ChatGPT ChatGPT launches Shopping Research

[Dopious]

OpenAI is launching Shopping Research today, November 24, 2025, a new feature in ChatGPT that helps users find and compare products online. Instead of searching and researching sites, filtering, and reading tons of reviews, users can describe their needs and ChatGPT attempts to build a complete, personalized purchase recommendation in minutes.

The launch is rolling out globally on both mobile and browser for all logged-in users, regardless of whether they are using Free, Go, Plus or Pro.

What is ChatGPT Shopping Research?

ChatGPT Shopping Research is designed for the more complex decisions, where budget, features, specifications, and personal preferences play a big role.

Shopping Research analyzes needs in several steps:

1. The user describes what they are looking for.
   Example: “Which quiet cordless vacuum cleaner is best for a small apartment?” or “Help me choose between these three bikes.”
2. ChatGPT asks follow-up questions
   Budget, target group, size, materials, functions or similar.
3. The model does research based on the entire web.
   It retrieves price, stock status, specifications, reviews and images in real time from reliable sources.
4. The user guides the process.
   Products can be marked as “not interesting” or “show more similar”, which fine-tunes the result.
5. A personal buying guide is created.
   The guide summarizes the products' strengths, weaknesses and current details from various retailers.

According to OpenAI, this is significantly more in-depth than a standard ChatGPT search and is particularly strong in electronics, home & garden, beauty, kitchen, sports and outdoor activities.

What is the technology behind ChatGPT Shopping Research?​

Under the hood, Shopping Research is powered by a specially trained variant of GPT-5 mini, optimized for:

• to read and understand product data from reputable sources
• to compare many attributes at the same time
• to cite and merge information from multiple websites
• to dynamically adapt recommendations based on feedback

OpenAI has developed a new internal benchmark for product accuracy where the Shopping Research model performs highest among all their mini-models.

What does Shopping Research mean for e-retailers?​

OpenAI emphasizes that for e-retailers, Shopping Research means:

• no chats are shared with resellers
• results are organic, based on open sources
• low-quality product catalogs are avoided
• retailers can request allowlisting to guarantee visibility

The model may still miss detailed information, such as price or stock status, and the user is advised to verify the information with the retailer.

Source: https://openai.com/index/chatgpt-shopping-research/

 

[29dogs]

The launch is rolling out globally on both mobile and browser for all logged-in users, regardless of whether they are using Free, Go, Plus or Pro.

Hmm makes you wonder if Amazon or someone paid for them to add this feature in.

Still I'm excited to try it out!

 

[chaktty]

There may be some lags in this, especially when products' prices change, and AI still uses old data.

Nevertheless, it's a good idea and precursor to a full blown shopping integration.