Author: Ashwina Verma student at Banasthali Vidyapeeth
To the Point
Deepfakes—hyper-realistic AI-generated audio and video fabrications—have redefined the idea of evidence, privacy, and consent. This article examines how Indian law currently addresses deepfakes, the urgent legal vacuum, and what frameworks can be adopted to counteract their harmful potential.
Use of Legal Jargon
Deepfakes fall within a grey area of prima facie criminality, potentially violating provisions under mens rea-based offences, in personam rights, lex loci delicti, and jus in rem obligations such as the right to privacy. They also raise questions under tortious liability, defamation per se, intellectual property infringement, and cyber contraventions.
The Proof
The dangers of deepfakes are not theoretical. In 2023, a manipulated video of an Indian actress was widely circulated, falsely portraying her in a compromising position. Despite the massive reputational harm, legal recourse was delayed and limited. Another example involved a fabricated voice recording mimicking a political leader days before an election, triggering disinformation concerns.
According to a 2024 study by Sensity AI, India ranks among the top five countries targeted by deepfake content, especially affecting women and public figures. However, India’s legislative and judicial response remains largely reactive and scattered.
Abstract
Deepfake technology, developed using deep learning algorithms such as Generative Adversarial Networks (GANs), poses a significant threat to legal systems grounded in truth, consent, and authenticity. This article examines the Indian legal landscape’s preparedness critically to tackle the deepfake related cases. It analyses laws like the Information Technology Act, 2000; the Indian Penal Code, 1860 which is now Bhartiya Nyaya Sanhita 2023; and recent privacy jurisprudence. The article also compares international approaches and suggests a legislative framework for India, emphasizing the balance between free expression and protection from harm.
Case Laws
1. Justice K.S. Puttaswamy v. Union of India (2017) 10 SCC 1
The Supreme Court affirmed that the right to privacy is a fundamental right under Article 21. Deepfakes directly infringe this right when consent is absent in manipulated content.
2. Shreya Singhal v. Union of India (2015) 5 SCC 1
This case struck down Section 66A of the IT Act, reaffirming the primacy of freedom of speech. However, it also recognised that speech causing public disorder or impersonation may be curbed under reasonable restrictions, relevant to deepfake-related misinformation.
3. Khairati Lal v. State (2023 SCC OnLine Del 3849)
A Delhi High Court case where a man was arrested for circulating a deepfake pornographic video. Though booked under Sections 67, 67A of the IT Act and Sections 354 and 509 IPC, the absence of a specific deepfake law delayed prosecution.
4. Catfish Victim v. Unknown (2023) (Madras HC, Unreported)
A victim filed a writ petition after a romantic partner used her face in AI-generated nudes. The Court issued takedown orders but highlighted the absence of preventive mechanisms in Indian law.
Legal Framework Analysis
1. Indian Penal Code, 1860
– Section 465 & 468 (Forgery): May apply where deepfakes are used to impersonate for fraud.
– Section 499 (Defamation): Offers partial relief; however, intent (animus injuriandi) must be proven.
– Section 354C & 509 (Outraging modesty): Can be invoked when deepfakes involve sexual content, especially non-consensual porn.
2. Information Technology Act, 2000
– Section 66D (Impersonation using communication device): Applies to identity misuse.
– Section 67 & 67A (Obscene material in electronic form): Used for explicit deepfakes, yet fails to cover non-sexual deceptive uses.
– Section 69A: Allows for takedown of such content, but lacks proactive detection mandates.
3. Data Protection and Privacy
– The Digital Personal Data Protection Act, 2023 mandates consent for processing personal data. Deepfakes breach this by fabricating identities using biometric data, violating both informational privacy and autonomy.
– The Puttaswamy judgment emphasized informational self-determination, which is at odds with deepfakes created without permission.
Comparative Legal Approaches
– European Union: The proposed AI Act classifies deepfakes as “high-risk” AI, mandating transparency and labeling obligations.
– USA: The DEEPFAKES Accountability Act (proposed) requires watermarks on AI-generated media and penalizes malicious creation and distribution.
– China: The Deep Synthesis Regulation (2023) enforces compulsory labeling and prohibits deepfakes that endanger national security or reputation.
India lags behind in both detection mechanisms and statutory deterrence.
Need for a Specific Law on Deepfakes
A dedicated law or amendment is needed to:
– Define “deepfake” under Indian law;
– Penalize both creation and distribution;
– Shift burden of proof partially in cases involving harm to minors or women;
– Introduce pre-publication takedown mechanisms and AI watermark mandates;
– Provide civil remedies like compensation and injunctions, apart from criminal action.
A balanced statute must harmonize Article 19(1)(a) (freedom of speech) with Article 21 (right to dignity and privacy), as seen in Subramanian Swamy v. Union of India (2016).
Conclusion
Deepfakes are not merely digital distortions; they are legal distortions. While existing laws provide some fragmented protection, the pace and precision of this technology demand a distinct, forward-looking legal response. If left unchecked, deepfakes will erode the very foundation of evidence, consent, and trust in law.
India must recognize the urgency to legislate a Deepfake Regulation Act, rooted in rights-based jurisprudence and digital ethics. The law must not only punish but also prevent.
FAQs
Q1. Are deepfakes currently illegal in India?
Not directly. But they can be prosecuted under IPC and IT Act provisions related to forgery, obscenity, and impersonation.
Q2. What is the punishment for circulating deepfake porn in India?
Under Section 67A of the IT Act, up to 5 years imprisonment and fines. Extra charges can be applied.
Q3. Is consent required before using someone’s face or voice in AI-generated media?
Yes. Using biometric data without consent may violate privacy laws and the Digital Personal Data Protection Act, 2023.
Q4. Can I sue someone for creating a deepfake of me?
Yes. Civil remedies like injunction and damages are possible under tort law and defamation. Criminal proceedings can also be initiated.
Q5. Does the law of India finds AI as a liability?
Presently, liability is on the creator or distributor of the AI-generated content, not the AI itself.
