As AI becomes embedded across the media ecosystem, marketers should focus less on the technology itself and more on the data powering it, according to Sean Black, general manager at Audience Path.

At EMARKETER’s Ad Buyer Strategies Summit, in conversation with our analyst Yory Wurmser, Black argued that marketers need to look beyond industry buzzwords and understand what AI systems are actually trained on.

While many companies are touting AI-powered planning and buying capabilities, Black warned marketers to scrutinize the underlying data sources.

"I believe that the core to any good, strong agentic platform or AI platform is what it's being trained on," Black said.

Signals replace static audience data

The shift toward signals represents a significant departure from traditional audience targeting approaches built on cookies and historical data. Black defines signals as "really understanding what people are doing now."

Black said because Audience Path analyzes billions of video streams, the company can continuously refresh its understanding of consumer interests and intent.

"For us, those signals are a core part of seeing 5 billion streams of video consumption that allows us to understand a signal in terms of: What are you watching? What's important to you, even real time," he said.

Rather than relying on cookies, he said Audience Path uses behavioral logic to determine which impressions matter most.

"What we're saying is that based on this behavior, based on what this person is watching, based on the types of other things that make sense to them, their intent, their interests, and so forth, these are the impressions that I want," Black said.

Agentic AI works best when specialized systems collaborate

Rather than having a single AI agent perform every task, Black envisions specialized agents working together.

"Not one agent is taking on the responsibility of everything, but several agents are communicating to each other as a team to come up with the best solution and the best audience," Black said.

Building those systems, however, requires significant investment and technical expertise.

"It's inherently difficult," Black said. "There's a lot of costs associated with these platforms. It's really difficult to have the resources."

AI will elevate strategy over execution

That evolution could reshape media buying workflows, but Black does not believe it will eliminate the need for human expertise. As AI increasingly automates campaign activation, marketers' value will come from strategic decision-making rather than execution.

"It's not about the activation part of it," Black said. "The strategy in which you go into this is inherently the most important part."

Black expects the industry to move steadily toward more automated buying systems, but believes that shift will place even greater demands on planners, agencies, and publishers.

"I think it's going to put more stress on us and the agencies and the publishers to better understand the strategy in which you're going to market," Black said.

CTV's next challenge is proving its contribution

The same strategic challenge applies to connected TV (CTV). While marketers continue searching for better ways to measure CTV's impact, Black believes the industry's focus should be on understanding how the channel contributes across the broader media mix.

"For us, the most important change is, how do you make CTV accountable to the other things you're running?" he said.

Rather than evaluating CTV in isolation, marketers should assess how exposure on CTV influences outcomes across channels.

Black also pointed to growing opportunities for interactivity in CTV, citing features such as QR codes and remote-enabled engagement tools that can make television advertising more measurable and actionable.

Ultimately, he sees both AI and CTV moving toward the same goal: helping marketers better understand audiences, improve accountability, and make smarter strategic decisions.

Watch the full session.

We prepared this article with the assistance of generative AI tools and stand behind its accuracy, quality, and originality.