Agentic ad buying has spent years promising that AI can plan, buy and optimize media so marketers don’t have to. The latest signals are more grounded: brands are writing small checks, a few campaigns are producing returns, and humans are still watching closely.
What changed in 2026
Apostra, which pivoted from sustainability to agentic buying, now lets AI agents discover, negotiate and approve deals—then execute them across digital publishers, linear TV and in-store audio. Marketers can manage campaigns across Meta, Google, Pinterest, Reddit, Spotify and ChatGPT through the platform’s agentic connectors.
The field is getting crowded:
- The Trade Desk is building a central AI conversational interface that routes buyers to specialized agents.
- PubMatic launched AgenticOS in January; Magnite has a buyer agent plus an orchestration layer across buy- and sell-side tools.
- Kargo introduced Project Kera, an AI planning and buying interface, in spring.
Outside media buying, Runway Ads can spin up ad variations and localize them for Meta, Google and TikTok, then pull performance data. At Dreamforce, Southwest Airlines said its AI agents will soon complete a flight ticket purchase. WPP opened Devon’s Point hub to automate parts of production.
The 5x that got attention
Rouge Care, a red-light therapy brand, spent $25,000 on CTV ads through PubMatic’s AgenticOS in Q2 2026 and credits AI agents with a fivefold return—over $125,000 in attributable sales. Apostra says it has already handled about $1.2 million in automated media buys for “very large” brands in the last three months.
Those numbers matter because they move the conversation from demo to budget. But the scale is still tiny. Gartner puts highly autonomous deployments at less than 5% of all marketing use cases in 2025, and predicts only 15% hit that mark by 2028. The real question is how far spend can scale and what infrastructure supports it.
Standards and accountability
Agentic advertising runs on rival protocols. Apostra uses AdCP, developed with Scope3 and other adtech firms; IAB Tech Lab backs AAMP; Amazon has Model Context Protocol. If buyers and sellers choose different standards, the industry could add another interoperability layer to an already fragmented supply chain.
Apostra is positioning itself as a neutral SaaS layer that doesn’t take a cut of media spend—its fee stays flat as ad dollars rise. Gartner analyst Andrew Frank says that could give advertisers a reason to favor a neutral coordination layer over walled gardens.
Accountability is the harder problem. “If agents reallocate, say $10 million from TV to creator inventory, and performance drops, who’s accountable?” Frank said. Agents can explain some reasoning, but the explanation may still fail. Expect budgets to stay in the experimental bucket until that question is answered—especially in premium, relationship-driven deals.
What to do now
Run a small, capped agentic test: one channel, a few thousand dollars, a clear revenue target. Audit the agent’s decisions and make vendors show which protocols they use. The early advantage will go to platforms that can show the work behind the numbers, not the ones that promise full autonomy.
Source: Adweek
