By 2028, AI Will Quietly Decide How 70% of Global Ad Spend Gets Priced

Gartner predicts that by 2028, more than 70% of global ad spend and 80% of U.S. ad spend will flow through self-serve advertising platforms where AI materially influences media buying, cost, and outcomes. This is “back-office” AI embedded in pricing and delivery algorithms, distinct from generative AI producing ad creative — and Gartner’s warning is specific: better platform economics from AI doesn’t automatically translate into lower costs for the advertiser. The firm’s three recommendations for CMOs all point the same direction — concentrate investment and talent on the platforms most strategically critical rather than spreading thin, build independent measurement instead of trusting platform-reported efficiency, and avoid over-relying on agency-built AI platforms given how often agency relationships turn over. The throughline is that as AI increasingly makes buying decisions behind the scenes, marketing leadership’s job shifts from running campaigns to maintaining strategic control and measurement independence.

That reframing echoes a point Forrester’s Rani Salehi has made about how CMOs are misreading AI’s organizational impact generally. Salehi argues the wrong question is “what roles do I need to hire or cut” — AI absorbs specific tasks like predictive modeling or first-pass creative without eliminating the roles around them, it just raises the bar for what those roles must deliver: reading AI output, challenging it, defending decisions to leadership. Applied to ad spend, that’s exactly the posture Gartner is asking CMOs to take toward AI-priced media buying — judgment and verification, not passive trust in whatever the platform reports back.

Verification is also the crux of OpenAI CFO Sarah Friar’s own framework for enterprise AI ROI. Her proposed metric, “Useful Intelligence per Dollar,” rejects cost-per-token as the yardstick in favor of true cost per successfully completed task — factoring in human review, retries, and rework, not just sticker price. It’s the same discipline Gartner wants CMOs to apply to ad platforms: don’t grade AI spend on the number the vendor hands you.

For consulting and enterprise leaders, the ad-spend prediction is really a preview of every AI-priced vendor relationship: the platforms doing the buying have no incentive to make their own math easy to audit.