"The new tool is built on Open AI's latest generative multimodal AI, which understands natural language and semantics, and can also parse images. The brand says the tool, which has been in development for a year, has been trained solely on the brand's own imagery and content. This includes product imagery from Ralph Lauren's online product catalogue archive, PDP imagery, editorial campaigns, style guides and brand language. The tool only recommends looks to customers based on inventory that is available in real time - a feature the brand says distinguishes it from large tech companies' universal AI shopping tools, which can sometimes recommend unavailable products."
"Starting today, when customers log into the Ralph Lauren app, they'll be presented with a white search bar and the prompt "ask me anything", where they can begin their "intent-based" search in natural language - for example, "I need an outfit for my cousin's wedding in Arizona this spring." The tool then responds with follow-up prompts to help refine its recommendations. Like all generative AI tools, the more detail a customer shares in the open-ended conversation, the more relevant the tool's product recommendations, which are ultimately presented to customers in an interactive shoppable carousel of looks."
Ralph Lauren deployed a generative multimodal AI tool that understands natural language and image content and was trained exclusively on the brand's own imagery and language. Training sources include the product catalogue archive, PDP imagery, editorial campaigns, style guides, and brand language. The tool only recommends items that are available in real time, preventing suggestions of out-of-stock products. Customers access the feature in the Ralph Lauren app via an "ask me anything" intent-based search, receive follow-up prompts to refine results, and view personalized looks in an interactive shoppable carousel. The feature currently covers the Polo catalogue with plans for broader rollout.
Read at Vogue Business
Unable to calculate read time
Collection
[
|
...
]