Below is a roundup of last week’s notable industry news, with summaries and our opinions. It’s protocols week! The Trade Desk is reworking RTB protocols, while others are working on agentic protocols…

The Ad Context Protocol Aims To Make Sense Of Agentic Ad Demand
Source: AdExchanger
October 15th, 2025

Summary: A new industry consortium has launched the Ad Context Protocol (AdCP) to standardize how AI agents communicate across the advertising supply chain. The founding members—including PubMatic, Scope3, Triton Digital, Yahoo, Optable, and Swivel—are each committing $10K annually plus time and resources to establish open-source protocols for agentic tools to negotiate, plan, and execute media campaigns. Scope3 CEO Brian O’Kelley wrote this blog post about the new initiative.

We're still a ways from this happening, but the goal is to help buy-side and sell-side AI agents talk to each other effectively. AI agents are AI tools that can perform human-like tasks without human intervention. Initially, AdCP will offer two tools. Audience Activation Protocol, which will help advertisers find the right audiences, and Media Buy Protocol, which will enable AI agents buy media. An advertiser might use Audience Activation Protocol to prompt: "I want to find parents shopping for back-to-school supplies with household income over $75K in the Northeast," and many different publisher agents might respond with matching inventory packages and audiences to fit these criteria.

Eventually, these agent-to-agent deals could expand beyond digital to include OOH, linear TV, and other channels. The consortium hopes to have solid documentation and practical demonstrations within a year. 

Opinion: So let’s get this straight. The AdCP idea is to make it easier for buyers to discover audiences and buy advertising across many different platforms and publishers in a standardized, scalable, and efficient way. Hmm. Where have we heard of that concept before?

Okay, so this is not some earth-shattering value prop. We’ve been doing this for a long time in digital advertising using ad tech and programmatic and RTB. Glad we got that out of the way. But, this is compelling from the perspective of standardizing the use of agentic AI in ad tech. Clearly all of ad tech is going to move towards agentic workflows. So tackling these problems right now, in an open-source, consensus-driven fashion, is important work.

The biggest issue with using agentic AI tools to perform tasks today is a lack of a common language (AKA standardization) and trust (because, you know, this).

An agentic protocol (theoretically) enables standardization and trust! 

And open-sourcing the protocol means it can’t become some kind of strategic wedge for the walled gardens (like auction logic became for Google until header bidding came along); it becomes democratized across the industry. 

What else do we think standardized, agentic AI protocol could enable for the industry?

In this, there are obviously major implications for companies and people. Some bad, like companies get crunched out because they aren’t agentic-first and people lose their jobs. And some good, like we get rid of a bunch of unnecessary ad tech intermediaries and cut the ad tech tax. It should yield more value for publishers and for advertisers, which is fantastic!

But then again, that’s what we said about programmatic many years ago…

Thank you for reading the U of Digital Newsletter.
Continue reading by upgrading.

The U of Digital Weekly Newsletter is intended for subscribers, but occasional forwarding is okay!

To subscribe visit Uof.Digital/Newsletters or contact us directly for group subscriptions.

And remember, U of Digital helps teams drive better outcomes through structured education on critical topics like programmatic, privacy / identity, CTV, commerce media, AI, and more. Interested in learning more about how we can supercharge your team?