The Last Holdout
For years, the advertising industry automated everything it could — media buying, audience targeting, bid management, attribution, reporting. The one thing that stayed stubbornly human was creative. Not because the industry was sentimental about it, but because no tool could reliably produce work that didn't look like it was made by a tool.
OpenAI is now making a direct move on that holdout.
According to reporting from Digiday, OpenAI is working to automate ad creative — not assist with it, not suggest variations, but produce the actual deliverables advertisers run. That's a meaningful distinction. Plenty of platforms already offer AI-assisted creative features. What OpenAI appears to be pursuing is something closer to end-to-end production.
Why This Is a Bigger Deal Than It Sounds
Creative production is where agencies have quietly rebuilt margin after programmatic buying gutted their media commissions. A brand might pay a holding company agency a fraction of what it once did to place media, but it still pays real money for the creative that runs in that media. Volume-based creative — social assets, display variants, video cuts for different formats — has become a reliable revenue line.
If OpenAI can automate that output at acceptable quality, it doesn't just change a workflow. It changes a pricing conversation. Clients who are already asking why creative costs what it does will have a new data point in that negotiation.
This is also a signal about where OpenAI sees its commercial future. Selling API access to developers is one business. Sitting inside the advertising production stack — where budgets are large, cycles are fast, and the appetite for cost reduction is structural — is a different and potentially much larger one.
What Agencies Should Actually Be Worried About
The immediate risk isn't that OpenAI replaces creative directors. It's that it replaces the production layer underneath them — the studios, the retouchers, the motion graphics teams, the social content farms. That layer is large, it's distributed across agencies and independent shops, and it has been the quiet employer of a significant portion of working creative talent.
Agencies that have already built proprietary AI creative tooling — or that have restructured their creative operations around AI-assisted production — have some insulation. They can argue they're adding judgment and brand governance on top of the automation, not just running the automation itself.
Agencies that haven't made that shift are now in a more uncomfortable position. The window to get ahead of this particular curve has been closing for two years. It may now be closed.
The Brand Side of This
For marketers, the pitch is obvious: faster creative, lower cost, more variants for testing. Performance-focused brands — direct-to-consumer, e-commerce, app marketers — will move toward this quickly if the output quality holds. Brand marketers with stricter guidelines and higher stakes on consistency will move more slowly, but they'll move.
The more interesting question is what happens to creative strategy when creative production becomes cheap. If making the ad costs almost nothing, the value shifts entirely to knowing what ad to make. That's a different skill set, and it lives in a different part of the organization than production does.