Safety Act14. The Online Safety Act is a significant legislative measure enacted by the United Kingdom to enhance digital safety. Its primary objective is to establish a regulatory framework that addresses illegal and harmful content online, thereby ensuring a safer digital environment for users, particularly children. The Online Safety Act represents a significant step by the UK government to create a safer online environment. Its success will depend on effective enforcement by regulatory bodies like Ofcom and the willingness of online platforms to comply with the new legal requirements.
Ongoing discussions and potential legislative amendments indicate that online safety remains a dynamic and evolving issue, requiring continuous attention to balance user protection with freedoms in the digital space. United States: Although there is no federal regulation in United States for dealing with the deepfakes but some states have their own laws dealing with this subject. California passes the law in 2019 to make the distribution of the deepfakes of political candidates illegal. It also bans the pornographic deepfakes made without consent. Texas and New York also have laws against deepfakes, it prohibits the dissemination and publication of the sexually explicit deepfakes. Virginia also penalised the pornographic deepfakes by legislation in 2019. In 2019, a federal
14 The Online Safety Act 2023
bill i. e. Deep Fake Accountability Act was introduced and reintroduced in 2023 in US. It is under consideration in Congress. It requires the deepfake creators to disclose the purpose of deepfake creations and to regulate the circulation of deep fake during the elections or which are made for the purpose of deceiving others. It also penalises the illegal creations of deepfakes and using it for the unlawful purposes. This bill also provides for the creation of a task force for the purpose of mitigating the effect of deepfakes on national security threats. Canada: Canada currently lacks specific legislation addressing deepfakes, existing criminal and civil laws provide avenues to address certain harmful uses. Ongoing legislative efforts, such as Bill C-6315, indicate a move towards more comprehensive regulation in response to the challenges posed by deepfake technology.
This bill seeks to establish a framework holding online platforms accountable for harmful content, including those involving deepfakes. If enacted, it would impose obligations on platforms to monitor and remove such content. Canada is fighting against this evil in three stages i. e. prevention, detection and response. Canada is primarily focusing on the prevention of deepfakes. For preventing, it is making the people aware about deepfake technologies, telling them the consequences of this evil and also developing the prevention technologies. Government has made a lot of investments for the research and developments to deepfake detection technologies. Government is also eagerly working for the new competent legislation for the responding against this evil by making it illegal and preventing it. Existing law bans the non-consensual publication of intimate pictures.
European Union: EU has recently passed the Artificial Intelligence Act16 for regulating the deepfake contents in multiple ways. It provides the Transparency, Classification, Accountability and Prohibitions techniques for regulating. The creators of deepfakes have to disclose that they are creating the deepfake contents and also have to disclose the techniques used for creation of such contents. This Act also classifies the deepfake contents on basis of risk caused by them. Contents which are defamatory and of political manipulation falls under the higher risk deepfake and they are provided with stricter punishments. Creators and distributors must have to maintain the data and records of process use for the creation of deep fakes. The ACT also prohibits the malicious utilisation of deepfake contents.
15 Online Harms Act 2024
16 The Artificial Act (AI Act) 2024
China: China has made Provisions on the Administration of Deep Synthesis of Internetbased Information Services. This enforces certain responsibilities upon the deep synthesis service providers which includes the protection of personal information and data security, transparency, content management and labelling, safety on technical security. Deep synthesis service providers must take necessary steps for the protection of personal data according to the existing legislations. They must establish mechanisms for recognising false and potentially harmful information to ensure the transparency and data management. They have to maintain mechanisms for dispel and removal of deepfakes and misinformation. The law requires the periodical review of the algorithms by the service providers and to conduct security assessments.
South Korea: It has passed the law to criminalise the creation, possession and dissemination of sexually explicit deepfake materials. Under the Sexual Violence Prevention and Victims Protection Act17 it provides a huge penalty for creating sexually explicit deepfakes with sentence up to 5 years of imprisonment and fine of about $ 38000. South Korean National Assembly passed a new bill which provides anyone who purchase save or watches any sexually explicit deepfake materials may be subject to imprisonment up to 3 years and fine up to $22,600. Legal challenges Although there are laws which regulates the deepfake but there are certain legal challenges which needs to be overcome for the proper regulations of deepfakes and to mitigate the negative effects of deepfake.
These challenges are as follows: • Absence of any specific legislation: IT Act and BNS deals with the deepfake but it requires a specific law which directly criminalize the creation and dissemination of deepfakes for the purpose of defamation, pornography, misinformation etc. • Strict implementation of privacy laws: the perpetrators breach the privacy of victims, steals their photos and videos and create deepfake content s of them. So, it is necessary to strictly implement the privacy laws like Digital Data Protection Act. • Jurisdictional Challenges: Some of the deepfake creators and distributer are sitting at
17 The Sexual Violence Prevention and Victims Protection Act 2010
the cross borders and committing the offences. They are outside the reach of law enforcement agencies. • Lack of awareness among law enforcement: Law enforcement agencies may not be adequately trained to identify or deal with deepfake-related crimes. So, it requires the proper training for the law enforcement to identify and investigate deepfakes and to equip them with advanced tools for forensic analysis. • Digital unawareness: The people are digitally unaware and they do not cross check that he video or image they are watching is real or fake. • Technological evolutions: The technology behind deepfakes is advancing rapidly, making it difficult for laws to keep pace. So, it requires to encourage the collaboration between legal experts and technologists to ensure adaptive and forward-looking legal frameworks.
Conclusion and Suggestions Though there are various provisions of different legislations deals with the various aspects of deepfakes but there are no consolidated legal framework for the regulation of deepfakes and existing legislations are capable enough to tackle with this emerging global evil. Government should frame a unified law for regulating deepfakes and to mitigate its possible effects. It should make provisions for strict enforcement and affirmative action for mitigating the harms caused by deepfakes. There should be a nodal officer for receiving the complaints regarding deepfake contents. As deepfakes are global issue so government should also make provisions for the international cooperation and collaborations to check the deepfakes. Government should adopt the four approaches for avoiding the significant threats caused by deepfakes to various aspects of society.
It should be prohibition, punitive, regulatory and spreading awareness. Under the first approach it should make provisions for restricting the public access to the applications, software, platforms and websites which helps in the creation of deepfakes. Under the second approach it should ensure the liability and provide punishments to the individual and the organisations for the creation and dissemination of deepfake contents for mala-fide purposes. Under the third approach it should issue guidelines and advisories to the various intermediaries and social media platforms for take measures for the detection of
deepfake contents, immediate actions against the complaints made by any individual and forthwith removal of such contents from their platforms. Under the fourth approach it should make provisions for the spreading awareness in the people about such contents so that they could cross verify the truthfulness of any such contents before believing it blindly.