ASCI Releases AI Advertising Guidelines 2026: Key Facts, Need for guidelines, Four Key Areas, AI Labelling, Valid Labels, Significance and Challenges

Why in news?

The ASCI released the Guidelines for Responsible Labelling of Synthetically Generated Content in Advertising on 29 September 2026.

Key Facts

What is the ASCI?

• ASCI was established in 1985. It is an independent, voluntary self-regulatory organisation of the advertising industry. It is not a government regulatory body. 
• Its purpose is to promote responsible advertising and protect consumer interests through advertising self-regulation.
• The ASCI Code expects advertisements to be legal, decent, honest, truthful, not hazardous or harmful and fair in competition.

What is Synthetically Generated Content?

• SGC broadly refers to audio, visual or audio-visual material artificially created, generated or materially altered using digital or computational technologies, including Artificial Intelligence.
• In advertising, this could include AI-generated people or influencers, Synthetic voices, AI-created locations or events, Deep fakes, Digitally generated product demonstrations, AI-generated sound effects and AI-enhanced visual representations. 
• The key concern is not simply whether AI was used, but whether its use can affect how consumers understand an advertisement or make a purchasing decision.

Why were the Guidelines needed?

• Reality vs Artificial Representation: Consumers may believe that an AI-created event, person or product demonstration actually exists.
• Fake Endorsements: Deep fake technology can potentially make it appear that a celebrity or another individual has endorsed a product when they have not.
• Consumer Manipulation: Realistic AI-generated content may influence purchasing decisions without consumers knowing that what they are seeing is synthetic.
• Trust Deficit: Undisclosed synthetic advertising can reduce trust in both brands and the wider digital advertising ecosystem.

Four Key Areas of the ASCI Guidelines

• Prohibited Content (AI Label Cannot Make It Acceptable): Certain advertisements remain unacceptable regardless of whether they carry an AI-generated label. These include advertisements that are illegal, contain misleading claims, violate the ASCI Code, exaggerate product results or features through misleading synthetic representations, present non-existent locations or settings as real in a misleading manner or use unauthorised copyrighted material. 
  • Eg. if AI is used to create a fake celebrity endorsement without consent, merely putting an “AI-generated” label on the advertisement does not make the advertisement acceptable.

When is AI Labelling Mandatory?

• Synthetic Influencers and Ambassadors: If an advertisement uses an AI-created influencer or brand ambassador who appears realistic, consumers should be informed about the synthetic nature of the representation.
• Non-existent Products: If AI is used to demonstrate a product that does not currently exist, the synthetic nature of the demonstration needs to be disclosed.
• AI-Generated Product-Relevant Sound: Labelling may be required where synthetic sound effects are directly relevant to a product's main features. Eg. Artificially generating sound to demonstrate the supposed audio quality of headphones could affect consumer perception of their actual performance.
• Sponsored AI Product Suggestions: Paid or sponsored AI-generated product suggestions must specifically carry the disclosure “Sponsored by [Brand]”. This provision is significant as AI-based recommendation systems increasingly influence consumer choices.

When is AI Labelling NOT Required?

• Routine Editing: Colour correction, Lighting adjustment, Noise reduction, Minor blemish removal, Decorative Elements. AI-generated background or ambient elements that do not materially alter the product representation generally do not require disclosure.
• Obviously Fantastical Effects: If an effect is clearly unrealistic or imaginary and an ordinary consumer is unlikely to mistake it for reality, disclosure is generally unnecessary.
• Accessibility Uses: Subtitles, Translations, Accurate audio descriptions. The guidelines attempt to prevent “label fatigue”, where consumers could begin ignoring disclosures if virtually every minor AI-assisted edit carried a warning.

What Labels Can Advertisers Use?

• Where disclosure is necessary, advertisers can use clear labels such as “Audio/Video created using AI” or “Audio/Video enhanced using AI”
• Advertisers may also use appropriate disclosure mechanisms provided by advertising platforms.
• Alternative labels are permitted as long as they accurately communicate the nature of the synthetic content to consumers.
• The disclosures should also comply with the ASCI Code's requirements relating to disclaimers wherever applicable.

Risk-Based Logic of the Guidelines

• AI/Synthetic Content Used in Advertisement → Does it violate law, rights or ASCI Code? → Yes (Prohibited even with AI label) → If No, Can it materially affect consumer understanding or decision-making? → Yes (Disclosure/Labelling Required) → If No, (Routine /Decorative /Accessibility Use, No AI Label Required)

Significance

• Tackling Deep fakes: Unauthorised use of a person's likeness or deep fake endorsements is specifically treated as unacceptable.
• Transparency in Digital Advertising: Mandatory disclosure can reduce information asymmetry between advertisers and consumers.
• Protection of Personality Rights: Restrictions on unauthorised replication of an individual's likeness or voice reinforce the importance of consent in AI-driven advertising.
• Trust in Advertising: Clear disclosure can help maintain consumer confidence as realistic AI-generated advertising becomes more common.

Challenges

• Detecting AI content: Highly sophisticated synthetic content may be difficult to identify.
• Rapid technological change: AI capabilities can evolve faster than advertising standards
• Deep fake detection: Increasingly realistic synthetic voices and videos make verification difficult
• Defining material influence: Determining when synthetic content meaningfully affects consumer decisions can involve context-specific judgments

FAQs