OpenAI’s new feature turns ChatGPT into a personalized shopping assistant: describe what you need, answer follow-up questions, and get a curated buyer’s guide – not just product cards, but expert-style comparisons and tradeoffs.
OpenAI has introduced shopping research, a dedicated mode in ChatGPT that helps users make informed buying decisions. Rather than scrolling through dozens of websites, you can explain your needs – for example, “a compact cordless stick vacuum” or “a gift for a 10-year-old who loves drawing” – and ChatGPT will ask clarifying questions to refine its shopping criteria.
Once clarified, the system researches broadly – browsing trusted retailers, checking specs, reviews, prices, and availability – and builds a personalized buyer’s guide. It shows top recommendations, tradeoffs, and why certain products may be a better fit based on your stated preferences.
How ChatGPT shopping research works
Users can initiate shopping research directly in conversation, or explicitly select the “shopping research” tool via the (+) menu. During the process, you can mark items as “Not interested” or “More like this,” which helps the system adapt the suggestions in real time. After a few minutes, you receive a structured summary with product options, key feature comparisons, and links to retailer sites.
Tech and trust
Shopping research uses a specialized version of GPT-5 mini, fine-tuned for shopping tasks. According to OpenAI, it was trained to read trusted product sites, cite sources accurately, and synthesize information from multiple retailers.
It’s designed, so the claim, with transparency in mind: OpenAI says it only reads publicly available retailer data, avoids low-quality sites, and does not share your chat data with merchants. OpenAI warns of some limitations: pricing, stock, or availability may change rapidly, so always double-check on the merchant’s website.
Agentic commerce has arrived
The feature targets complex purchase decisions – categories like electronics, home appliances, kitchen gear, and even sports equipment — where trade-offs matter. For consumers, this reduces the time spent comparing products across sites and empowers them with a curated, trustworthy guide. For OpenAI, it’s a move towards agentic commerce: blending conversational AI with commerce in a way that could disrupt traditional e-commerce search.
Strategic context
This launch follows earlier updates to ChatGPT Search that added rich product cards (with images, reviews, and price links) across web, mobile, and even non-logged-in users. With shopping research, OpenAI is leaning into the idea of ChatGPT as not just an assistant, but a platform for product discovery and decision-making, especially during the holiday season. The rollout includes nearly unlimited usage through the holiday period for all users – Free, Plus, Pro, and Go – to help attract users during peak shopping time.
Limitations
Whilst the technology is advanced, it’s still possible to surface incorrect or outdated information – especially for fast-moving inventory or new product launches. The feature may favor more established retailers or listings that are well-indexed; niche or small brands may be harder to surface. There’s no built-in checkout (yet) for all merchants – whilst ChatGPT may provide links to retailers, you often still need to complete the purchase on the merchant’s site.
The bottom line
AI is becoming a deeper part of the consumer decision-making process, potentially reshaping how shoppers discover and compare equipment, apparel, and accessories – not just a query interface.
