Published on: 19th April 2026
Authored by: Anagha Suresh Manian
Asian Law College, Noida
ABSTRACT
Artificial intelligence development has progressed rapidly which created new methods for producing and sharing digital content that enabled the invention of deepfake technology. Deepfakes use machine learning techniques to create extremely realistic fake content which includes altered images and videos and audio recordings that show people doing things they never actually did. The technology provides valid uses for entertainment and education and digital accessibility but it creates major problems because it enables people to spread false information and steal personal identities and commit financial crimes and invade other people’s private lives. The legal system in India lacks established laws for deepfake technology which leads to the usage of current laws that include the Information Technology Act 2000[1] and the Digital Personal Data Protection Act 2023[2]. The legislation controls cybercrime and identity theft and data security matters but it was never intended to control synthetic media technologies. This paper investigates the technological basis of deepfakes and their effects on society and the legal developments happening in India which demonstrate the need for impartial regulatory solutions.
Keywords: Deepfakes, Artificial Intelligence, Cyber Law, Privacy Rights, Synthetic Media, Digital Regulation, India
INTRODUCTION
Digital media’s production and consumption have changed sharply due to artificial intelligence. One standout shift emerges from deepfake systems, capable of generating fake visuals and sound clips that seem authentic. These fakes come about not just by code but also through guided training of algorithms. A regular image or clip shifts shape once software reshapes facial movements or speech patterns. Reality twists subtly when such tools go to work. The outcome often indistinguishable from truth challenges how easily people trust what they see.
The term “Deepfake” combines “deep learning” and “fake” to create its meaning[3]. Trained on vast amounts of data, advanced algorithms study how faces move, shift, or sound in real life. Because of this, what people see often feels believable. A person could seem to say something they did not utter, while their features get mapped convincingly onto someone else’s frame. Even though it looks genuine, every detail comes from software-driven changes. What seems true is actually built by machines.
Spreading fast, deepfakes ride on the rise of social networks paired with smart algorithms. The creation of fake videos which show public figures and ordinary citizens has become a frequent occurrence in recent times. The videos are used for two purposes because they function both as entertainment content and as tools for deceiving voters and conducting financial fraud. Hidden behind realistic images, these fakes stir fear about stolen identities and rigged elections. Trouble grows as truth becomes harder to spot online. India presently does not have a dedicated deepfake legislation which results in regulatory authorities depending on current cybersecurity and privacy protection laws such as the Information Technology Act, 2000 and the Digital Personal Data Protection Act, 2023.
- TECHNOLOGICAL FOUNDATIONS AND SOCIETAL IMPLICATIONS OF DEEPFAKES
Deepfakes use Convolutional Neural Networks (CNNs) because this deep neural network model excels at processing visual data. The deepfake technology operates through CNNs, which require training on extensive datasets that include authentic visual content from images and videos. The system learns to duplicate human physical characteristics together with their natural movements, which forms the basis of deepfake technology[4].
Deepfake algorithms use encoder-decoder models or generative adversarial network (GANs). The encoder extracts and compresses facial features into a latent vector. The decoder reconstructs a face. GANs improve visual authenticity through their capacity to introduce new textures and enhance lighting and eliminate visual defects. The GAN discriminator distinguishes real from fake, forcing the generator to improve. The adversarial process makes GANs highly popular among users[5].
Deepfake technology, which has dangerous applications, provides advantages to multiple industries. AI tools assist disabled users by providing them with individualized accessibility solutions. The system provides interactive educational experiences through its historical reenactments and delivers affordable visual effects solutions for artists and filmmakers. Deepfake technology enables journalists and activists to protect their identities while they maintain their right to free speech and private life[6].
Deepfake technology creates multiple digital security threats which include disinformation attacks and financial fraud together with ethical violations and privacy breaches. The system generates realistic fake material which disseminates false information while diminishing trust in authentic sources. The process of authenticating real online content has become more challenging. The distribution of unauthorized dangerous material results in three negative impacts which include damage to reputation and protection of personal information and emotional distress[7].
Deepfakes spread false information through online platforms which makes it hard to verify what is true. Deepfakes create false audio and video content which creates problems for the justice system because it makes court cases untrustworthy and damages the credibility of witnesses. Deepfakes create serious privacy concerns because they allow people to use another person’s facial appearance and voice which leads to reputation destruction. Deepfake scams generate financial losses for companies because they create security threats. The rising use of deepfakes leads to a decline in public confidence and personal privacy together with the loss of trust in online information[8].
LEGAL AND REGULATORY FRAMEWORK GOVERNING DEEPFAKES IN INDIA
The technological complexity of deepfake technology together with different national regulations creates challenges for laws which attempt to control the false information and deception and personal boundary violations enabled by deepfakes. The technology creates dangers which threaten essential human rights therefore regulations must be established through careful consideration. Amendments made recently to India’s digital guidelines reflect persistent conflict in managing these problems.
Digital platforms must remove dangerous online content which includes deepfakes according to new regulations. Social media companies must comply with court or law-enforcement takedown orders within three hours and remove non-consensual explicit content within two hours while resolving user complaints within seven days. The platform accountability system requires these measures to increase its effectiveness while ensuring law enforcement agencies receive better support and digital harmful material gets immediate attention[9].
Deepfakes infringe Article 21[10] privacy by misusing likenesses without consent, which becomes worse through mandatory provenance tracking that enables surveillance of legitimate creators. The Article 21 reputation protection suffers damage from uncorrected defamatory material because the speed of content removal creates a chance for incorrect content removal. Article 19(1)(a)[11] free speech conflicts with the established rules because pre-upload verification and labelling procedures create a risk of censorship against satire and artistic expression and journalistic work. The experts identify “regulatory overreach” as a threat that will force content creators to adopt censorship practices instead of maintaining their artistic freedom[12].
The present legal framework of India gives the nation capacity to tackle various problems which include identity theft and online fraud and personal privacy violations and deepfake technology violations. The current legal system uses two main legislations to handle deepfake technology violations which include the Information Technology Act of 2000 and the Digital Personal Data Protection Act of 2023.
1. Information Technology Act, 2000:
Section 66C[13] criminalizes identity theft through the use of fake identity by people who use another person’s likeness to commit identity theft, punishable by up to 3 years imprisonment and/or ₹1 lakh fine. Section 66D[14] addresses cyber fraud through cheating by personation which uses computer resources for voice-cloned scams, with identical penalties of up to 3 years imprisonment and/or ₹1 lakh fine. Section 67[15] prohibits the distribution and transmission of obscene as well as sexually explicit materials which include non-consensual deepfake pornography; first offense incurs up to 3 years imprisonment and ₹5 lakh fine, which increases for subsequent offenses under Section 67A[16] to 5-7 years and ₹10 lakh. Section 79[17] establishes safe harbour protection for intermediaries who remove unlawful deepfake content after receiving proper notice, but they must conduct due diligence according to IT Rules[18].
2. Digital Personal Data Protection Act, 2023:
Sections 6[19] and 8[20] require organizations to obtain explicit permission before they can process personal data which includes biometric information for facial and voice recognition purpose. Data fiduciaries, including AI developers, commit fundamental violations when they use information without proper authorization. Section 33[21] gives authorities the power to impose fines for organizations which conduct data breaches or process personal information without user consent. People can pursue legal action against privacy violations that occur through deepfakes which operate without proper legal authorization.
3. EMERGING DEEPFAKE CASES IN INDIA
A doctored clip featuring performer Rashmika Mandanna spread fast online, grabbing public attention. This incident set off widespread conversation about how AI methods can be twisted, especially to target women through digital abuse and damage to personal standing. Though not the first, it sharpened focus on weak safeguards in place[22]. Actor Anil Kapoor sought Delhi High Court’s protection against unauthorized AI use of his image and voice. The court issued an interim injunction which prohibited websites and people from using his identity without his permission to protect his personality rights in the digital era[23].
When fake videos featuring actor Akshay Kumar surfaced, judges took a close look at how synthetic media can be misused. Legal responsibility, courts noted, does not vanish just because technology evolves. Profiting from someone else’s image without consent might lead to court actions. Though tools change, holding people answerable remains firm[24]. Cyber fraudsters used a deepfake voice call to deceive a farmer of Roorkee into believing his son had been arrested for a serious crime which resulted in his loss of 6 lakh rupees[25]. The cases demonstrate how urgent the need has become for establishment of clearer legal systems which can effectively deal with synthetic media technology misuse.
CONCLUSION
Deepfake technology enables new creative possibilities in entertainment and education and access services yet it creates major threats which endanger both individual privacy rights and public trust toward digital content. India currently lacks dedicated laws about deepfake technology which forces authorities to use existing regulations from the Information Technology Act, 2000 and the Digital Personal Data Protection Act, 2023. The existing laws lack sufficient authority to control synthetic media because dedicated legal regulations need to be developed for handling upcoming challenges. The creation of a dedicated legal system which protects essential rights while supporting technological progress is necessary to solve the problems created by deepfakes.
[1] Information Technology Act, No. 21 of 2000, India Code (2000)
[2] Digital Personal Data Protection Act, No. 22 of 2023, India Code (2023)
[3] Deepfake, Unit21 Fraud & AML Dictionary, https://www.unit21.ai/fraud-aml-dictionary/deepfake
[4] Understanding Deepfakes, Data Security Council of India – Cybersecurity Centre of Excellence, https://ccoe.dsci.in/blog/understanding-deepfakes
[5] What Is AI Deepfake and How Does It Technically Work?, Milvus AI Quick Reference, https://blog.milvus.io/ai-quick-reference/what-is-ai-deepfake-and-how-does-it-technically-work
[6] Applications of Deepfake Technology: Positives and Dangers, KnowledgeNile, https://www.knowledgenile.com/blogs/applications-of-deepfake-technology-positives-and-dangers
[7] Gabija Stankevičiūtė, What Are Deepfakes? Good vs Bad Use Case Examples, iDenfy Blog (July 31, 2024), https://www.idenfy.com/blog/what-are-deepfakes/
[8] Deepfakes and Their Impact on Society, Mea Digital Evidence Integrity (Aug. 16, 2024), https://www.mea-integrity.com/deepfakes-and-their-impact-on-society/
[9] Vineet Upadhyay & Sumit Kumar Singh, India’s New 3-Hour Deepfake Removal Rule: Experts Urge Strict Compliance, The Indian Express (Feb. 17, 2026), https://indianexpress.com/article/legal-news/indias-new-3-hour-deepfake-removal-rule-experts-urge-strict-compliance-10528122/
[10] INDIA CONST. art. 21
[11] INDIA CONST. art. 19, cl. 1(a)
[12] https://skvlawoffices.com/indias-2026-amendment-to-it-rules-regulation-of-deepfakes-ai-content-and-the-three-hour-takedown-regime/
[13] Information Technology Act, No. 21 of 2000, § 66C, India Code (2000)
[14] Information Technology Act, No. 21 of 2000, § 66D, India Code (2000)
[15] Information Technology Act, No. 21 of 2000, § 67, India Code (2000)
[16] Information Technology Act, No. 21 of 2000, § 67A, India Code (2000)
[17] Information Technology Act, No. 21 of 2000, § 79, India Code (2000)
[18] Krish Arya, Deepfake Regulation India 2025: MeitY’s Comprehensive IT Rules Amendment, Khurana & Khurana Blog (Dec. 16, 2025), https://www.khuranaandkhurana.com/deepfake-regulation-india-2025-meity-s-comprehensive-it-rules-amendment
[19] Digital Personal Data Protection Act, No. 22 of 2023, § 6, India Code (2023)
[20] Digital Personal Data Protection Act, No. 22 of 2023, § 8, India Code (2023)
[21] Digital Personal Data Protection Act, No. 22 of 2023, § 33, India Code (2023)
[22] DC Correspondent, 2024’s Top 10 Victims of AI Deepfakes: From Celebrities to Power Players, Deccan Chronicle (Jan. 16, 2025), https://www.deccanchronicle.com/entertainment/2024s-top-10-victims-of-ai-deepfakes-from-celebrities-to-power-players-1854508
[23] Biranchi Naryan P. Panda & Isha Sharma, Deepfake Technology in India and World: Foreboding and Forbidding, Law & Humanities Quarterly Reviews (2025), https://www.asianinstituteofresearch.org/lhqrarchives/deepfake-technology-in-india-and-world%3A-foreboding-and-forbidding
[24] Maaz Shareef, Someone Made Deepfake Using Your Face or Voice? What Indian Law Says – Lessons from Celebrity Cases, Trademarkia Blog (Feb. 20, 2026), https://www.trademarkia.in/blog/legal-info/deepfake-face-voice-law-india
[25] TOI News Bureau, Deepfake Call Cheats Roorkee Farmer of Rs 6 Lakh, The Times of India (Feb. 14, 2026), https://timesofindia.indiatimes.com/city/dehradun/deepfake-call-cheats-roorkee-farmer-of-rs-6-lakh/articleshow/128124805.cms
