Marketing Dive Trendline on AI and Marketing Data

The age of artificial intelligence is here, and marketers in 2026 face a new landscape where the technology is woven into nearly every aspect of the advertising industry. AI is changing how campaigns are created, deployed and measured, and could help marketers finally deliver on holy grails like personalization at scale and closed-loop measurement as emerging channels, from connected TV to retail media, become performance driven.

But there’s still a lot of work that needs to be done to bring those promises to fruition. As stakeholders jockey for position, often promising one-stop shops and turnkey solutions powered by AI, the savviest marketers will need to navigate the changed ecosystem by separating reality from hype — especially around emerging areas like agentic AI.

When it comes to AI in 2026 and beyond, marketers would do well to embrace a version of the serenity prayer: accepting what they cannot change, courageously changing what they can and wisely knowing the difference between the two.

“You look at some of these structures, and you start to say, when everything works, it truly is like magic,” said Jacob Davis, executive director and global head of performance at Crossmedia. “It very infrequently works perfectly.”

In the last few years, major agencies, ad platforms, media conglomerates and ad-tech companies have rolled out AI-powered solutions that seek to automate much, if not all, of the advertising process. When launching campaigns, advertisers can pick and choose from Meta’s Advantage+ suite, Google’s Performance Max and Amazon’s full-funnel campaign offering; WPP Open, Publicis’ CoreAI and Omnicom’s Omni; and publisher offerings from giants NBCUniversal and Disney. 

While they have different capabilities and components, these solutions look to decrease costs and increase performance, often at the cost of transparency into their machinations. In that way, AI supercharges pre-existing machine-driven automation systems — “the algorithm on steroids” said Mathieu Roche, co-founder and CEO at ID5. 

“It's still very much a black box, which works if you're just interested in outcomes, however they define outcomes,” Roche said. “I don't think it works for the top-end of the pyramid of advertisers, but for small- to long-tail advertisers who are looking for traffic to websites or app downloads… there is a sliver of the market that really welcomes that.”

Platform, agency challenges

Over the last few years, AI has been integrated into marketing workflows around creative, planning, targeting and optimization. Each area represents a different challenge for AI and a level of tolerance for outsourcing by marketers. For example, brands are likely to retain control over their creative but more likely to hand over audience planning to AI, Roche claimed. The way media planners and buyers work could change the most as platforms roll out plain-language chatbots that assist in generating media plans.

However, these AI-powered platforms are usually mass-market solutions, not purpose-built for individual marketers. The question remains whether they can truly address brand needs. 

“Do they work the way they are supposed to work? Because these platforms are so big, they're not really changing things based on what a brand wants, even if they're willing to spend money,” said Unni Kurup, director of client consulting and strategy at Theorem.

As brands look to navigate this new AI-powered landscape, the role of agencies could evolve, turning the adland players into a layer of connectivity and enablement for environments like Meta and Amazon — walled gardens that have used the AI moment to strengthen their positions, explained Nicole Greene, vice president and analyst at Gartner.

“Every one of these platforms is going to have their own data. They're going to own the experience, and now they're going to own the creative, optimization and measurement. … You're playing by their rules,” Greene said. “Agencies might be a great way for brands that otherwise don't have the capacity to do it on their own to have visibility across those platforms.” 

Amid the push for automation, marketers still face questions around identity, a space that remains in flux despite Google’s decision not to deprecate third-party cookies. Marketers must decide whether to spend on media or on media plus identity resolution, which is potentially more effective. Currently, many are choosing the former, and possibly making a mistake, Davis said.

“If I could put $100,000 into LiveRamp, or I could put $100,000 into Meta, the smart choice would be $100,000 in LiveRamp,” Davis said, using the data-collaboration platform that recently secured a strategic partnership with Publicis as an example.

The ability to justify performance media offered by AI-powered ad platforms to other C-suite stakeholders could make it an easier choice, especially as marketers become wary of high fees in programmatic and suspicious of additional players in the media supply chain taking a cut of media spend. Through it all, marketers continue to face issues pinpointing which media drives conversions and value.

“Was it because we had a Kargo SSP overlay with PubMatic? Was it because of the creative that we ran? Was it because of LiveRamp, or was it because we ran it through The Trade Desk?” Davis posited, illustrating a common marketer quandary.

The rise of the agents

Just as marketers were getting comfortable with a new AI-powered status quo, the next step of the automation revolution has begun with the growth of agentic AI: fully autonomous systems that can coordinate without human intervention and could simplify and optimize applications like programmatic advertising.

WPP and Omnicom began 2026 by announcing new agentic AI offerings. PubMatic debuted an agentic operating system intended to solve pain points around programmatic that launched with partners including WPP Media, Butler/Till, Wpromote and MiQ, and is part of a consortium that launched the new Ad Context Protocol. Meanwhile, the IAB rolled out frameworks and roadmaps for the agentic future — an opportunity for adland to learn from the mistakes of past technical developments and bake in standardization from the start.

“Although AI solutions will increasingly take shape, the industry should expect several false starts in deploying agentic solutions,” said IAB Tech Lab CEO Anthony Katsur in a statement. “The promise of agentic AI is real and meaningful, yet its practical application will require years of market experimentation, standardization and alignment across platforms, agencies and publishers.”

Marketers that adopt agentic AI and let agents plan, test and continuously optimize campaigns will save time and resources on laborious tasks, giving them time to work on big-picture ideas like goal setting and experimentation.

“Agentic AI will change marketing by shifting the burden of execution,” said Upwave CEO Chris Kelly in emailed comments. “The real advantage comes from agents making thousands of small optimization decisions in real time… far more than any team could manage manually. Brands that adopt these systems early will simply move faster and learn faster than their competitors.”

But like other applications of AI in advertising, agentic AI success isn’t guaranteed. Along with sorting through a new alphabet soup of agent protocols — AdCP, MCP, UCP — marketers will need to have their own data, APIs and other tech stack elements prepared.

“Do you have quality APIs where you can pass through data? That's going to be really important, to make sure that you're agent ready, so that as these different platforms begin to consolidate and allow for more accessibility, you're able to deliver into these environments,” said Gartner’s Greene. “People [who say] it's all going to connect and be seamless… if that happens, I can probably sell you a bridge.”