IAB develops AI transparency guidelines as platforms show inconsistent standards

The Interactive Advertising Bureau (IAB), a digital marketing trade organization, is developing comprehensive AI transparency and disclosure guidelines for advertising, aiming to establish industry-wide standards for when and how AI use should be disclosed to consumers. The initiative comes as marketers grapple with reputational risks from AI-generated content errors and the launch of OpenAI’s Sora app, which can create photorealistic videos from text prompts.

What you should know: The IAB’s approach focuses on meaningful disclosure rather than labeling everything touched by AI.

• Caroline Giegerich, the IAB’s vice president of AI, emphasized that over-labeling risks desensitizing consumers and could “dilute trust instead of strengthening it.”
• The guidelines aim to flag AI use only when it could genuinely mislead audiences, balancing transparency with practical implementation.
• A working group of brands, agencies and platforms is establishing shared baseline principles for disclosure across the advertising ecosystem.

Why this matters: Current AI labeling policies across major platforms like Meta, TikTok and YouTube are inconsistent, creating potential brand safety issues.

• Detection is unreliable unless AI metadata is embedded directly in the creative content itself.
• The IAB wants to create “portable and consistent” transparency standards before regulators impose their own requirements.
• As Giegerich noted: “If we don’t align now we’ll end up with 20 different versions of transparency, and one of them will mean anything.”

Industry response: Multiple organizations are developing complementary AI standards and guidance.

• The Media Ratings Council (MRC), which sets measurement standards for digital advertising, is creating a comprehensive standalone AI standard, targeting market release within the next year.
• The World Federation of Advertisers (WFA), a global association of national advertiser associations, operates an AI community of over 900 senior marketers and a steering team of 10 advanced brands including L’Oreal, Unilever, and Diageo.
• WFA’s Gabrielle Robitaille explained their approach as “less about setting standards and more about surfacing” current best practices.

Key details: The MRC’s upcoming AI standard will address six critical areas.

• Updates for measurement with zero-click search and agentic AI (AI systems that can act independently to complete tasks).
• Invalid traffic considerations related to AI user agents and human pattern simulation.
• Brand safety requirements for AI-generated content.
• Transparency requirements for AI buying agents in auction systems.
• Content provenance measurement and labeling (tracking the origin and authenticity of digital content).
• Training, monitoring and disclosure requirements for AI detection systems.

The agency perspective: Media agencies are already steering ad dollars away from AI-generated “slop” content.

• “Our guidance to advertisers has been, ‘we’ll keep an eye on it and create [exclusion lists] where we can,'” said David Dweck, general manager at Go Fish Digital, a media agency.
• Karen Ram from Canvas, a creative agency, emphasized that “the whole thesis that people trust people, goes away when you bring AI into it.”
• Some agencies are building or buying creator vetting systems, with New Engen partnering with DoDilly for brand safety monitoring.

What they’re saying: Industry leaders acknowledge the complexity of AI content oversight.

• Scott Sutton, CEO of Later, a social media management platform, sees faceless AI creators as viable for “bottom-of-funnel campaigns where the goal is sales and product discovery more than it is brand awareness.”
• Dweck compared the current situation to programmatic advertising’s early challenges, noting the reliance on human judgment to identify when AI creators have “crossed the line.”