Fashion Brands Who Want AI Agents to Recommend Them. Here's What They're Actually Looking For

I've been thinking about something a researcher said recently in a podcast conversation about the ChatGPT ad rollout. She spent close to two decades tracking the growth of social media as a commercial channel and was one of the first analysts to formally call it as a major ad market back when Facebook was two years old. She's seen this pattern before. And she made an observation, entirely unprompted, that struck a chord with me

She said the foundation for everything, organic AI recommendations and paid AI advertising, is the same thing: structured product data. Before a brand can think about placing ads on AI platforms, they need product feeds and data that AI engines can accurately read. Without it, neither the organic presence nor the paid placements work properly.

The AI advertising layer is evolving faster than most people realise. It’s evolving on every major platform simultaneously. ChatGPT launched its first ad test roughly a month ago to considerable drama. Amazon has been building advertising quietly inside Rufus. Walmart has Sparky. Google has ads adjacent to AI Overviews and AI Mode. Microsoft's Copilot has had basic search ads for nearly a year without much attention. Perplexity tested ads, pulled back, and pivoted to commerce. Every significant AI platform is building some version of a commercial layer right now, and the formats and rules are still being figured out in real time.

The instinct for most fashion brands will be to wait. Discover which platforms establish themselves. See what the ad formats settle into. Then decide where to put budget. That's probably the right call on direct ad spend for now. But there's a more important story underneath the ad rollout headlines.

Trust is the central problem every platform is navigating, and how they handle it tells you a lot about where this goes.

ChatGPT has been deliberate about keeping ads clearly separated from organic responses. The current formats are basic by design. They look like search ads because that makes performance comparison straightforward and because users recognise them. The concern is obvious: if advertising starts to influence what the AI recommends, the thing that makes the platform valuable disappears. OpenAI knows that.

The cautionary tale is sitting right there for everyone to see. Spend any time searching for a specific product on Amazon and you're essentially navigating sponsored content to find the item you actually searched for. Google isn't far behind. The accumulated weight of ad density over years has degraded the organic experience on both platforms enough that a significant portion of users started going elsewhere for initial research and ChatGPT grew partly on the back of that frustration.

Every AI platform building an ad product is managing the same tension: the organic experience is what makes the ad inventory worth buying, and the ad inventory slowly erodes the organic experience. It's not a solvable problem, it's a managed one. And the brands waiting for a clean, settled ad format to appear are waiting for something that probably never fully arrives.

The structural implication most brands are missing is this: the foundation for organic AI recommendations and the foundation for paid AI advertising are the same thing.

The same product data completeness that earns you an organic recommendation in a ChatGPT response is the same foundation that makes a paid placement on that platform perform. You don't build one for organic and a different one for paid. You build one thing. A data layer that AI engines can read accurately and confidently and it works across both channels. Retailers figured this out early. In the first weeks of ChatGPT advertising, something close to a quarter of the early advertisers were retailers, not because they had the biggest budgets but because they understood the risk of being left out and wanted to establish the foundation before the space got crowded.

This is the commercial case for the Agentic Trust Layer. It's not purely about organic discovery. It's about building data infrastructure that every AI commercial channel runs on and doing it now, before the paid layer matures to the point where brands are scrambling to retrofit a foundation they should have built months earlier.

The most interesting prediction I've heard about where all of this goes. and it's the one I'd encourage ecommerce brand owners to sit with, is that advertising will eventually stop being aimed at human eyeballs altogether.

Right now, an ad is something a person sees. It creates awareness, prompts consideration, influences a search. In an agentic world, the agent handles discovery, comparison, and increasingly the transaction. The human may never see an "ad" in any recognisable sense. Which means the commercial question shifts from how do I get in front of a customer to how do I get an agent to choose me.

The answer to that question isn't an ad format. It's your product data quality. Your review signals. Your entity coherence across the web. The degree to which an AI agent can make a confident recommendation for your brand because everything it can find about you tells a clear, consistent story.

I don't know exactly what AI advertising looks like in three years. Nobody does, the analyst I mentioned at the top said she's found that with AI, every month is a new year. Which I think is about right. But I'm fairly confident about which brands will be well positioned regardless of how the formats settle. It's the ones treating their product data with the same obsession they treat their product.

The brands waiting for the ad unit to arrive are going to find that the window to build the foundation underneath it has already closed.