AI Search News · NEWS

ChatGPT Shopping Adds Virtual Try-On and Saved Products: An APAC Merchant Guide

ChatGPT Shopping now supports virtual try-on for clothing and accessories and lets users save products in Library. OpenAI also explains the product and merchant data used in shopping results.

Content typeNEWS
MarketGlobal
Updated2026.10.30
PublisherAI Search Lab
AI shopping journey from product discovery through virtual try-on, saving and merchant choiceNEWS · AI SEARCH LAB
Answer

OpenAI’s official shopping help page, updated October 29, 2026, says people can virtually try on clothing and accessories and save products to Favorites or folders in ChatGPT Library. Shopping results can use the query, conversation context, Memory, Custom instructions, structured product data and public third-party content. OpenAI states that organic product results are selected independently from ads and are not influenced by partnerships.

What changed in ChatGPT Shopping?

A Try on button can appear on listings for clothing and accessories. The user takes or uploads a selfie, and ChatGPT Images creates a visualization. A clothing image can also be uploaded directly in a conversation. OpenAI warns that the result may not represent the product or the person’s appearance exactly and does not guarantee size or fit.

Reference photos can be kept for later try-ons and managed under Personalization. Products can be bookmarked into Favorites or organised in Library folders. The official page describes these capabilities as available on mobile and web.

If your customer moves from a generated image to purchase, the merchant page still needs accurate measurements, material details, final price and returns. A polished visualization cannot replace these transaction facts.

How are shopping results separated from ads?

OpenAI says product results are not ads and are not influenced by its partnerships. ChatGPT Ads runs separately. An advertiser should not treat paid delivery and organic product discovery as one visibility metric.

The model can interpret budget, preferences, use case and restrictions from the conversation. Memory or Custom instructions may also influence relevance. Users can refine the conversation when a result misses their preferences.

Track organic product appearances, sponsored placements and answer citations separately. Connect them at the product ID and destination level when visits, carts and purchases are measured. That makes it easier to see whether the constraint sits in product data, paid eligibility or public content.

What product information can ChatGPT use?

The help page lists structured metadata from first-party and third-party providers, including price and product descriptions, along with other public third-party content. The model response produced before new search results and OpenAI safety policies can also affect the output. A stated budget can raise the importance of price, while a detailed use case can make suitability more important.

ChatGPT may simplify titles and descriptions from different merchants. Labels such as Budget-friendly or Most popular can be model-generated and are not guarantees or a complete market assessment. Review summaries come from public reviews, and OpenAI does not verify the reviews or ratings.

APAC product teams should keep brand, model, variant, material, dimensions and restrictions consistent across destinations. Large differences in naming or identifiers can make it harder to compare the same item across merchants and markets.

How are prices and merchants presented?

The price in the first response usually comes from the first listed merchant and may not be the lowest available price. A product detail view can show additional sellers, and a Best price label may appear for the lowest option ChatGPT knows. OpenAI warns that price and shipping updates can take time to appear.

Product data, reviews and user conditions flowing through a comparison process to merchant choices
ChatGPT can combine product data with conversational context, while buyers should still confirm price, stock and returns on the merchant page.

Merchant order can use availability, price, quality and whether the seller is the maker or primary merchant. The list can draw on direct merchant data and third-party metadata. OpenAI expects the ranking process to evolve and become more personalised.

Check a regional price today. Taxes, fees, shipping, stock, colour and size choices, returns and warranty should reflect the same effective date. A stale feed can place an attractive price in the comparison even though the checkout page has already changed.

What should APAC merchants prepare?

Start with stable product identifiers. Maintain SKU, GTIN, brand, model and variant rules across owned stores and marketplaces. Put an update time on price, stock, shipping coverage and return windows. The structured data and visible page should report the same facts.

For fashion and accessories, provide a detailed size table and material description because virtual try-on does not guarantee fit. Front, back, side and material-detail imagery can help buyers verify what the generated view cannot establish.

OpenAI says Shopify merchant data is already integrated through Shopify Catalog, so individual Shopify merchants do not need additional integration work. That does not remove the need to maintain the source catalog. Other merchants can review the Product Feed documentation and apply for direct-feed access.

Regional governance for Korea, Japan and wider APAC

A global team can share product identity and asset standards across Korea, Japan, Taiwan, Hong Kong, India and Australia. Currency, tax, delivery, returns, legal entity and support language still require local verification. A translated global page rarely completes every local buying task.

Japanese companies entering Korea should align the headquarters catalog with Korean won pricing, local stock, Korean-language measurements and the Korean seller of record. If the same product appears under unrelated titles or identifiers, merchant comparison and analytics become less reliable.

Healthcare organisations should not transfer a fashion try-on use case to medical devices or treatment outcomes. Health products may be subject to separate law and platform policy. This article is not legal or medical advice.

What changes for GEO, AEO and SEO?

The official explanation confirms that structured product data, public web content and conversational context can all contribute to product selection. It does not support a promise that one schema type or repeated keyword will secure placement. Merchants need accurate source pages, current feeds, identifiers and transaction terms.

Try-on and saved products also mean the buyer may not click to a merchant immediately. People can compare and save inside ChatGPT, then return later. A first-click report can miss part of that journey, so teams need product-level measurement for repeat visits and later conversion.

Open a product page now. Can ChatGPT simplify the title without losing brand and model identity? Can the buyer verify the final price if a feed is delayed? Is the reason to choose the owned store visible when several merchants offer the same item?

AI Search Lab view

AI shopping now covers discovery, comparison, visual trial, saving and merchant selection. Commerce teams should focus on product-data accuracy, useful imagery, sizing and the update path across sellers rather than guessing a recommendation phrase. Paid ads and organic product results still belong in separate reports.

LeadGenLab’s GEO, AEO and AI commerce services review query sets, product feeds, detail pages and merchant information together. Global teams can scope a Korea-first or wider APAC programme through the project contact.

Official sources

Checked October 30, 2026 against OpenAI’s official help page and developer documentation.

FROM NEWS TO EXECUTION · LEADGENLAB

분석을 실제 실행으로 연결하려면

AI Search Lab은 공식 출처와 편집 기준에 따라 변화와 실행 기준을 설명합니다. 진단, 기술·콘텐츠 개선, 산업별 GEO·AEO 지원이 필요하다면 운영사 LeadGenLab의 관련 서비스와 상담 페이지에서 다음 단계를 확인하세요.