The news: The Interactive Advertising Bureau (IAB) is working on a measurement framework to attribute ads that are served to AI bots, per Digiday.

The solution, set for a November 12 debut, will help marketers quantify AI’s role in purchase decisions, “especially when traditional signals are minimized,” said Caroline Giegerich, vice president of AI at the IAB.

Differences in opinion between working group members, which include agencies, brands, measurement vendors, publishers, and tech firms, shows how much uncertainty surrounds the logistics of AI advertising. The group is having a difficult time agreeing on “everything,” Giegerich said.

A resulting tool will likely split AI bot measurement based on awareness on one end (e.g., results in AI assistant output) and purchase decisions on the other.

Zooming in: The IAB’s framework release is only the beginning—the guidelines will transform over time, similar to how its retail media measurement standards evolved from its September 2023 debut to the final release the following December.

Like other IAB frameworks, we can expect AI measurement standards to cover transparency into data quality and processing; incrementality via modeling or control groups; reporting standards; and audience measurement such as reach, ad delivery, and viewability to prove ad exposure.

The IAB is addressing a major pain point for the 6 in 10 US marketers who say AI-driven discovery platforms are the most difficult to attribute accurately, per EMARKETER and Partnerize. But even when IAB standards are finalized, AI platforms may stay opaque, making measurements less than 100% accurate.

Implications for marketers: ROI and return on ad spend (ROAS) metrics depend on attribution—a measurement that has become cloudy in the past three years since ChatGPT hit the mainstream. Marketers’ budget decisions are based in part on attribution, but as genAI use grows as part of the discovery layer, they need a reliable way to measure it.

The IAB’s solution has the potential to both transform and split advertising metrics based on ads served to humans and those served to bots, which would also need to be weighed against each other to determine full impact. Until then, marketers will need to rely on standard attribution frameworks while separately ensuring AI engines are surfacing their brands through thorough research and AI visibility models.