Artificial Intelligence (AI) has enabled the creation of highly realistic synthetic content, commonly known as deepfakes, including manipulated audio, video, images, and text. While AI offers immense opportunities, deepfakes pose serious challenges such as identity theft, misinformation, financial fraud, privacy violations, electoral manipulation, and cybercrime.

Recognizing these threats, the Government of India has strengthened its legal and regulatory framework to ensure a safe, trusted, accountable, and open cyberspace, particularly through amendments to the Information Technology Rules in 2026.

What are Deepfakes?

Deepfakes are AI-generated or AI-manipulated digital content that imitates the appearance, voice, or actions of real individuals, making fake content appear authentic.

Major Risks

Identity theft
Online impersonation
Financial fraud
Spread of misinformation and disinformation
Non-consensual intimate imagery
Child sexual exploitation
Damage to reputation
Threats to national security and public order

Existing Legal Framework Against Deepfakes

1. Information Technology Act, 2000

The IT Act provides several legal provisions applicable to AI-generated unlawful content.

Important Sections

Section 43

Penalty and compensation for damage to computer systems.

Section 66

Punishment for computer-related offences.

Section 66C

Identity theft.

Section 66D

Cheating by impersonation using computer resources.

Section 66E

Violation of privacy.

Sections 67 & 67A

Publishing or transmitting obscene or sexually explicit content.

Section 69A

Blocking access to unlawful online information.

Section 79

Removal of unlawful information by intermediaries.

Sections 78 & 80

Investigation powers for police authorities.

2. Bharatiya Nyaya Sanhita (BNS), 2023

The BNS also provides criminal penalties relevant to deepfake-related offences.

Key Provisions

Section 319

Punishment for cheating by personation.

Section 336

Punishment for forgery, including false electronic records.

Section 353

Penalizes false statements, rumours, and misinformation causing public mischief.

Section 111

Covers organised cybercrime involving deepfake content.

Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021

The IT Rules impose due diligence obligations on intermediaries.

Users are prohibited from sharing content that:

Belongs to another person without authorization
Is obscene or pornographic
Invades privacy
Promotes hate or violence
Harms children
Infringes intellectual property rights
Misleads users regarding the origin of messages
Spreads misinformation
Impersonates others through AI
Threatens national security or public order
Violates any applicable law

Intermediaries must clearly inform users about penalties such as content removal, account suspension, or account termination for violating these rules.

Additional Obligations for Significant Social Media Intermediaries (SSMIs)

Platforms with 50 lakh or more registered users in India must:

Assist law enforcement in tracing originators of serious messages.
Deploy automated tools to detect unlawful content.
Publish periodic compliance reports.
Appoint grievance officers and compliance officers in India.
Offer voluntary user verification.
Provide appeals and fair hearing mechanisms.

Grievance Redressal Mechanism

Every intermediary must appoint a Grievance Officer.

If users are dissatisfied, they can appeal before the Grievance Appellate Committee (GAC).

Police investigations are conducted by State and Union Territory law enforcement agencies since Police and Public Order are State subjects under the Constitution.

Major 2026 Amendments to IT Rules

On 10 February 2026, the Government significantly strengthened regulations relating to AI-generated content.

1. Mandatory AI Content Labelling

Platforms must ensure:

Clear labels for AI-generated content.
Traceable metadata.
Easy identification of synthetic content.

2. Stronger User Accountability

Platforms must educate users regarding legal consequences of unlawful AI-generated content.

3. Expanded Protection

The amendments specifically address:

Child sexual abuse material
Non-consensual intimate imagery
AI-based impersonation
Other harmful synthetic content

4. Faster Content Removal

Timeline reduced:

Removal of unlawful content: from 36 hours to 3 hours
General grievance resolution: from 72 hours to 36 hours
Sensitive grievances (nudity/impersonation): from 24 hours to 2 hours

5. Mandatory Technical Measures

Intermediaries must deploy:

Automated detection tools
Other suitable technological mechanisms

to prevent unlawful AI-generated content.

6. Proactive Detection

SSMIs must proactively identify content depicting:

Rape
Child sexual abuse
Previously removed unlawful content

Failure to comply results in loss of Section 79 safe harbour protection, making intermediaries liable under applicable laws.

AI Governance Framework

India follows a balanced and risk-based approach toward AI governance.

The India AI Governance Guidelines, released on 5 November 2025, promote:

Safe AI
Trusted AI
Responsible AI
Proportionate regulation

rather than blanket restrictions.

IndiaAI Mission

The Safe & Trusted AI pillar under the IndiaAI Mission promotes indigenous AI governance, standards, evaluation mechanisms, and responsible deployment.

Approved Deepfake Detection Projects

Saakshya (IIT Jodhpur & IIT Madras)
AI Vishleshak
Real-Time Voice Deepfake Detection System (IIT Kharagpur)

Institutional Mechanisms

Grievance Appellate Committees (GACs)

Provide appellate review against intermediary decisions.

Takedown Notices

Governments may direct intermediaries to remove unlawful content through valid legal processes.

Indian Cyber Crime Coordination Centre (I4C)

Coordinates cybercrime investigations nationwide.

SAHYOG Portal

Provides centralized automated removal notices to intermediaries.

National Cyber Crime Reporting Portal

Citizens can report:

Deepfakes
Cyber fraud
Content misuse

The cybercrime helpline 1930 is also operational.

Police

Investigate offences under applicable laws.

Samanvaya Platform

Supports interstate cybercrime investigations using analytics-based criminal linkages and the Pratibimb module.

Government Advisories

The Government has issued:

Advisory on responsible handling of religious information (9 February 2026)
Advisory against abusive and misleading AI-generated information (16 March 2026)
Standard Operating Procedure (2025) for tackling Non-Consensual Intimate Imagery (NCII)

Public Awareness Measures

Awareness campaigns include:

National Cyber Security Awareness Month
Safer Internet Day
Swachhta Pakhwada
Cyber Jagrookta Diwas

CERT-In also regularly publishes advisories and awareness material on deepfake threats and cybersecurity.

Conclusion

The Government’s strengthened regulatory framework reflects a shift from merely reacting to unlawful AI-generated content toward proactive governance. By mandating AI content labelling, reducing takedown timelines, enhancing intermediary accountability, promoting indigenous AI governance, and strengthening cybercrime coordination, India seeks to balance technological innovation with user safety and legal accountability in the age of artificial intelligence.

Summary: Quick Revision Notes

Key Highlights

AI-generated deepfakes regulated through IT Act, BNS, and IT Rules.
Mandatory AI content labelling introduced.
Traceable metadata required.
Unlawful content removal timeline reduced: 36 hrs → 3 hrs.
Sensitive grievance disposal: 24 hrs → 2 hrs.
Automated AI detection tools made mandatory.
SSMIs required to proactively detect unlawful AI-generated content.
Non-compliance leads to loss of safe harbour under Section 79.
IndiaAI Mission supports indigenous deepfake detection research.
Key institutions: GAC, I4C, SAHYOG Portal, CERT-In, National Cyber Crime Reporting Portal, Samanvaya Platform.

Reference: PIB

AI-Generated Deepfakes: India’s Legal, Institutional, and Policy Response