Meta tests shopping research feature in its AI chatbot

The chatbot responds with a carousel of product images that include captions with information about the brand, website, and price

Meta Platforms Inc. is testing a shopping research feature in its artificial intelligence chatbot, rivalling a similar tool offered by OpenAI’s ChatGPT and Google’s Gemini.

AI chatbots are quickly morphing from Q&A tools into discovery engines. The feature, which allows requests for product suggestions, is being rolled out to some US-based users of the Meta AI web browser. The chatbot responds with a carousel of product images that include captions with information about the brand, website, and price. It also offers a brief explanation of its recommendations in bullet-point form. A Meta spokesperson confirmed that the shopping tool is being tested, but declined to share further details.

Chief Executive Officer Mark Zuckerberg has made it a goal for Meta to build “personal superintelligence” as it competes with other popular chatbots like ChatGPT and Gemini, which have also begun incorporating e-commerce features to make money from their tools. During an earnings call in January, Zuckerberg said the company will start shipping new products in the coming months that can show Meta’s ability to provide a “uniquely personal experience” based on people’s history, interests, content and relationships.

When applicable, the chatbot’s recommendations are tailored to what Meta already knows about the user’s location and to the gender it infers from their name, Bloomberg News found when testing the feature. For example, when asked to find puffer jackets, Meta AI’s response referenced the author’s location in New York and offered options for women’s puffers. There is no checkout or payment option within Meta’s chatbot, but users can click on the provided merchant links for further browsing.