OpenAI has introduced a new “shopping research” feature in ChatGPT — an AI-driven experience designed to help users find suitable products more efficiently. Rolling out on web and mobile for all logged-in users across Free, Go, Plus and Pro plans, the tool aims to streamline product discovery and comparison during a period of heavy seasonal shopping.
A more conversational way to choose products
According to OpenAI, hundreds of millions of people already use ChatGPT to explore and understand products. The new tool builds on that behavior by offering deeper, more structured assistance for complex purchasing decisions. It can handle tasks such as comparing detailed specs, identifying trade-offs and filtering products to match budgets or preferences. Users describe what they need — for example, a quiet cordless vacuum or a specific type of stroller — and the system asks clarifying questions before returning a personalized buyer’s guide.
OpenAI says the feature draws on up-to-date information from “high-quality sources,” pulling in details like pricing, availability, reviews, specs and images. It also adapts in real time based on user actions such as marking items as “Not interested” or requesting “More like this.” The company positions this as a way to simplify research-heavy categories including electronics, home and garden, appliances, beauty, and sports equipment.
Shopping research also integrates with ChatGPT Pulse — a personalized assistant currently offered to Pro users. When relevant, Pulse may preemptively suggest buyer’s guides based on past conversations, such as highlighting accessories if a user has been researching e-bikes.
What powers the tool
The feature is built on “GPT-5 mini,” a model trained with reinforcement learning specifically for product research. OpenAI states that it designed the system to read trusted retail websites, cite reliable sources and synthesize information from multiple pages. The tool runs iterative updates during the conversation, adjusting recommendations as users add constraints or react to suggestions.
The company emphasizes that chats are “never shared with retailers,” and that all results come from publicly accessible retail websites rather than paid placement. Merchants who want their products included must request allowlisting.
Potential drawbacks and limitations
Despite the intended improvements, OpenAI acknowledges limitations. The tool may still present inaccurate details, including outdated prices or availability information, and users are encouraged to check retailer sites directly for confirmation. The model, while outperforming others on OpenAI’s internal evaluations for product-detail accuracy, “is still not perfect,” the company notes.
There are also broader questions for the ecommerce sector:
- AI-generated comparisons may reduce direct traffic to retailer websites, potentially impacting merchants that rely on search-driven product discovery.
- Smaller brands not on the allowlist could find themselves excluded unless they actively apply, raising visibility and fairness considerations.
- With personalized recommendations influenced by ChatGPT memory, transparency around data usage will remain an area of close scrutiny.
Looking ahead
OpenAI frames the launch as “just the start,” outlining plans to expand coverage into more product categories and refine the system’s understanding of user preferences. The company also signaled upcoming support for direct purchases inside ChatGPT for participating merchants — a shift that could position ChatGPT as a deeper ecommerce intermediary.
Why this matters to ecommerce merchants
For online retailers, ChatGPT’s shopping research tool introduces both opportunities and potential risks. It could streamline product discovery for consumers, increasing purchase confidence and improving conversion rates for merchants whose products appear in guides. But it also creates a new layer between shoppers and brands, meaning visibility may depend on how well a merchant’s content is understood — and indexed — by AI systems. Even beginners in ecommerce should track these developments: tools like this may quickly reshape how shoppers compare products and where they begin their buying journey.
Get ecommerce news you can use
Sign up below and we’ll send you email digests of all the latest news and developments in the world of ecommerce, tips on how to grow an online business, and special offers.
Matt Walsh is the Head of Content at Ecommercetrix. He also creates video content for the site and its associated YouTube channel.
Having gained a degree in Literature and Philosophy from Trinity College Dublin and an MA in International Relations from Dublin City University, Matt worked in the marketing world for a decade before taking up an education role — teaching languages, literature and creative writing in London.
Matt is an Adobe Certified Associate in Photoshop and Illustrator, and also holds a certificate in digital marketing from DMI (Digital Marketing Institute), along with a postgraduate certificate in Design Thinking, Innovation, Entrepreneurship and Creative Thinking from Trinity College Dublin.