By Kris Smith, Director of Paid Media, Stratacomm
For years, media professionals have been asking the same question: What will AI mean for the future of paid media?
This week, we got part of our answer, and it came with a price tag and a programmatic on-ramp.
ChatGPT’s partnership with Criteo marks a pivotal moment in the evolution of conversational AI as an addressable media channel. While the ad industry has a history of both overhyping and underestimating emerging platforms, this development deserves serious attention from anyone responsible for media strategy.
A Familiar Pattern, An Unfamiliar Platform
If you’ve been in this business as long as I have, you’ve seen this movie before. Cable. Paid search. Social media. Connected TV. Every major media channel started as an experiment. It was expensive, uncertain and accessible only to organizations willing to take a calculated risk.
ChatGPT ads are following the exact same trajectory. And honestly? I’m here for it.
What’s different this time is the conversion signal is already there. Early data suggests ChatGPT advertising can convert at 1.5 times the rate of other digital channels. That’s a business case waiting to be tested.
The Criteo partnership begins to lower the cost barrier that has kept this channel largely inaccessible. Up until now, the economics were so prohibitive that no responsible media strategist would recommend it without significant platform incentives. That equation is starting to change.
The Muddy Landscape Ahead
Here’s what we know: There’s a strong signal of intent when someone is actively conversing with an AI to solve a problem. That context creates a unique advertising opportunity potentially more valuable than passive social scrolling or even active search behavior.
Here’s what we don’t know yet: How targeting works at scale. What creative constraints will look like. How end users will respond compared to other platforms. Whether CPMs will stabilize at economically viable levels.
The landscape is muddy. But that’s exactly when the most valuable learning happens, and when the organizations willing to get their hands dirty gain the biggest advantage.
Test and Learn, Not Wait and See
This is where agencies and organizations will split into two camps: early adopters and skeptical observers.
Those who wait for the “right time” will eventually enter a mature market with established benchmarks, proven best practices and intense competition. They’ll have certainty, but they’ll pay for it in higher CPMs and missed learning opportunities.
Those who test now — strategically, with appropriate budgets and clear success metrics — will be defining what works. They’ll understand targeting nuances, creative formats and audience behaviors before their competitors even start paying attention.
For me, it needs to be test and learn, not wait and see. I don’t have patience for the sidelines when there’s this much potential on the field.
What This Means for Your Media Strategy
I’m not suggesting every client should immediately allocate budget to ChatGPT ads. That would be reckless. But dismissing this channel as “too early” or “too expensive” without evaluation is a strategic mistake.
The right approach depends on your client’s appetite for innovation and their ability to commit “test and learn” budget without guaranteed returns. If you have clients who can stomach that investment, this is the time.
Yes, CPMs will likely be higher than you’d prefer. That’s the price of early entry. But consider the upside: What if the ROIs pay off exponentially because you learned how to leverage this channel before the market matured?
The worst thing that could happen isn’t wasting budget on a failed experiment. It’s waiting too long and missing the window. I’ve seen too many organizations make that mistake.
The Broader Shift: From Search to Conversation
This signals something larger: the evolution from search engine marketing to conversational AI engagement. Just as SEO evolved into GEO (generative engine optimization), we’re now seeing SEM transition into paid conversational placements.
The organizations that understand this shift will have a significant advantage as AI-powered interfaces become the primary way people discover information and make decisions.
The Bottom Line
We’re at an inflection point. Conversational AI advertising is moving from experimental to accessible, from theoretical to measurable.
My recommendation? If you have the budget, the client appetite and the strategic rationale, test it now. Learn what works. Build institutional knowledge while the costs of entry are still declining and the competition is still figuring out whether to pay attention.
Because the agencies and organizations that lead are the ones willing to learn in the messy early days, not the ones who wait for someone else to figure it out first.
And trust me, I’ve been doing this long enough to know which camp I’d rather be in.
What’s your take? Are you testing conversational AI advertising, or taking a wait-and-see approach? Let us know.
