AI Agents Are Now Buying Real Ad Dollars — About $1.2 Million of Them

AI agents have completed roughly $1.2 million in automated media buys over the last three months for what Apostra calls "very large" brands — a tiny number by ad industry standards, but hard evidence that agentic buying now moves real money, not just slideware.

September brought a wave of agentic product launches and updates across the adtech stack. Apostra, a company that pivoted from sustainability to agentic advertising, now lets AI agents discover, negotiate, approve, and execute deals across digital publishers, linear TV, and in-store audio. Its agentic connectors support campaign management across Meta, Google, Pinterest, Reddit, Spotify, and ChatGPT.

Other players are assembling comparable systems:

• The Trade Desk is building a central AI conversational interface that routes media buyers to a growing roster of specialized agents.
• PubMatic launched AgenticOS in January.
• Magnite has introduced a buyer agent plus an agentic orchestration layer spanning buy- and sell-side functions.
• Kargo rolled out Project Kera, an AI-backed planning and buying interface, in spring.

Beyond media buying, Runway — the AI video startup — entered the ad business with Runway Ads, an agent that creates variations of existing ads, localizes them for Meta, Google, and TikTok, and then pulls performance data. At Dreamforce, Southwest Airlines said its AI agents will soon handle more complex tasks, including completing a flight ticket purchase. WPP opened Devon's Point, a hub using AI-powered tools to automate parts of development and production.

What results are advertisers actually seeing?

One data point stands out. Red-light therapy brand Rouge Care spent $25,000 on CTV ads and drove over $125,000 in attributable sales — a fivefold return on ad spend — with the awareness campaign facilitated through PubMatic's AgenticOS in the second quarter of 2026.

That is an official figure credited by the brand itself, not an agency case study embellished for a keynote. It is also, so far, an outlier rather than a trend.

Gartner's data tempers the hype: highly autonomous deployments represent less than 5% of all marketing use cases in 2025, and the firm predicts that by 2028 only 15% of agentic AI deployments will be highly autonomous. For practitioners, the near-term implication is clear — agents can execute, but humans are still watching closely, and budgets are staying small.

Who coordinates the agents — and who pays for the layer?

Apostra runs on AdCP, a protocol developed by a group of adtech companies including Scope3 that gives agents a common language to interact. It is one of several competing standards, alongside IAB Tech Lab's AAMP (Agentic Advertising Management Protocols) and Amazon's Model Context Protocol.

Multiple protocols create a fragmentation risk. If buyers and publishers back different standards, the industry could inherit yet another interoperability problem in a sector already notorious for fragmentation — and that friction directly caps how far agentic spend can scale.

Apostra's business model is notable for buyers: it takes no cut of media spend, positioning itself as SaaS with a flat fee that doesn't rise with ad dollars. Andrew Frank, VP and distinguished analyst at Gartner, said that neutrality could give advertisers a reason to favor an independent coordination layer over platforms like Amazon or Meta.

Who is accountable when an agent reallocates $10 million?

The accountability question may matter more than the plumbing. Frank framed it directly: "The problem is agents are supposed to make strategic decisions about budget allocations. If agents reallocate, say $10 million from TV to creator inventory, and performance drops, who's accountable? Agentic buying introduces this whole new decision layer whose recommendations may not be fully transparent."

That opacity is a practical budget constraint. Even when an AI reveals some of its reasoning, the reasoning may not work out — and that risk keeps brand spend in the experimental bucket.

Premium advertising poses the steepest barrier. Large deals there remain heavily relationship-driven, Frank noted, and publishers and buyers have spent years building their own infrastructure — the segment least likely to hand control to agents soon.

The open question for the next 18 months is not whether advertisers will fund agentic buying — they already are — but how far that spending scales, and which protocol ultimately becomes the infrastructure underneath it.