Labelling the Artificial: ASCI’s New Framework for Synthetically Generated Content in Advertising

Introduction

As artificial intelligence (“AI“) becomes increasingly integrated into advertising, the distinction between authentic representations and computer-generated content is becoming harder to identify. From virtual brand ambassadors to AI-generated product demonstrations, advertisers can create realistic representations of individuals, events and product experiences that may never have existed in reality. While these technologies offer significant creative and commercial opportunities, they also raise concerns regarding consumer deception, authenticity and the reliability of advertising claims.

Against this backdrop, the Advertising Standards Council of India (“ASCI“), on September 29, 2026, issued the Guidelines for Responsible Labelling of Synthetically Generated Content in Advertising (“Guidelines“), following consultation on its draft guidelines issued in May 2026. Effective from December 29, 2026, the Guidelines establish a framework governing the use and disclosure of synthetically generated content (“SGC“) in advertisements. Rather than requiring disclosure merely because AI has been used, the Guidelines distinguish between prohibited content, content requiring mandatory labelling and content that does not require labelling.

Understanding the Scope of the Guidelines

The Guidelines define SGC to include audio, visual or audio-visual content that is artificially or algorithmically created, generated, modified or materially altered using a computer resource to appear real or authentic, and which depicts an individual or event in a manner likely to be perceived as genuine.

The definition focuses on the realistic appearance and potential perception of the content, rather than the particular technology used to produce it. Accordingly, the use of an AI tool does not, by itself, determine whether labelling is required. For instance, an AI-generated depiction of a celebrity endorsing a product may require assessment under the Guidelines, whereas a stylised animation that does not portray a real individual or event as genuine may fall outside the definition.

The regulatory focus is therefore on the effect of synthetic content on consumer understanding and purchasing decisions, rather than on AI adoption as such.

Three Categories of Synthetic Content

1. Prohibited Content: A Disclaimer Cannot Cure a Misleading Advertisement

The Guidelines prohibit advertisements that are illegal, infringe third-party rights, contain misleading claims or otherwise contravene the ASCI Code. Importantly, adding an SGC label does not make otherwise impermissible advertising acceptable.

For example, an advertisement featuring an AI-generated person demonstrating extraordinary weight-loss results may remain misleading if the representation conveys an unsubstantiated or unrealistic product outcome. Similarly, a synthetic depiction of a public figure endorsing a product without authorisation may raise concerns regarding fabricated endorsements and third-party rights.

Advertisers must therefore examine both, the claims conveyed by synthetic content as well as the manner in which those claims are communicated. The prohibition reinforces an important principle: transparency regarding the use of AI is not a substitute for truthful, substantiated and legally compliant advertising.

2. Mandatory Labelling: When Disclosure Matters to Consumer Choice

Under the Guidelines, labelling is required where SGC materially influences consumer decisions and non-disclosure would mislead consumers. The Guidelines identify, among other examples, personalised endorsements using a real person’s replicated face or voice with consent, demonstrations of products that do not yet exist, and AI-generated sponsored product recommendations.

Consider a company promoting a prototype product through a realistic AI-generated demonstration of a feature that has not yet been developed. If presented as an actual demonstration of the product’s capabilities, the content may mislead consumers. Disclosure of its synthetic nature would therefore be material to an informed purchasing decision.

The Guidelines also address AI-generated sponsored product recommendations. For instance, where an AI-generated virtual influencer recommends a particular product pursuant to a paid brand collaboration, the advertisement must clearly identify the sponsorship using the formulation “Sponsored by [Brand]”. A generic disclosure that the content was created using AI would not, on its own, communicate the commercial relationship.

ASCI indicates that disclosures such as “Audio/Video created using AI” or “Audio/Video enhanced using AI” may be used, with other appropriate formulations permitted depending on the context. Applicable ASCI disclaimer requirements will continue to apply.

3. No Mandatory Label: Avoiding Unnecessary Disclosures

The Guidelines recognise that not every synthetic modification warrants disclosure. No label is required where modifications are minor or do not materially affect a consumer’s ability to make an informed choice. Illustrative examples include routine edits that do not alter the substance or core claims of an advertisement, decorative synthetic elements unrelated to the product’s capabilities, and accessibility features such as subtitles and closed captions.

For example, an advertiser using AI to change the background of a product photograph without altering the product’s appearance or represented characteristics may not need to disclose the modification. This illustrates that the Guidelines do not impose a blanket disclosure obligation on all AI-assisted advertising. Instead, they distinguish between inconsequential creative enhancements and synthetic representations that could materially distort consumer perception.

Compliance Implications for Advertisers

The Guidelines apply to all advertisers using SGC, and advertisements remain subject to the ASCI Code in its entirety. Non-compliance may result in recommendations for modification or withdrawal of the advertisement.

Advertisers should consider incorporating an SGC assessment into their existing advertising approval processes. This could include documenting the nature of the synthetic modification, evaluating whether it affects a product claim or endorsement, recording the rationale for labelling decisions, and coordinating with advertising agencies, influencers and digital platforms on disclosure requirements.

A further consideration is the interaction between the Guidelines and the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021, as amended in February 2026 (“IT Rules“). While the Guidelines adopt a materiality-based approach to labelling in advertising, the IT Rules impose separate labelling obligations concerning synthetically generated information on intermediaries. Advertisers should assess their obligations under each framework independently, rather than assuming that compliance with one necessarily satisfies the other.

Conclusion

The Guidelines introduce a structured approach to synthetic content in advertising, balancing consumer transparency with the legitimate use of emerging technologies. Their practical application, however, will depend on how advertisers assess materiality, distinguish misleading representations from permissible creative enhancements, and determine the appropriate form of disclosure.

As AI-generated advertising becomes more sophisticated, the principal challenge for advertisers will be to ensure that technological innovation does not compromise the accuracy of the representations made to consumers. A documented, context-sensitive assessment of synthetic content will be central to navigating this evolving regulatory framework.