India Orders Platforms to Label and Rapidly Remove Deepfakes Under New AI Rules

India has issued new directives requiring social media platforms to identify, label, trace, and rapidly remove deepfakes and other AI‑generated content in one of the most consequential regulatory responses to date. 

According to a report by TechCrunch, the rules will take effect on February 20, 2026 and mandate clear labeling of synthetic audio and visual media and tightly compressed takedown timelines for non‑consensual or otherwise prohibited content.

These changes come as part of a broader Indian government effort to combat the spread of manipulated media on platforms with hundreds of millions of users. 

TechCrunch notes that officially flagged deepfakes and legally unlawful AI‑generated material must be removed by platforms such as Meta, YouTube, and X within three hours of a notification, down sharply from the previous 36‑hour window. 

Certain urgent cases, like non‑consensual imagery, will require action within two hours, and compliant labeling and traceability standards are now statutory.

India’s position as one of the world’s largest internet markets means these requirements could influence global moderation practices. Large platforms will likely need to deploy technical systems that can verify synthetic content disclosures, detect deepfakes at scale, and embed traceable metadata into allowed AI content. 

What the Indian Rules Require

The amended IT rules create formal obligations for platforms that host user‑generated audio‑visual content. Under the new framework platforms must ensure clear labeling of synthetic content with embedded provenance data, so users and moderators can see when media has been created or significantly altered by algorithms.

Meanwhile, users must disclose when they upload AI‑generated media, and platforms are expected to deploy verification tools to validate these disclosures.

Certain categories, including deceptive impersonations and non‑consensual intimate content, are explicitly prohibited and must be removed.

There will also be rapid takedown deadlines that bind platforms to remove deepfakes flagged by courts or authorities within hours of notification, a significant compliance challenge in practice.

Regulators and lawmakers say these steps aim to curb misinformation, digital harms, and abuse while reinforcing accountability in online ecosystems. Platforms that fail to meet these requirements risk losing safe‑harbor protections that shield them from certain legal liabilities under Indian law.

Deepfake Abuse and Its Impact on Children

The urgency of these regulatory changes is underlined by recent findings from the United Nations Children’s Fund (UNICEF) highlighting how deepfake technology has been abused to generate harmful and sexually explicit imagery involving minors. 

According to UNICEF‑led research conducted across 11 countries, at least 1.2 million children reported that their real images were manipulated into sexually explicit deepfakes in the past year, a scale equivalent to “one child in a typical classroom” of 25 students in some regions. 

These manipulated images, often created without consent, are a form of child sexual abuse material under international standards and fuel a broader crisis of online harms targeting minors.

UNICEF has called on policymakers to expand protections and criminalize the creation and dissemination of such content, noting that AI‑driven deepfake tools have dramatically lowered the barriers to producing exploitative imagery. The agency emphasizes that this abuse is not just technological but deeply harmful to the dignity and safety of children globally.

The Indian government’s new rules, by demanding proactive labeling and rapid removal of deepfakes and abusive content, reflect a recognition that synthetic media can be weaponized at scale and that moderating it effectively may require both legal obligations and technical safeguards.

The move in India echoes wider conversations about deepfake regulation and industry accountability. TechCrunch previously reported on efforts by AI and digital rights communities calling for stronger anti‑deepfake legislation and on platforms facing pressure to refine their labeling and moderation strategies. 

For example, industry figures have advocated for distinct categories within policy frameworks to better address non‑consensual and exploitative AI‑generated imagery.

Implications for Users and Platforms

For users and creators, India’s new framework means that social media platforms must become more transparent about synthetic media and responsive to violations of digital safety norms within constrained timelines. 

From an operational perspective, platforms will need to bolster automated detection and labeling systems, invest in round‑the‑clock moderation infrastructure, and align compliance mechanisms with a statutory definition of deepfakes and synthetic content.

As deepfake technology continues to mature and proliferate, legal and regulatory efforts like India’s amended IT rules may become reference points globally, especially in markets grappling with rising misinformation and online abuse. The intersecting concerns of privacy, platform accountability, user rights, and child protection will shape ongoing debates about how to govern synthetic media responsibly in a digital society.

---