India's ad regulator ASCI orders AI labels on ads that could mislead

India's ad regulator ASCI orders AI labels on ads that could mislead

29 September 2026: The Advertising Standards Council of India (ASCI) has published guidelines that require advertisers to label AI-generated images, video and audio when the content materially influences consumer decisions and leaving it unlabelled could mislead. The guidelines take effect three months after publication. MediaNama reports that ASCI released them on 29 September; the document itself is signed 17 September.

Not yet independently verified. The PDF is dated 17 September while MediaNama reports a 29 September release, and no exact start date is stated. ASCI's decisions are self-regulatory and carry no legal penalty. We will update this when it can be confirmed, and remove this note.

ASCI defines synthetically generated content as audio, visual or audio-visual material created or materially altered by a computer so that it appears real and shows a person or event in a way people could take for genuine. Brands are asked three questions: whether the content is central to the persuasive message, whether its use misleads about claims or performance, and whether missing disclosure would create a false impression.

Labelling is mandatory in listed cases, among them synthetic influencers and ambassadors, a real person's face or voice replicated for personalised messages even with consent, fabricated events or settings, and demonstrations of products that do not yet exist. The guidelines also cover paid AI chatbot suggestions: if a brand has sponsored a recommendation, it must be shown as "Sponsored by [Brand]".

Some uses are prohibited whatever the label: fabricated endorsements, exaggerated results shown through generated images, non-existent locations presented as real, and unauthorised copyrighted work, deepfakes or likenesses. No label is needed for routine edits such as colour correction, decorative backgrounds, obviously fantastical effects, copy drafting or accessibility aids like captions and translations. Acceptable labels include "Audio/Video created using AI" or "Audio/Video enhanced using AI", or labels supplied by ad platforms.

ASCI Secretary General and CEO Manisha Kapoor said a label alone may not make a misleading advertisement acceptable, and that responsibility stays with advertisers. MediaNama points out two gaps: the guidelines do not say whether an ad stitched from many AI clips needs one label or several, and they set no size or duration for labels. It also notes that platforms already face separate labelling duties under India's IT Amendment Rules 2026 since 20 February.

Sources

• ASCI: Guidelines for Responsible Labelling of Synthetically Generated Content in Advertising (PDF, dated 17 September 2026), 17 September 2026
• MediaNama: India's new AI ad labelling guidelines, what brands must disclose, 6 October 2026