Deepfakes - Legal Implications In The Age Of Synthetic Media
Author: LawSphereX
Category: General Law
Published:
Updated:
DEEPFAKES: LEGAL IMPLICATIONS IN THE AGE OF
SYNTHETIC MEDIA Ms. Sangeeta Chakravarty, Ph.D. Scholar, School of Law Bennett University, Greater Noida, Uttar Pradesh, India
ABSTRACT
“A new crisis is emerging due to deepfakes” – Ashwini Vishnaw, Union Minister of Electronics and Information Technology, The Hindu Deepfakes created in 2017 by a reddit user, is an useful innovation that could be used for various purposes like to resurrect old actors, portray aging, create parodies, but it also poses a threat to society by creating fake images, audios, and videos of individuals leading to violation of several rights and creating distrust. The rapid growth of deepfakes has sparked concerns about their potential misuse. Malicious individuals could exploit them for harmful purposes such as extortion, harassment, and blackmail, posing significant threats to people, businesses, and governments. This research studies the legal implications of deepfakes and the challenges posed to govern it.
This is a Doctrinal method of research, and Data is collected from primary and secondary sources. Keywords: artificial intelligence, misinformation, digital distrust, privacy, regulation.
Introduction
According to Meriam Webster Dictionary Deepfakes means “an image or recording that has been convincingly altered and manipulated to misrepresent someone as doing or saying something that was not actually done or said. ”1 Deepfakes are AI-generated media so realistic that they blur the line between reality and fiction. They can manipulate videos and audio, making it seem like people said or did things they never did. This technology, with significant potential for misuse, is expected to grow into a $38. 5 billion industry by 2030. 2 Deepfakes we first created in 2017 by a Reddit user who used machine learning to graft famous faces onto pornographic videos.
3 According to the Report of European Parliament, EU, ‘Tackling deepfakes in European Policy’ deepfakes are defined as “manipulated or synthetic audio or visual media that seem authentic, and which feature people that appear to say or do something they have never said or done, produced using artificial intelligence techniques, including machine learning and deep learning. ” Despite the absence of a universally accepted definition, deepfakes are generally understood to have three main components: media altered through deep learning, the impersonation of an individual's identity within that altered media, and the presentation of this untrue media as genuine to a casual viewer. Deepfakes adopt the form of face reenactment, face generation, face swapping, and speech synthesis.
The capabilities of this technology are such that the differentiation between a deepfake and authentic content frequently surpasses the discriminatory capacity of unaided human visual perception. Rapid growth of deepfakes has raised concerns about misuse, as malicious actors may use them for extortion, harassment, blackmail, and other harmful
1 Meriam Webster Dictionary,
https://www.merriamwebster.com/dictionary/deepfake#:~:text=A%20deepfake%20is%20an%20image,transparent%3A%20deepfake s%20are%20not%20real 2 Neil Banerjee, ‘Deepfakes in India: Analyzing AI-powered deception and its far-reaching consequences’, SME Futures, (2024) https://smefutures.com/deepfakes-in-india-analysing-ai-powered-deception-and-its-far-reachingconsequences/ 3 Team EMB ‘The impact of Deepfakes: Navigating truth in the digital age’, Technology: EMB Global, (2024) https://blog.emb.global/impact-of-deepfakes/ activities, posing risks to individuals, companies, and governments. Against this backdrop, this research is an attempt to study social and legal challenges posed by deep fakes.
1. Impact of Deepfakes on society
Deepfakes can be both incredibly useful and extremely dangerous. The accessibility of advanced AI tools and the abundance of publicly available data are key drivers in the propagation of deepfakes.4 Deepfakes used to create parodies and resurrect actors, de-aging to show an actor’s past in the movie, protecting the identity of individuals in films with sensitive issues. For example, Deepfake of Shahrukh Khan for Cadbury advertisement. Deepfake content has permeated almost every social media platform. There are three types of deepfakes viz. face swaps, puppet master and Generative Adversarial Networks (hereinafter GANs). GANs function through the interaction of two AI models that compete to create a deepfake output that is as 'accurate' or authentic as possible.
1.1 Misuse
Misinformation and disinformation It is important to understand deepfakes in the digital landscape due to their potential for misinformation and disinformation. Misinformation is false information regardless of intent, while disinformation is intentionally false with the goal of deception. The deepfakes are easily created with readily available tools and can cause both intentional and unintentional harm, particularly in areas like politics and revenge porn. The primary threat stems from people’s tendency to trust what they see, making even unconvincing deepfakes effective in eroding trust in news and social media. 5 Deepfakes of Bollywood actresses including Rashmika Mandanna, Katrina Kaif, Alia Bhatt went viral resulting in identity theft and invasion of privacy.
According to research by Deep trace, an AI firm, the deepfakes are negatively and disproportionately impacting women globally. There is overwhelming prevalence of nonconsensual pornographic content, with a striking majority involving female politicians during Neill Jacobson ‘Deepfakes and their impact on society’, C. P. I.
Open Fox, (2024) https://www. openfox. com/deepfakes-and-their-impact-on-society/ 5 Neil Banerjee, ‘Deepfakes in India: Analysing AI-powered deception and its far-reaching consequences’, SME Futures, (2024) , https://smefutures. com/deepfakes-in-india-analysing-ai-powered-deception-and-its-farreaching-consequences/ elections and actresses leading to identity theft and invasion of privacy. 6 The sources emphasize that the primary application of this technology, in practice has been as a tool for harassment due to the creation of content without individual consent. 7 In 2023, roughly 500,000 video and void deepfakes were shared on social media around the world. 8 million deepfakes by 2025 are expected. 8 Deepfakes can be both incredibly useful and extremely dangerous.
One significant challenge in tackling deep-fakes is the efficiency of social media platforms to detect them due to a lack of technology. To address this issue companies like Facebook, Amazon and Microsoft have initiated the Deepfake Detection Challenge aimed at developing detection methods. Deepfake technology poses a significant and growing threat across various sectors. Based on a survey carried out by Deloitte in 2024, approximately 26% of executives reported deepfake incidents that their organizations had experienced targeting financial and accounting data in the past year, while 50% of all respondents anticipated an increase in the near future. More than 1 in 4 executives revealed that their organizations had experienced one or more deep-fake incidents. Some of the real-world examples of deepfake demonstrate the devastating financial impact.
For example, in early 2024, the British engineering firm Arup fell victim to a deepfake scam that led to a loss exceeding USD $25 million. 9 In February, fraudsters employed deepfake technology to mimic the company’s CFO and other employees during a video call, deceiving a finance staff member into transferring the funds. 10This case highlights the increasing sophistication of digital fraud and weaknesses in identity verification systems. [Hongkong Police] A Mumbai-based doctor, Dr. K. H. Patil, lost Rs. 7. 1 lakh in a stock market scam after being
6 Tom Simonite ‘Most Deepfakes are porn, and they’re multiplying fast’ Wired (7Oct, 2019),
https://www.wired.com/story/most-deepfakes-porn-multiplying-fast/ 7 Reuters ‘Pakistani, Bangladeshi politicians are new targets of deepfake’, 90 percent of videos online are pornographic, (2023) https://www.tribuneindia.com/news/trending/from-rashmika-mandanna-to-bangladeshipolitician-film ed-in-a-bikini-90-per-cent-of-deepfake-videos-online-are-pornographic-571782
8 Neill Jacobson ,2024, Ibid.
9 Kathleen Magramo, ‘British engineering giant Arup revealed as $25 million deepfake scam victim’, CNN Business (17 May 2024), https://edition. cnn. com/2024/05/16/tech/arup-deepfake-scam-loss-hong-kong-intlhnk/index. html 10 Incode, ‘$25 Million stolen using Deepfakes in Hong Kong: Incodes Passive Liveness Technology’, “Shield against Advanced Fraud,” (2024) , https://incode. com/blog/25-million-deepfake-fraud-hong-kong/ misled by a deepfake video of Mukesh Ambani. She encountered the video on Instagram on April 15, where the altered footage showed Ambani endorsing a trading group and urging investment in BCF Academy for high returns. Trusting the content, she reached out to the academy online and invested the amount over May and June, as stated in her FIR. 11 Deepfake technology also poses significant danger to legal proceedings, particularly in courtrooms.
Deepfake content crafted with malicious intent to spread false information could be a significant danger if introduced as evidence in legal proceedings. The ease with which evidence can be manipulated using deepfakes raises concerns about their potential to influence judicial decisions and leave courts astray. The UK child custody case (2020) demonstrates how a fabricated audio recording was used in an attempt to discredit one party as he was making violent threats towards his wife, though it was ultimately identified as fake and dismissed by the court. This case underscores the critical need to scrutinize digital evidence to prevent such manipulation from impacting decisions. The court dismissed the recording as fake as it was found to be a manipulated evidence. 12
2. Governing Laws
India presently lacks specific legislation that explicitly addresses deepfakes, and existing legal frameworks do not provide a clear definition for them. In the absence of exclusive legislation on deepfakes several existing legal frameworks can be applied to certain aspects of deepfakes related issues.
2.1 Information Technology Act, 2000
The Information Technology Act, 2000 includes several provisions to address the misuse of deepfakes. Section 66D punishes cheating by impersonation through digital means, such as using deepfakes to commit fraud. Sections 67 and 67A cover the transmission of obscene material, including explicit deepfakes, holding both creators and distributors legally accountable. Section 66E deals with privacy violations, penalizing those who share private images without consent. Offenders may face fines up to ₹2 lakh, imprisonment for up to three 11 Business Today Desk, “Mumbai- based doctor falls prey to deepfake video to Mukesh Ambani, loses Rs. 7 lack in stock market scam “ Business Today,(2024), https://www. businesstoday.
in/india/story/mumbai-based-doctorfalls-prey-to-deepfake-video-of-mukesh-ambani-loses-rs-7-lakh-in-stock-market-scam-434303-2024-06-22 12 Nishith Desai Associates, ‘Unmasking Deepfakes: Legal, Regulatory and Ethical Considerations’(2024) https://www. nishithdesai. com/fileadmin/user_upload/pdfs/research_Papers/Unmasking-Deepfakes. pdf years, or both. Additionally, the section targets identity theft committed with malicious intent using computer resources. Section 66F addresses cyber terrorism and applies to deepfakes used to spread false information that incites hatred, undermines public trust, or provokes violence against the government. 13
2.2 Intermediary liability
The Information Technology Act offers a "safe harbor" to online intermediaries, protecting them from being held legally responsible for user-generated content, provided they meet certain conditions outlined in Section 79. These conditions require intermediaries to: (i) avoid initiating the transmission of online content, (ii) refrain from selecting or modifying any thirdparty content, and (iii) follow the due diligence guidelines set by the government. However, this immunity is not absolute intermediaries can still be held accountable if they fail to meet these obligations. Additionally, Section 79 allows social media platforms to take down deepfake content upon receiving a complaint or a court directive.
They remain exempt from liability as long as they do not control the transmission process or alter the content, and they continue to act in accordance with the prescribed due diligence norms. 14 The Supreme Court, in the landmark Justice K. S. Puttaswamy case (2017), recognized the right to privacy as a fundamental aspect of the ‘right to life and dignity’ under Article 21 of the Constitution of India. The ruling emphasized "informational privacy," affirming that individuals have the right to control how their personal data is shared. Any unauthorized use or disclosure of such information amounts to a violation of fundamental right. 15 Currently, privacy is regulated under the ‘Information Technology’ Act and ‘Information Technology (Reasonable Security Practices and Procedures and Sensitive Personal Data or Information) Rules, 2011’ (Hereinafter SPDI Rules).
According to SPDI regulations, biometric information is considered sensitive personal data. An individual’s face, voice patterns, fingerprints, eye retinas, and irises are all considered biometrics. Artificial Intelligence (AI)-generated deepfakes would be subject to privacy violations under the aforementioned clauses since they utilize sensitive personal information such as biometric information, voice and facial patterns. Deepfakes made with publicly accessible data, however, 13 Shinu Vig ‘Regulating Deepfakes: An Indian perspective’, Journal of Strategic Security (2024) 17(3)70-93.
14 Ibid. 13
15 SCO,’Fundamental Right to Privacy’, (2017), https://www. scobserver. in/cases/puttaswamy-v-union-of-indiafundamental-right-to-privacy-case-background/; Case No. WP(C) 494/2012 might not be covered by the Digital Personal Data Protection Act, since it does not protect such information. Under Section 353 of Bhartiya Nyaya Sanhita,2023 (BNS) attention is drawn to statements made with intent to cause public mischief, which penalizes individuals who willfully and intentionally disseminate false information, rumors, or reports, including electronic media. It encompasses misinformation designed to provoke public, induce fear or alarm, and promote enmity, hatred, or ill will among differing groups. For instance, the use of deepfakes in propagating false content.
Section 356 of BNS addresses defamation, whereby an individual tarnishes another’s reputation through spoken or written words, intentionally to harm the reputation of the latter, is considered defamatory and punishable under the Act. Deepfakes, by spreading fabricated content that a person would not typically endorse, can be deemed defamatory, as it impacts their standing within society upon viewing such content on various platforms. Section 334(4) of the BNS penalises forgery using electronic record documents that can be interpreted to include deepfakes within its ambit. Section 318 clarifies the act of cheating, whereby an individual deceives another through dishonesty or fraud without their consent, thus committing an offense that is also punishable under Sanhita.
In July 2022, in the State of Kerala deepfake fraud case was reported, where a 73-year-old man named Radhakrishnan lost ₹40,000. He was deceived by a scammer who used AI to impersonate his former colleague on a WhatsApp call, persuading him to transfer funds. After realizing he had been defrauded, Radhakrishnan reported the incident to the police. It was found in the police investigation that the fraudster used social media to gather personal information. The police were able to track the money to an account in Maharashtra and are currently investigating the case. 16
2.3 International Scenario
In June 2022, the European Union introduced “the Strengthened Code of Practice on Disinformation” (hereinafter 2022 Code”). This code requires online platforms utilizing AI systems for content generation or modification, including deepfakes, to adhere to transparency standards. The Digital Services Act (DSA), enacted in the same year, reinforces the 2022 Code through its regulations on online disinformation, thereby providing it with legal authority 16Pranav Dabas ,2024; Supra concerning internet safety and platform responsibility. The DSA envisions empowering users to identify deepfakes by obligating online platforms to offer tools that allow users to flag content they believe has been maliciously manipulated using synthetic media to deceive others. In the US, both federal and state levels of government have taken action to confront the problems associated with deepfakes.
On the federal level, one significant step is the introduction of the ‘Deepfakes Accountability Act,’ 17 “The Deepfakes Accountability Bill” was proposed in 2023 (“Deepfakes Bill”) and it aims to establish a comprehensive framework to regulate deepfake technology and address the potential harm it poses. It includes the establishment of a “Deepfakes Task Force” places obligations on online platforms to maintain minimum technical capabilities to trace the origin of hosted content, provide disclosures, and have an active deepfakes detection system. The Deepfakes Bill outlines both criminal and civil penalties for online platforms that fail to comply with its regulations.
Furthermore, in January 2024, the “No Artificial Intelligence Fake Replicas and Unauthorized Duplications (No AI FRAUD) Bill, 2024” (“No AI Fraud Bill”) was introduced that aims to safeguard individuals’ rights against the harmful applications of artificial intelligence, with deepfakes being a key concern. Several states in the US have proactively addressed the issue of deepfakes through legislation. For example, California has banned the spread of manipulated videos within two months of an election and has made non-consensual deepfake pornography a criminal offense. Similarly, Texas has criminalized the creation and distribution of deepfakes designed to impact election outcomes.
Parallel to these state-level legal actions, major technology companies, including Facebook (which launched the Deepfake Detection Challenge), X, and Google, have implemented stringent policies and initiatives to fight the proliferation of deepfake content. 18
3 Combating Deepfakes
Government Strategies In response to a legal petition by India TV Chairman Rajat Sharma regarding the lack of deepfake regulation, the Delhi High Court issued a notice to the Union Government. In August 2024, the High Court warned that deepfake technology poses a serious threat to society and 17 Pranav Dabas ‘The Impact of Deepfake Technology: Legal Risks and Regulatory Solutions’, Meta legal Advocates, (2024), https://www. metalegal. in/post/the-impact-of-deepfake-technology-legal-risks-andregulatory-solutions 18Pranav Dabas ,2024 ;Ibid stressed that technological solutions are essential to counter AI misuse. Subsequently, the Union Ministry of Electronics and Information Technology (hereinafter MeitY) formed a committee on November 20, 2024, to address deepfake-related issues.
The MeitY)presented a status report to the Delhi High Court on March 24, 2025, addressing concerns about deepfake technology's spread in India. 19 Honorable High Court of Delhi gave following remarks20: “the deepfake technology is going to be a serious menace in society and the government should start action and the antidote for AI would be technology only. ” The report noted that many stakeholders believe that the existing legal mechanisms including the ‘Information Technology Act, 2000’; Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021’; and the ‘Bhartiya Nyaya Sanhita, 2023’, are adequate to address malicious deepfake use, but they require more robust enforcement and better forensic tools.
The Deepfakes Analysis Unit (DAU), working under the Meta-backed Misinformation Combat Alliance (MCA), identified two troubling trends: targeted deepfake attacks on women during state elections and a surge in AI-powered scams after elections. The DAU also pointed out the ongoing difficulty in detecting manipulated audio content and emphasized the need for cooperative detection systems and clear regulations. There is absence of a standardized definition for "deepfake," which complicates regulatory efforts. To combat deepfakes, the report further emphasized the necessity of extensive public awareness programs focused on educating its users on identifying and understanding deepfakes. The significance of developing indigenous data sets and tools for detecting and analysing deepfakes in Indian languages and contexts was emphasized.
The focus should be on curbing malicious actors rather than limiting the innovative use of deep-fake technology. The committee has been examining the concept of "mandatory intermediary compliance," which would establish the responsibility of digital platforms in managing AI-generated content. This approach aims to strike a balance between platform accountability and the protection of free speech.
4. Critical Analysis
There is a need for legal definition of ‘deepfakes’for deepfakes that would enable lawmakers in distinguishing them from other forms of digital alterations, thereby paving way for specific regulatory frameworks. Obligate online intermediaries to immediately observe, identify, and report deepfake actions. Creating or disseminating deepfakes with malicious intent can lead to
19 Chaitanya Rohilla vs. Union of India, HC Delhi, Writ Petition (C) No. 15596 of 2023
20 August ,2024
severe legal consequences, including substantial fines and imprisonment. Guarantee prompt identification and action to reduce damage. Technological Methods to Fight Deepfakes. These measures may assist India in establishing a solid framework to tackle the issues presented.
5. Conclusion
To navigate the challenges posed by rapidly advancing deepfake technology to information integrity and digital trust, a collaborative and timely response is crucial. Tech companies, policymakers, educators, and the public must work together to safeguard society from this AIdriven deception and uphold the integrity of our digital environment. International cooperation and conventions to regulate deepfakes and standardize countermeasures globally offer a potential path forward. Social media platforms have a key role in identifying and flagging potentially harmful deepfake content, removing it, and informing users of its manipulated nature. Ultimately, our ability to ensure privacy and security alongside technological progress and free speech hinges on effectively addressing the threat of deep fakes.
Bibliography
1 Anurag Roushan & Varun Sharma, ‘Government submits status report on deepfakes
AI scams under scrutiny’ India TV, (2025), https://www.indiatvnews.com/news/india/govtdeepfake-report-ai-scams-election-misuse-govt-report-delhi-high-court-2025-03-27-
2. Bhaumik Aaratrika, ‘Regulating deepfakes and generative AI in India)’, The Hindu, (2023),
https://www.thehindu.com/news/national/regulating-deepfakes-generative-ai-in-indiaexplained/article67591640.ece
3. Chaitanya Rohilla vs. Union of India, HC Delhi, Writ Petition (C) No. 15596 of 2023
4. European Parliament Report, ‘Tackling deepfakes in European Policy’ , (2021),
https://www.europarl.europa.eu/RegData/etudes/STUD/2021/690039/EPRS_STU(2021)6 90039_EN.pdf
5. Felipe Romero Moreno, ‘International Review of Law, Computers & Technology’, Vol. 38,
NO. 3, 297–326 (2024), https://doi.org/10.1080/13600869.2024.2324540
6. Fraud Detection, ‘Top 5 Cases of AI Deepfake Fraud From 2024 Exposed’ (2024)
https://incode.com/blog/top-5-cases-of-ai-deepfake-fraud-from-2024-exposed/
7 Indian Cyber Squad, ‘Case Study
Kerala’s First Deepfake Fraud’ (2023), https://www.indiancybersquad.org/post/case-study-kerala-s-first-deepfake-fraud
8 Neil Banerjee, ‘Deepfakes in India
Analyzing AI-powered deception and its far-reaching consequences, SME Futures, (2024), https://smefutures.com/deepfakes-in-india-analysingai-powered-deception-and-its-far-reaching-consequences/
9. Neil Jacobson, ‘Deepfakes and their impact on Society’, CPI Openfox (2024),
https://www.openfox.com/deepfakes-and-their-impact-on-society/
10 Pranav Dabas, ‘The Impact of Deepfake Technology
Legal Risks and Regulatory Solutions, Meta legal Advocates, (2024) https://www.metalegal.in/post/the-impact-of-deepfaketechnology-legal-risks-and-regulatory-solutions
11 Sara H. Jodka, ‘Manipulating reality
the intersection of deepfakes and the law’, Reuters, (2024), https://www.reuters.com/legal/legalindustry/manipulating-reality-intersectiondeepfakes-law-2024-02-01/
12 Shinu Vig, ‘Regulating Deepfakes
An Indian Perspective’, Journal of Strategic Security 17(3), (2024), https://digitalcommons.usf.edu/cgi/viewcontent.cgi?article=2245&context=jss
13 Sidhartha Mishra, ‘AI & Deepfakes
Legal Issues and Challenges in India’, Indian Journal of law and Legal Research, 4(2) ISSN: 2582-8878, (2024)
14 The Financial Express, ‘Explainer
Generative AI and Personal Data’, (2024) https://www.financialexpress.com/business/digital-transformation-explainer-generativeai-and-personal-data-3503632/