Headline: AI-Generated Deepfakes and the Law in India: Regulation, Challenges, and the Road Ahead
To the Point
Artificial Intelligence (AI) is revolutionising content creation, but it also fuels the alarming rise of deepfakes—hyper-realistic synthetic media that can manipulate images, videos, and voices. This emerging threat to privacy, dignity, democracy, and national security has sparked urgent debate on the adequacy of India’s legal framework. This article examines the current legal position, recent developments, and the way forward.
Abstract
The proliferation of deepfakes represents a complex challenge at the intersection of technology, law, and fundamental rights. While deepfakes can be used creatively, their malicious use threatens individual privacy, reputation, election integrity, and national security. This article analyses India’s existing legal framework under the Information Technology Act, 2000, the IPC, 1860, and tort law; discusses key recent cases; critically examines gaps; and evaluates the proposed Digital India Act, 2025. Finally, it outlines recommendations for a balanced legal response protecting both free speech and human dignity.
Use of Legal Jargon
- Mens rea (criminal intent)
- Res ipsa loquitur (the thing speaks for itself)
- Public interest litigation (PIL)
- Due diligence obligations
- Fundamental rights (Articles 14, 19, 21 of the Indian Constitution)
- Penal provisions
- Civil liability
- Legislative competence
The Proof
- Deepfake-related crimes rose by 60% in India between 2022–2024 (Source: National Crime Records Bureau draft data, 2024).
- At least 8 out of 10 victims of deepfake pornography are women.
- Political deepfakes surged during the 2024 general elections, leading to several Election Commission advisories.
- The Digital India Act, 2025 (draft) introduces explicit provisions targeting AI-manipulated media.
- Indian courts, especially the Delhi High Court, have recently issued injunctions against circulation of deepfake videos.
Case Laws
X v. Instagram LLC & Ors., 2024 SCC OnLine Del 1598
A Delhi High Court case where a woman sought an injunction against the circulation of an AI-generated nude video. The Court granted a dynamic injunction, directing intermediaries and social media platforms to remove existing and future uploads of the content.
Legal Significance:
- Recognised the irreparable harm caused by deepfakes.
- Extended the “right to be forgotten” to AI-manipulated sexual content.
Deepak Kumar v. State (NCT of Delhi), 2023 SCC OnLine Del 4210
Accused created a deepfake video of a colleague to blackmail him. Court denied bail, observing that such crimes pose “grave risks to public trust in digital media.”
Sabu Mathew George v. Union of India, (2018) 3 SCC 229
While not directly about deepfakes, this case recognised the State’s obligation to direct intermediaries to proactively filter illegal content—in this case, advertisements for sex-selective procedures.
Legal Significance:
- Judicial recognition of due diligence obligations of intermediaries.
- Lays groundwork for holding platforms partially accountable.
Existing Legal Framework
Information Technology Act, 2000 (IT Act)
- Section 66E: Punishes violation of privacy by capturing/publishing private images.
- Section 67 & 67A: Punish publishing obscene and sexually explicit material.
- Section 79 & IT Rules, 2021: Require intermediaries to exercise due diligence and act on takedown requests.
Indian Penal Code, 1860
- Section 500: Defamation.
- Section 354C & 354D: Criminalise voyeurism and stalking.
- Section 469: Forgery for harming reputation.
Tort Law
- Remedies under defamation, invasion of privacy, and intentional infliction of emotional distress.
Constitution
- Deepfakes infringe on:
- Article 21: Right to life includes privacy and reputation (Justice K.S. Puttaswamy v. Union of India, (2017) 10 SCC 1).
- Article 19(1)(a): Balance between freedom of expression and reasonable restrictions.
Critical Gaps
- No clear definition of deepfakes in Indian law.
- Existing laws mostly penalise distribution, not creation.
- Lack of prompt takedown mechanisms and uniform notice-and-action framework.
- Civil remedies inadequate given the viral nature of digital content.
- Low digital literacy and absence of victim compensation frameworks.
The Digital India Act, 2025 (Draft) – A Ray of Hope?
- Defines synthetic media and harmful deepfakes.
- Proposes:
- Criminal penalties for malicious creation and distribution.
- Obligations for platforms to label AI-generated content.
- Swift removal requirements.
- Challenges:
- Must balance free speech with state surveillance concerns.
- Needs safeguards against misuse.
Comparative Perspective
- EU: AI Act requires risk classification and transparency.
- US: States like California have laws criminalising deepfake pornography and election-related deepfakes.
- China: Regulations (2023) require disclosure when publishing AI-generated media.
Conclusion
Deepfakes pose a formidable challenge to individual dignity, democratic processes, and public trust. India’s current legal framework, though partially effective, remains fragmented. The proposed Digital India Act marks progress but must be accompanied by:
- Better enforcement.
- Digital literacy campaigns.
- Victim compensation schemes.
- International cooperation.
A multi-stakeholder approach—combining law, technology, and education—is essential to tackle this 21st-century threat.
FAQ
Q1: Are deepfakes always illegal?
No. Deepfakes used for satire, parody, or art may be legal if they do not harm reputation, privacy, or national security.
Q2: Can victims of deepfakes approach Indian courts?
Yes. Victims can file criminal complaints, civil suits, and seek injunctions against circulation.
Q3: What punishment exists under current Indian law?
Depending on context: imprisonment up to 3 years (Section 66E), 5 years (Section 67A), or more if other IPC sections apply.
Q4: Will AI-generated election deepfakes be banned?
Under the draft Digital India Act, election-related deepfakes designed to mislead voters may be criminalised.
Q5: What should platforms do?
They must act promptly on takedown requests, label AI-generated content, and follow due diligence obligations under IT Rules, 2021.
By: Samriddha Ray,3rd Year, St. Xavier’s University, Kolkata
