Short answer. You cannot upload a Google Shopping feed to OpenAI as it is, but you can convert it in an afternoon. OpenAI reads JSONL, CSV or TSV rather than XML, renames most fields (id to item_id, link to url, image_link to image_url, item_group_id to group_id), requires one field Google never asked for (seller_name), and spells one availability value differently (pre_order). The values themselves carry over.
The four differences that matter
1. The container. Google’s specification is built around XML (RSS or Atom), with tab-delimited text as the alternative. OpenAI’s specification accepts JSON Lines, CSV and TSV, gzip-compressed or not, and says XML is not supported. A Google XML feed therefore has to be re-serialised, not just re-labelled.
2. The field names. Most fields keep their meaning and change their name; the table below is the whole map. Two fields keep both name and meaning (title, description) and even the same limits, 150 and 5,000 characters.
3. The one new required field. seller_name, the name of the seller supplying the offer, is required on every row and has no Google counterpart, because Merchant Center already knows who you are from the account. It should be the store name shoppers recognise, not a platform handle.
4. One spelling. Availability values match (in_stock, out_of_stock, backorder) except that Google’s preorder is OpenAI’s pre_order, and OpenAI adds unknown. A converter that copies the column untouched fails on the first preorder product.
The field map
| Google Shopping | OpenAI | Note |
|---|---|---|
id | item_id | Same rule: stable and unique per variant. |
title | title | 150 characters in both. |
description | description | 5,000 characters in both. |
link | url | Renamed. |
image_link | image_url | Renamed; OpenAI names JPEG or PNG. |
additional_image_link | additional_image_urls | Renamed; an array in JSONL, comma-separated in CSV. |
brand | brand | Required by OpenAI for every row. |
— | seller_name | New. Required. Your store’s name as shoppers know it. |
availability | availability | preorder becomes pre_order; unknown is added; the rest match. |
price | price | Same “amount CURRENCY” shape. |
sale_price | sale_price | Same idea; update when a sale starts or ends. |
item_group_id | group_id | Renamed, plus listing_has_variations and variant_dict on the rows. |
color / size | variant_dict | Options move into one object keyed by option name. |
— | is_eligible_search / is_eligible_checkout / is_ads_eligible | New. Flags for where the product may appear. |
What to convert and what to add
Converting is mechanical: rename the columns, re-serialise to JSONL or CSV, translate preorder, and move colour and size into variant_dict with group_id and listing_has_variations on the parent. Adding is the part that needs a decision: seller_name for every row, and the eligibility flags, which say whether each product may appear in search, be bought inside ChatGPT (only with the checkout integration enabled) and be used in ads. A feed with everything converted and no flags set is valid and does nothing.
The defects carry over too. A Google feed with missing images, placeholder brands or ungrouped variants converts into an OpenAI feed with the same problems under new names, so run the free feed check on the Google feed first and fix the source; both files inherit the fix. FeedRobin renders the OpenAI JSONL and the Google XML from one catalogue, with the renaming, the availability spelling, seller_name and the variant fields handled in the renderer rather than in a spreadsheet.
Questions merchants ask
Can I upload my Google XML to OpenAI and let it convert?
No. OpenAI’s specification states XML is not supported. The file has to be JSONL, CSV or TSV with OpenAI’s field names. A converter, a feed tool or a spreadsheet does the renaming; the values mostly survive unchanged.
Which value breaks most often in the conversion?
Availability. Google writes preorder as one word; OpenAI writes pre_order. Everything else in the four-value set matches, so a feed converted by renaming columns is fine until the first preorder product.
Do I have to give OpenAI a seller_name if my brand is my store?
Yes; the field is required on every row regardless. Use the store name a shopper would recognise, and use the same value on every row.
Does Google’s Merchant Center approval carry over to ChatGPT?
No. The two programmes are independent: OpenAI has its own application and its own validation of the file. A clean Google feed is a head start on the data, not on the approval.
Sources
OpenAI’s products feed specification and its product feeds for ChatGPT Ads; Google’s product data specification and availability attribute page.