ASCI’s New AI Ad Labelling Rules: What Marketers Need to Know

India’s ad watchdog has just forced a question many brands would rather avoid: when does synthetic content cross from harmless production help into hidden persuasion? The Advertising Standards Council of India (ASCI) has released guidelines for labelling Synthetically Generated Content in advertising, and the answer is refreshingly behavioural: label it when it materially shapes what a consumer believes or decides.

The material influence test

The guidelines, effective in three months, do not demand a label every time a marketer touches an AI tool. They mandate disclosure for high-impact uses—virtual influencers, voice replication and synthetic product demos. Routine editing and AI copy generation sit outside the labelling requirement. Unauthorised deepfakes remain strictly prohibited, with or without a disclosure.

That focus on material impact rather than underlying technology is what makes the framework interesting. ASCI is not asking “Did you use AI?” but “Would this change how a reasonable consumer decides?”

Why this is a consumer psychology story

Labels are not just legal furniture. In decision-making terms, disclosure restores a small piece of consumer agency. A virtual influencer that feels human borrows trust from real social proof; a synthetic product demo can overpromise performance. When that influence remains invisible, brands are essentially running a persuasion test without informed consent.

Manisha Kapoor, CEO and Secretary General of ASCI, framed the rule around consumer centrality:

“The use of AI in advertising is evolving rapidly, and our approach must evolve with it while keeping the consumer at the centre. The responsibility remains with advertisers to ensure that the end communication is honest, transparent, and compliant with the ASCI Code.”

For brand builders, that is a useful mental model: AI does not change the ethics of persuasion, it raises the stakes on disclosure.

What to do before the rules bite

Marketers have a three-month window to operationalise this. The practical move is an AI content audit, not a panic review.

• Inventory virtual influencers, voice clones and synthetic demonstrations; these are the clearest label triggers.
• Use clear phrasing where labels are required, such as “Audio/Video created using AI” or “Audio/Video enhanced using AI”, following standard ASCI disclaimer rules.
• Treat unauthorised deepfakes as a non-negotiable line: no label makes them acceptable.

The trust upside

The brands that get this right will not just dodge compliance risk. They will be able to say “AI-assisted, clearly” as a trust cue, not a confession. In a market where synthetic content is growing cheaper by the week, transparency may soon become the scarce input. The three-month countdown is less a deadline and more a chance to make disclosure a habit before it becomes a headline.

Source: ETBrandEquity.com