Myntra has introduced ‘MyFashionGPT’, powered by ChatGPT, to enable natural language-based queries for seamless product discovery and get various options across related categories to complete looks.
‘MyFashionGPT’ enables shoppers to search for specific outfit needs by typing text in a manner that closely resembles natural speech. Based on the nature of the query, customers will be shown up to 6 ensemble options across multiple categories from top wear, bottom wear, footwear, accessories to makeup.
For instance, under women, when asked for “what can I wear for a wedding in Jaipur”, the feature will show multiple results that include sarees, blouses, salwars, jhumkas, lipsticks, and footwear from Myntra’s collection of over 2.1 million styles.
Also read: Why Indian businesses are rushing to adopt ChatGPT
While interacting with MyFashionGPT, users can ask for outfit queries based on popular events like IPL, travel locations like beaches, celebrity looks from popular movies, occasions like weddings and festivals and many more. The AI-based assistant can keep refining its results as per the follow-up queries posted by the user.
Developed in-house by the Myntra team,the new product discovery feature is powered by OpenAI’s large language model, ChatGPT 3.5. The user search is sent to the ChatGPT model to fetch the looks for the user, creating appropriate prompts. The ChatGPT response is then processed by Myntra’s search ecosystem to show curated lists of products for the selected look to the user. ChatGPT’s recommendations use internet-trained datasets and are based on the semantic understanding of the user’s query.
Also read: For Myntra, 55% of orders come from Tier 1, metros in beauty category: CEO
“It is a special launch as we are arguably the first fashion, beauty and lifestyle platform, globally, to roll out this feature to the entire customer base at this scale,” said Raghu Krishnananda, Chief Product and Technology Officer, Myntra, on the launch of MyFashionGPT. Adding, “In our journey to democratise fashion using technology, this latest innovation will empower our customers to express their fashion needs and allow them to choose looks from over 2 million styles. We believe this feature will redefine how customers will explore and embrace fashion, making every wardrobe choice a statement of individuality.”
Going forward, the plan is to support voice search, conversational interaction just like talking to a sales assistant, and personalisation of outfit recommendations. For example, if the user is searching for a party attire, the outfit recommendations will use the customers’ previous shopping history, styles and brand preferences and the price points, in addition to allowing users to further refine the results based on additional inputs in a conversational style.