India’s Deepfake Rules: The New Compliance Challenge for Big Tech and AI Startups

 

REGULATION & BUSINESS  |  DEEP DIVE

India’s 2026 IT Rules amendments bring realistic synthetic media into a stricter compliance framework. For technology platforms and AI businesses, the challenge is balancing user protection, operational demands and innovation.

KEY FACTS

• Notification: 10 February 2026 (G.S.R. 120(E)).

• Effective date: 20 February 2026.

• Core focus: realistic synthetic audio-visual content, including qualifying deepfakes, with platform due-diligence obligations, labelling and traceability requirements.

• Important distinction: the three-hour removal timeline applies to unlawful information following a valid court order or reasoned government intimation; it is not a blanket rule that every AI-generated post must be removed within three hours.

The news behind the headline

India’s synthetic-media rules are not a brand-new announcement dated 22 October. The Ministry of Electronics and Information Technology notified the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Amendment Rules, 2026 on 10 February, with the amendments taking effect on 20 February. Government statements in March and later updates have continued to describe the framework and related advisories. The editorial story, therefore, is about what the rules mean for business—and whether any later announcement changes the obligations—not a claim that the rules were passed today.

What counts as synthetic media?

The amended framework defines “synthetically generated information” (SGI) around audio, visual or audio-visual material created, generated, modified or altered using a computer resource so that it appears real or authentic and portrays a person or event in a way likely to be perceived as indistinguishable from reality. Depending on the facts, this can include deepfake video, realistic AI-generated images and voice cloning.

The definition is not intended to sweep in every routine use of software. The Ministry’s explanatory FAQ describes exclusions for good-faith editing and enhancement that does not materially distort meaning, document or presentation preparation that does not create a false document, and accessibility or quality improvements that do not manipulate material parts of the underlying content. The legal test and the specific facts matter.

What platforms must do

The amendments strengthen intermediaries’ due-diligence obligations. Among other things, platforms must use reasonable and appropriate technical measures to address unlawful synthetic content; clearly label permissible synthetic material and support traceability through associated metadata; and take steps to inform users of relevant restrictions and legal consequences. The framework specifically addresses risks including impersonation, misleading content, child sexual exploitation material and non-consensual intimate imagery.

The rules also shorten certain response timelines. Government explanations state that unlawful information must be removed within three hours after a qualifying court order or reasoned intimation from the appropriate government or its agency. This should not be described as an automatic three-hour takedown clock for every complaint, nor as a requirement to remove all AI-generated content.

The compliance challenge for Big Tech

Large social platforms will need dependable workflows across detection, labelling, metadata, user reporting, escalation and legal review. Automated detection can help at scale, but synthetic media changes quickly and tools can produce false positives and false negatives. A platform must be able to respond rapidly without treating every edited image, parody, accessibility feature or clearly fictional work as a harmful deepfake.

The commercial burden will vary by platform size, product design, user volume and the content it hosts. Rather than assume a uniform cost, businesses should assess staffing, tooling, vendor contracts, incident records, escalation procedures and the ability to preserve relevant metadata.

What AI startups, brands and publishers should consider

AI startups should map where their products generate or modify realistic audio-visual material, identify which legal role they perform, and review product labels, provenance information, user terms and abuse-reporting channels. The rules do not mean that every developer or every AI tool has identical obligations; applicability depends on the service, role and relevant provisions.

Brands and agencies should review campaigns that use synthetic presenters, cloned voices or realistic AI-generated scenes, especially where a real person could be impersonated or audiences could be misled. Publishers should strengthen verification and disclosure practices, preserve source records and avoid presenting synthetic material as documentary evidence without appropriate checks.

Innovation versus accountability

The policy objective is to make harmful deception harder while preserving legitimate uses of generative AI. That balance will depend on implementation: clear labels and traceability can help users assess content, while vague processes or overly aggressive filtering can suppress lawful material. Businesses should monitor official clarifications and relevant legal developments rather than rely on summaries alone.

A practical checklist for businesses

☐ Map where realistic synthetic audio, images or video enter your products and workflows.

☐ Define who owns detection, labelling, metadata, complaints and urgent legal escalations.

☐ Document how you respond to valid orders and government directions, including the applicable deadlines.

☐ Review user disclosures, vendor contracts, retention practices and incident-response procedures.

☐ Train marketing, editorial, trust-and-safety and customer-support teams on impersonation and synthetic-media risks.

☐ Seek qualified legal advice for obligations specific to your role, product and content.

BIZ MASALA TAKEAWAY

The rules turn synthetic-media governance into an operational business issue—not just a content-policy debate. Companies should move from ad hoc deepfake responses to documented systems for labelling, verification, escalation and legal compliance, while protecting legitimate AI use.

Verification note

As checked for this article on 22 October 2026, official government material confirms the 10 February notification and 20 February commencement date, and also records subsequent government advisories and explanatory updates. We found no basis in the checked official sources to describe the amendments themselves as newly passed on 22 October. The story is therefore framed as a business explainer on the existing framework.

Sources & further reading

Ministry of Electronics and Information Technology — FAQ on the 2026 Amendment Rules (including commencement date and definition of SGI)
https://www.meity.gov.in/static/uploads/2025/10/065b6deb585441b5ccdf8be42502a49c.pdf

Press Information Bureau — Government has issued multiple advisories emphasizing due diligence obligations (25 March 2026)
https://www.pib.gov.in/PressReleasePage.aspx?PRID=2245053&lang=1&reg=1

Press Information Bureau — Government strengthens regulatory framework to address AI-generated deepfakes (6 August 2026)
https://www.pib.gov.in/PressReleasePage.aspx?PRID=2295500&lang=1&reg=1

Official MeitY amendments and policy documents page
https://www.meity.gov.in/documents/act-and-policies/amendments-to-the-information-technologyintermediary-guidelines-and-digital-media-ethics-code-rules-2021-it-rules-2021-IjN4QjMtQWa?pageTitle=Amendments-to-the-Information-Technology-

BIZ MASALA

Business. Markets. People. Insight.

Comments

Post a Comment

Popular posts from this blog

THE GREAT INDIAN CINEMA RESET

Image
  BIZ MASALA | BUSINESS. MARKETS. PEOPLE. INSIGHT. From Bollywood's dominance to the pan-India blockbuster: how regional cinema, streaming and changing audience behaviour rewrote the business of Indian films By Murthy ASR | Biz Masala | 17 October 2026 Hero artwork created for this story; conceptual illustration, not a documentary photograph. The opening scene: A business no longer controlled by one city For decades, Mumbai occupied the commanding heights of Indian cinema. Hindi films enjoyed cultural and commercial reach, while Telugu, Tamil, Malayalam and Kannada cinema had powerful home markets but more limited national distribution. S. S. Rajamouli's Baahubali: The Beginning (2015) helped demonstrate that a film rooted in one linguistic industry could travel nationally when storytelling, scale, marketing and distribution worked together. Baahubali 2 and later successes from Telugu and Kannada cinema reinforced the lesson. A decade later, dubbed releases are ...
Read more

SINGARENI COAL PUZZLE : THE CONTRACT QUESTION

Image
  What the CAG found in Singareni’s coal operations—and what remains to be answered By Murthy ASR   |   Biz Masala Investigative Series The audit trail is public. The follow-through is the next question. Singareni Collieries Company Limited (SCCL) is jointly owned by the Governments of Telangana (51%) and India (49%). It is not simply a mining company: its coal feeds power generation and industry, and its operating decisions have consequences for workers, consumers and public finances. A subject-specific compliance audit by the Comptroller and Auditor General of India (CAG), covering overburden removal, coal transportation and stowing-material contracts during 2018–19 to 2022–23, put several contract-design and operational questions on the public record. The report was tabled in the Telangana legislature on 30 March 2026. Its findings are historical and specific to the contracts and periods examined; they do not, by themselves, establish the cause of later pro...
Read more

WHO TESTS THE MEDICINES INDIA MAKES?

Image
  BIZ MASALA SIGNATURE STORY Inside the business of pharmaceutical testing laboratories — the money, the market and the science behind a medicine’s quality By Murthy ASR   |   18 October 2026 Illustrative editorial image: pharmaceutical analytical testing laboratory. A medicine may be manufactured on a modern production line, packed in a clean facility and shipped across the world. But before a batch can be trusted, its quality has to be demonstrated. Behind that assurance is a less visible industry: laboratories that test ingredients, finished formulations, impurities, stability and, where relevant, biological characteristics. For India, this is more than a technical service. The country’s pharmaceutical manufacturers supply domestic patients and regulated markets overseas. Their products must meet specifications, documentation requirements and regulatory expectations that vary by medicine and destination. Testing laboratories help generate the evidence on ...
Read more