- Product results aren't ads, and that's the point
- How ChatGPT picks the products you see
- Labels, summaries and reviews: what they mean
- Why the first price you see may not be the lowest
- Virtual try-on, reference photos and Favorites
- A shopping routine that avoids the traps
- Is ChatGPT shopping worth using?
- Frequently asked questions
ChatGPT shopping is the set of product results ChatGPT shows when your question sounds like you want to buy something, with images, product details and links to the sites where you can find out more or buy. As of October 2026, OpenAI says those results are chosen independently by ChatGPT, aren't ads and aren't influenced by OpenAI partnerships, and you can virtually try on clothes and save products to your Library on mobile and web.
You don't switch anything on. Ask something like OpenAI's own example, help me find costumes for my dog, and ChatGPT decides it's a shopping question and shows a carousel of options. From there you can compare, refine and click through to a merchant.
Product results aren't ads, and that's the point
This is the part most people get wrong. OpenAI states that product results are selected independently by ChatGPT and aren't ads. Ads in ChatGPT are a separate thing and sit apart from the product results. If you want to know how the paid side works, my guide to ads in ChatGPT covers it.
That doesn't mean the results are neutral in every sense. They're shaped by what ChatGPT knows about you, by product data from other companies, and by OpenAI's own policies. Knowing what goes into them helps you read them better.
How ChatGPT picks the products you see
A product appears in the carousel when ChatGPT thinks it's relevant to what you want. It looks at your question and your context, which includes Memory and Custom instructions. So if you've told ChatGPT you dislike clowns, it may leave clown costumes out of a costume search, even if you'd have been open to one this time. If you want to check what it remembers about you, here's how ChatGPT memory works.
OpenAI lists three inputs when it decides which products to surface:
- Structured metadata from first-party and third-party providers, such as price and product description, plus other third-party content.
- Model responses ChatGPT generates before it looks at any new search results.
- OpenAI's safety standards and product policies.
For a goofy dog costume it may weigh options, price, reviews and ease of use. Give it a budget and it focuses more on price. Leave price out and it may focus on other things. Not every available product is shown, so OpenAI's own advice is to check that a product meets your needs before you buy.
Labels, summaries and reviews: what they mean
Merchants often use different titles for the same item, so ChatGPT may write simplified titles and descriptions from third-party data to make results easier to read. Three other things are worth knowing:
- Labels such as Budget-friendly or Most popular are generated by ChatGPT. They aren't guarantees or verified statements. Budget-friendly might mean reviewers often mention good value, not that it's the cheapest.
- Review summaries are written by the model from reviews on public websites, to pull out common likes and dislikes.
- Reviews and ratings aren't verified by OpenAI.
Why the first price you see may not be the lowest
Prices come from third-party providers. The price in ChatGPT's first answer usually reflects the first merchant listed, which may not be the lowest available. Open the product and you may see other merchants selling it at the same or different prices, and ChatGPT tries to highlight the lowest price it knows about there. When more than one merchant sells it, that option may carry a Best price label.
Prices and shipping terms can also lag. When a merchant changes them, there may be a delay before ChatGPT shows the update.
Merchants are listed from merchant and product data, and ranked on things like availability, price, quality and whether they're the maker or main seller. Shopify merchants are already included through Shopify Catalog, with no extra work for each shop. Other merchants can apply to send OpenAI a direct product feed.
Virtual try-on, reference photos and Favorites
| Feature | How it works, as of October 2026 |
|---|---|
| Try on | A button on clothes and accessories listings; take or upload a selfie and ChatGPT Images creates the try-on |
| Try on from a photo | Upload a picture of clothing or accessories in a chat and ask how it could look on you |
| Reference photos | Saved for future try-ons; change or delete them in Settings, Personalization, Reference photos |
| Favorites and folders | Select a product's bookmark icon to save it, or create a folder in your ChatGPT Library |
| Availability | ChatGPT on mobile and web |
Try-on images may not show the product or your appearance exactly, and they don't guarantee fit or size. Check the merchant's measurements, product details and return policy before you order. To remove a saved product, select its filled bookmark icon.
Want AI doing the heavy lifting in your marketing?
I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.
A shopping routine that avoids the traps
- Ask for the product with your budget, preferences and any must-haves in the first message.
- Ask ChatGPT to compare the options, explain the trade-offs or suggest alternatives, and tell it to adjust if the results miss.
- Use Try on for clothes and accessories, or bookmark products to Favorites to come back to.
- Open the product and look at every merchant listed, not just the first price.
- On the merchant's site, confirm the final price including taxes, fees and shipping, plus availability, size or colour options, and the return and warranty policies.
If something looks wrong, use the thumbs-up or thumbs-down icon on the product, or report it from the product menu.
I'd pair this with ChatGPT Error Messages.
Is ChatGPT shopping worth using?
I'd use it to narrow a long list down to three or four options, especially when you can describe what you want in plain words and need the trade-offs explained. Try-on is fun for a quick feel of colour and style. I'd skip treating it as a price checker, because the first price is often just the first merchant, and labels such as Budget-friendly are ChatGPT's own words. Do the research in ChatGPT, then do the buying on the merchant's page with your eyes open.
To see how it all fits together, read the ChatGPT Age Verification guide.
This guide is part of my ChatGPT Prompts and Limits: 30 Guides to Fun Prompts, Photo Edits and Daily Caps.
Frequently asked questions
Are ChatGPT shopping results ads?
No. OpenAI says product results are selected independently by ChatGPT, aren't ads and aren't influenced by OpenAI partnerships. Ads are separate from product results.
How does ChatGPT virtual try-on work?
Select the Try on button on a clothes or accessories listing, take or upload a selfie, and ChatGPT Images creates your try-on. It doesn't guarantee fit or size.
Does ChatGPT show the lowest price?
Not always at first. The price in the first answer usually reflects the first merchant listed. ChatGPT tries to highlight the lowest price it knows about on the product detail page.
Where do my saved products go?
Products you bookmark go to Favorites, or to a folder you create, in your ChatGPT Library.
Which merchants appear for a product?
ChatGPT lists merchants from merchant and product data and ranks them on factors like availability, price, quality and whether they're the maker or main seller.
