The Interactive Advertising Bureau has released a document titled "Measuring Visibility in the AI Era," a guidance paper aimed at advertisers and publishers trying to track how consumers find brands through AI-powered discovery tools such as chatbots and AI search summaries.

The release matters because the measurement problem is real and, so far, unsolved. When a chatbot recommends a product, no ad server records an impression. When an AI summary replaces ten blue links, the click that publishers counted on simply does not happen. Advertisers are spending against audiences they can no longer count.

The announcement itself carries little detail. The IAB has not yet published, in the material available at time of writing, the specific metrics, formulas, or standards the paper proposes. What we know is the title, the publisher, and the intended audience.

What does the paper claim to do?

According to the IAB, the document "helps navigate AI-powered discovery." That phrasing is doing heavy lifting. In practice, trade-body papers of this kind typically do three things:

• Define shared vocabulary for a new channel
• Propose measurement frameworks that vendors can implement
• Set baselines that media buyers can put into contracts

Whether this paper delivers on all three, or on any of them, cannot be confirmed from the announcement alone. The IAB has a track record here — its measurement standards for viewability and invalid traffic became industry defaults — but standards take years to harden, and AI discovery is moving faster than any standard-setting process on record.

Why is the industry asking for this now?

Search behavior is shifting. Publishers report declining referral traffic as AI-generated answers absorb queries that once ended in a click. Advertisers report the mirror problem: they cannot tell whether an AI assistant mentioned their brand, a competitor's, or neither.

That gap is not a rounding error. Brand visibility inside AI outputs — sometimes called AI visibility or answer-engine presence — has spawned a cottage industry of vendors selling "AI visibility scores" within the past two years. The IAB's entry suggests the trade body sees enough advertiser confusion to warrant an official framework.

A caveat belongs here: no third party has validated any of these measurement approaches, including whatever the IAB proposes. An industry standard is a starting point for verification, not verification itself.

Who should read it?

The primary audience is advertisers, agencies, and measurement vendors deciding what to buy, sell, and build. Publishers have a parallel stake: if visibility inside AI answers becomes a billable unit, publishers whose content trains or feeds those answers will want a share of it.

The paper does not, based on available information, address compensation for content used by AI systems. That question sits with courts and licensing negotiations, not standards bodies.

What is not said

Three absences stand out:

• No named methodology has been disclosed in the announcement
• No participating vendors or advertisers are listed as endorsers
• No timeline for adoption or certification appears in the public material

Standards without named implementers tend to gather dust. The IAB's viewability standard succeeded because major ad exchanges adopted it within roughly two years of publication. The test for this paper is the same: who ships against it first.

Mart Signal will follow up when the full document and its methodology become available.

By Nathan Brooks