OpenAI expands ChatGPT Search with shopping features

OpenAI is rolling out new shopping features for ChatGPT Search, primarily to assist users with their purchases. The search engine, which debuted last year, has been gaining popularity on a steady basis.

OpenAI says that it saw 1 billion web searches on ChatGPT in the last week alone. The chatbot is used by millions of users who access it to research about various things, including products that they like to buy. But, ChatGPT did not offer the same experience that a user would get on Google Search. This is something that the company has addressed with its latest update.

ChatGPT now displays improved product results with visual details which is fairly similar to the product cards that you see on Google, Bing, etc. It shows the images, and pricing of the product, and also shows star-ratings aka user reviews for the product from popular sites like Amazon, Best Buy, Reddit, to help you decide whether it could be worth buying. In addition to this, ChatGPT Search's shopping feature also includes direct links to buy the product that you queried about.

OpenAI's product lead, Adam Fry, told The Verge that these results don't have any ads and that the results are not sponsored. The design resembles Google's product cards, because it is a format that users are familiar with. OpenAI is working with partners to show up-to-date prices, though it is unclear who these partners are. While a demo of the feature, featured in the screenshot, seems impressive, the actual results could vary.

The AI-powered search engine reportedly failed to deliver accurate information about Nintendo Switch 2, and worryingly, also displayed results from some rather unscrupulous sources, and also included an overpriced eBay listing, likely from a scalper. This indicates that the feature still needs refinement.

ChatGPT's shopping improvements are rolling out to Plus, Pro, Free, and logged-out users everywhere.

Thank you for being a Ghacks reader. The post OpenAI expands ChatGPT Search with shopping features appeared first on gHacks Technology News.