Google's New AI Ad Agents Force Marketers to Choose Between Control and Performance
Google announced a suite of AI-powered advertising tools at Google Marketing Live yesterday, asking marketers to cede more control to machines in exchange for better campaign performance. The tools, built on Google's Gemini language model, include AI-generated product explainers, shopping ads with AI descriptions, and "Ask Advisor," an agent that can pull product details, set up campaigns, and surface insights across Google Ads and Analytics.
The rollout puts marketers in a bind. They want the performance gains AI promises. They don't want to surrender visibility into how their ad dollars work.
The transparency problem
Google's Performance Max platform has spent years as the poster child for black box advertising. Last year, agencies told clients to walk away from PMax entirely, shifting tens of millions of dollars off the platform over concerns about where ads actually ran and how Google optimized them.
Now Google is asking for more trust, not less. "The more autonomous these systems become, the more trust becomes the product," said John Geletka, founder and chief experience officer at creative agency Geletka+. Marketers must accept AI recommendations "without fully understanding the mechanics underneath them."
Google announced AI Performance Insights, a tool meant to show how brands appear in AI search results compared to competitors. The company also said it would integrate Meridian, its open-source marketing mix model, into Google Analytics 360 to help marketers measure performance across channels.
One agency president said he's noticed "more of an open, honest, transparent conversation" from Google's dedicated teams than in the past. But that's a low bar.
Marketers still spending despite doubts
Transparency gaps aren't stopping ad spending. AI-powered ad spend is expected to hit $57 billion this year, roughly 12% of the $475 billion U.S. ad market. None of the 13 agency executives interviewed for this report said clients were pulling money from Google because of the black box effect.
Performance gains matter more than granular insights to most marketers. Search queries have become longer and more conversational, making the detailed reporting that Google Search trained agencies to expect increasingly irrelevant, according to Ryan Schuster, director of search and social at Exverus.
Still, marketers want safeguards. They're asking for manual kill switches to account for AI hallucinations and requesting better reporting on things like AI inventory and where ad dollars perform best. Clients want sizable growth but "don't trust that AI built by media owners will deliver," one anonymous agency executive said.
For more on how AI is reshaping marketing operations, see our AI for Marketing resources or explore the AI Learning Path for Marketing Managers.
Your membership also unlocks:
