ASCI Unveils Draft Framework for Transparent AI Content Labelling in Advertisements

Chennai: The Advertising Standards Council of India (ASCI) unveiled draft guidelines today establishing standards for the transparent disclosure of synthetically created material in advertising campaigns. The framework harmonizes with the recent amendments to the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, which took effect on February 10, 2026, seeking to maintain transparency while preventing excessive labelling that could overwhelm consumers.

As companies increasingly harness artificial intelligence for advertising initiatives, the proposed guidelines embrace a risk-based framework that prioritizes consumer protection rather than constraining technological deployment. According to the framework, AI-driven advertising constitutes misleading or damaging content only when it generates unrealistic product expectations, targets susceptible demographics, portrays hazardous scenarios, or uses an individual's image or voice without authorization.

Disclosure obligations for AI-generated advertising stem from the potential consumer impact. The guidelines establish a three-tier classification system for AI-generated promotional material based on risk severity.

High-Risk Category (Prohibited): These advertisements breach legal standards, violate rights, present false information, or contravene ASCI regulations—regardless of labelling. Prohibited activities encompass fabricating celebrity endorsements, overstating product capabilities through imagery or descriptions, constructing artificial backdrops with convincing realism, employing deepfakes or unauthorized likenesses, and producing fictional authority personas with identifying traits (such as computer-generated medical professionals endorsing wellness items).

Medium-Risk Category (Mandatory Labelling): These advertisements employ AI in ways that substantially shape purchasing decisions, with non-disclosure constituting consumer deception. Mandatory disclosure applies to synthetically produced influencers and brand representatives, voice or likeness replication (even with consent) for targeted communications, artificially generated visuals depicting product performance that do not match actual functionality, entirely computer-generated realistic scenarios, demonstrations of unavailable products, exaggerated sound design integral to product features, and sponsored product recommendations requiring specific "sponsored by" designation.

Low-Risk Category (No Labelling Required): These advertisements incorporate minor modifications or AI applications with negligible influence on informed consumer decision-making. Exempt from labelling are routine technical adjustments including colour grading, acoustic cleaning, standard blemish correction, and minor illumination modifications that preserve advertising substance; purely ornamental computer-generated elements like decorative backdrops, fictional horizons, and background audio unconnected to product capabilities; conspicuously fantastical imagery such as mythical creatures that audiences recognize as fictional; and administrative applications including copywriting generation, accessibility audio descriptions, and legitimate document preparation.

When disclosure becomes necessary, organizations may employ standardized language such as "Audio/Video created using AI" or "Audio/Video enhanced using AI," provided such labels meaningfully communicate the nature of the content to consumers. Applicable disclaimers must adhere to ASCI's established disclaimer protocols.

The proposed framework invites public examination. ASCI is accepting submissions from business representatives, consumer advocates, and interested parties through June 13, 2026, following which the organization will finalize the guidelines. Comments may be directed to [email protected].