AI Strategies   //   May 12, 2026

OpenAI makes it easier to run shopping ads in ChatGPT

This story was first published by Modern Retail sibling Digiday.

OpenAI is making it easier for e-commerce companies to run ads on ChatGPT, adding automation that lets retailers generate ads directly from their product catalogs rather than building them one by one. 

To be clear, the ad itself looks the same to the user. The so-called “product feed” campaign will still appear in the same placement as any other ChatGPT ad — below the response, clearly labeled as sponsored. What’s new is what’s happening behind the scenes. Retailers connect their product catalog, set filters for which products are eligible and let the platform handle the rest, generating ads automatically from product names, images and attributes in the feed. 

For a brand with thousands of products, that’s what makes running ads there actually viable. Until now, retailers have been able to upload their product catalogs to ChatGPT so the platform can surface accurate prices, availability and product information when someone asks a shopping question. But there was no way to connect that data to paid ads. Brands wanting to advertise had to build their campaigns manually, product by product. Now the same catalog that powers their organic presence can generate their ads automatically. 

The setup will be familiar to anyone running shopping campaigns on Google. Retailers largely repurpose the same structured product file they already send Google, lowering the barrier to entry considerably. At least one retail brand has already gone through the process via Criteo, OpenAI’s first ad tech partner, and found it more straightforward than expected, according to an ad buyer running those campaigns. To win over others, OpenAI is currently asking new e-commerce partners to submit a sample of 100 products before sharing their full catalog. The platform can handle up to a million SKUs per advertiser, said another ad exec briefed on the update. No public launch date has been given. 

“Adding automations to generate ads from a product catalog is essentially table stakes in the age of AI,” said Sonata Insights’ founder and chief analyst Debra Aho Williamson. “What OpenAI is doing is somewhat similar to what Google, Meta and Amazon already offer retailers. The difference is that ChatGPT is serving ads based on conversational intent rather than signals from search behavior, social engagement or browsing in a marketplace environment.”

That in itself says a lot about OpenAI’s shopping ambitions. Weeks ago, the company killed off its instant checkout feature, a move that looked to some like a retreat from shopping as a category. The product feed capability suggests a different reading. OpenAI is abandoning shopping so much as changing how it plans to make money from it. Rather than taking a cut of transactions, it’s going after the advertising budgets that e-commerce brands already spend trying to reach shoppers.

And it won’t just be money coming into OpenAI directly. 

“We have the same feature, so once OpenAI supports it, we post our feed to their feed. And that’s just an example of, like, a seamless connection,” said Yang Han, co-founder and CTO of StackAdapt, one of the ad tech vendors advertisers can buy ads in ChatGPT from. 

It’s the latest in a series of efforts to lay the groundwork for a run at performance ad dollars. The tech company launched cost-per-click buying earlier this year, letting advertisers pay per click rather than per impression. Cost-per-action models, where payment is tied to an actual purchase, are in development. David Dugan, the Meta veteran hired to lead the ads business, was one of the main architects of one of online advertising’s most successful performance ad businesses. The product feed format is where that ambition becomes concrete. It has also launched conversion tracking features, while third-party measurement features are on the way.

OpenAI declined to comment on this article.