Adtech Standardization in the Age of Agentic AI and Contextual Ads
As agentic AI and contextual advertising move into the adtech mainstream, changing circumstances arguably call out for new standards to establish a working level of consistency and interoperability across the industry lest proprietary standards and incompatible secret sauces continue to drive players and packagers. Sargeway Owner Sarge Sargent, Dataxis Principal Analyst Ophélie Boucaud, and Servers.com Global Sales Executive Adtech Bradley Lewington discuss why standards and integrations are needed for widespread adoption and review the current and emerging state of specs and standardization across the CTV ecosystem in this clip from Streaming Media Connect 2026.
Bots to Bots Talking Apples to Apples
Boucaud begins the conversation by bringing up questions she wants to ask Sargent about standardization at the industry level: “Can you tell us a bit more about what’s at stake? Why do we need standardization? How can we achieve that? What are the different paths that you’re exploring?”
Sargent says that IAB Tech Lab “published their agentic spec and it’s leaning directly into OpenRTB. So you have bots talking to bots, whereas before you had a person talking to maybe a bot on the other end—they wouldn’t necessarily know—but now you have bots talking to bots to make deals in real time.” He’s noticing that each AI company has its own “secret sauce.” For example, fellow panelist Crystal Foote’s Digital Culture Group has its own LLM, while Vionlabs and TwelveLabs each also have a proprietary secret sauce. “I talked to, for example, Bitmovin, who’s been in the media game for a long time,” he says, and “they have a player, they have a packager, and then they build into their components some AI that can also pull out scene context and then signal that in line.”
When bots talk to other bots, but they have different secret sauces, “these bots need to talk apples to apples” and do so “at a homogenized layer so that everybody knows exactly what they’re talking about when it comes to these ad assets, or when it comes to this ad break, that I need to make a decision on,” Sargent explains. These are the questions that need to be asked: “What’s in the inventory? Can I actually match the inventory based on just the tone or the culture or the vernacular in the actual scenes?” His work focuses on “creating that baseline so everyone can talk and [do a] transaction at the same level. So we’re talking the same language, but inside, you don’t have to change your publishing workflows. Everything could be your own secret sauce” that no one else can access.
Reducing the Complexity
Boucaud summarizes, “So the idea is not to actually get rid of the complexity, but rather make it interactive between those different complex layers that every environment is developing at the moment, just so that the industry still makes sense of common currencies and common standards.”
Sargent believes it’s important to keep this in mind when looking at TV and movies. “You could have a Disney movie, but you can see that Disney movie on Netflix or HBO Max. You have different TV series that will be across multiple different streamers. Those TV series, those publishers create metadata about those assets, but they may not create metadata that talks to the scene level,” he says. “If two people are watching the same asset at the same time and they hit the programmatic marketplace, how is that asset being described in the marketplace or a specific scene in that asset at the ad break? How’s it being described in the marketplace? Is it being described the same or is each publisher’s different way of talking about that asset being sent to the marketplace?” He reiterates that simplification is the goal.
100% Fragmentation
Boucaud asks Lewington for an overview of the fragmentation of understanding context and what he thinks about the efforts for standardization.
Lewington replies, “Scene-level contextual is way ahead of so many of the standards that were originally designed for page- and content-level contextual. And what I see at the moment is the biggest risk in moving forwards in this kind of AI and contextual era is 100% fragmentation. Like [Sargent] said, if you have almost every vendor on the demand/supply side creating their own signals, models, and AI stack from a buyer perspective as either an agency or as a publisher, trying to compare like for like for different platforms and different providers becomes so, so tough.” He acknowledges that this differentiation is inevitable, though, in a healthy marketplace.
A common interface that different providers can use to talk to each other and to take measurements would benefit the marketplace, Lewington believes. “Realistically, if consuming contextual signals requires bespoke integrations with every single provider, it’s just going to put up such a barrier of adoption as we move more into this kind of era,” he notes.
Join us November 9–11, 2026 for more thought leadership, actionable insights, and lively debate at Streaming Media Connect 2026! Registration is open!
Related Articles
Streaming Media Connect returned this week with six dynamic keynotes from NASA+, Google, TiVo, BallySports, Wowza, and TwelveLabs, and peak-form panels on refining your live streaming playbook, tuning latency, matching ads to content, scaling streaming architecture, curbing churn, leveraging AI in personalized content discovery, innovating streaming UX, optimizing QoS and QoE, and a timely dissection of Disney's 4K EU blackout.
14 Aug 2026
AI makes it possible to model delivery across live, VOD, and FAST together and to update forecasts as signals come in, instead of planning channel by channel.
23 Jun 2026
As TV viewing moved to streaming, audiences fragmented across apps and services. The mass audience did not disappear. It spread. In response, the homescreen became one of the most important surfaces in media, bringing viewers back together at the point where every TV journey begins. It is the gateway to streaming content and brand storytelling, and increasingly the place where fragmented viewing behavior reconnects. In that sense, it has helped solve one of modern TV's biggest challenges. But solving audience fragmentation has exposed another layer of complexity.
16 Apr 2026
At Streaming Media Connect 2026, Tavant brought together a group of media and advertising leaders for an honest, wide-ranging conversation about what AI is actually doing for the industry today. What emerged was a clear pattern: the companies gaining the most from AI are not treating it as a feature layered onto existing systems—they are building core advertising operations around it.
09 Apr 2026