KEYWORDS: ChatGPT, GPT-4, OpenDataset, FashionDatasets, EcommerceDatasets, PersonalizedFashion
The fashion industry is consistently at the forefront of adopting new technologies and adapting to changing consumer tastes.
With the rise of Artificial Intelligence (AI) and Machine Learning (ML), the fashion industry is undergoing a revolution in personalization.
In this case, maadaa.ai has already written an article about the affection of AI in personalization, please click the link below to read more.
However, the impact of AI on fashion personalization has recently evolved with the launch of ChatGPT last November.
ChatGPT, developed by OpenAI, is an open-source artificial intelligence chatbot based on the GPT-4 system that can significantly impact the personalization of the fashion industry by leveraging its Natural Language Processing (NLP) capabilities, as well as its massive amount of data sources.
Here are some application scenarios of how ChatGPT/GPT-4 can impact the personalization of the fashion industry, such as virtual styling, tailored product recommendations and personalized customization.
Virtual Styling
Several fashion companies have already integrated GPT-4 into their websites or mobile apps to provide personalized styling recommendations, including H&M, Tommy Hilfiger, Adidas, Nordstrom, Saks Fifth Avenue, etc. [1]
Learn more about AI-driven virtual styling:
So the question right now is how ChatGPT will perform in virtual styling compared to existing AI stylists.
When asked for ideas for a spring wedding dress, ChatGPT’s text-only stylist responds with long-winded, generic suggestions, making it difficult for shoppers to visualize products and understand fit and style.
Unlike OpenAI, YesPlz.AI’s AI stylist can remember previous shopper conversations and have information at the ready for personalization that is “always fashionable, rooted in what the user prefers,” according to YesPlz.AI.
As YesPlz.AI mentions, in addition to using Open AI’s technology to generate prompt responses, they also use technologies such as AI image tagging, NLP, and Proprietary Fashion Transformer to generate the best product suggestions according to shoppers’ unique styles, needs, and thematic preferences.[2]
Another GPT-driven AI stylist is called MyFashionGPT, launched by Myntra, one of India’s fashion and beauty e-commerce giants.
Based on ChatGPT, as users type their queries, Myntra’s search engine processes and responds to display the curated list of products.
Since ChatGPT doesn’t work with images, all suggestions and responses are in text and it’s up to the user to visualize the rest. This is where Myntra comes in by providing visual suggestions.
So, we believe that as AI technology continues to iterate, virtual stylists will become a part of people’s lives.
2. Personalized Product Recommendations
Let’s take virtual styling one step further.
With more personal information such as size, skin&body type, preferences, previous purchases, and style preferences of users, there is no doubt that ChatGPT-like AI chatbots are able to tailor product recommendations.
For example, Nordstrom’s virtual assistant “Nordy” and Saks Fifth Avenue’s virtual assistant “Saksbot” can help customers find the right size, recommend items to complete an outfit, and provide styling tips.
Beyond fashion brands, there are several AI tools with personalized product recommendation capabilities.
According to Vue.ai, it allows each retailer can customize the AI personalization models that work best for their business. The systems build nuanced style profiles for each shopper by learning from their actions on the site and mapping them to rich product data.
Then AI analyzes and displays the products in the catalog that to meet customers’ needs and interests. [4]
In fact, AI systems like ChatGPT can do more than clothes, shoes, bags and accessories.
Many beauty brands are also using it to recommend beauty products.
By entering the skin type, age, and shopping experience, AI chatbots can recommend personalized beauty products and makeup.
For instance, brands such as Sephora, L’Oréal, and Ulta Beauty use ChatGPT-powered chatbots to enhance the customer experience. All of these virtual beauty advisors can understand natural language input and provide personalized product recommendations.[5]
Virtual fitting rooms, as a remarkable application scenario, which is transformed by ChatGPT/GPT-4.
In fact, even before the release of GPT-3, fashion brands and retailers were using AI technologies to develop virtual fitting rooms. Fortunately, maadaa.ai has covered such a topic, please click the link to read more.
According to a German online retailer Zalando, announced its plans to pilot a virtual fitting room experience across all its 25 markets.
In a dedicated campaign environment, millions of shoppers can create a 3D avatar by entering their height, weight and gender to try on a selection of jeans from different brands. [6]
3. Personalized Customization
There are two parts of personalized customization: Product Customizer and AI models.
Companies like Zakeke, Kickflip and Infinite Options help brands and retailers to add a powerful and complete visual commerce tool for their store, enabling the customers to custom-design any products at their fingertips.
We believe that such a small change will boost sales by making it fun and easy to purchase custom products.
In addition, after getting used to seeing models with similar shapes, personalized custom models are now becoming known with the development of AI technology.
Denim brand Levi’s will test AI-generated clothing models to increase diversity as a “complement to human models.”
According to Levi’s, the AI clothing models could be more body-inclusive, allowing customers to see how an item of clothing would look on a variety of models spanning a wide range of body types, ages, sizes, and skin tones. [7]
In addition, Chinese E-Commerce giant Taobao, owned by Alibaba, has developed an AI virtual model called Taji.
Based on Deep Learning algorithms, Taji has multi-size fitting functions, garment liquefaction deformation, 3D texture mapping, etc. At the same time, Taji has the ability to provide exclusive models, multiple background options and diversified generation capabilities combined with body size for enterprises.[8]
4. Fashion and E-Commerce datasets from maadaa.ai (Open & Commercial)
Trying to leverage your business to embrace personalization in the fashion and E-Commerce industries?
As we all know, the current ChatGPT does not perform well in such an area, not to mention its inaccuracy and ethical issues.
Therefore you need a large amount of high-quality and specific data for your business.
Based on our comprehensive accumulation of Fashion and E-Commerce technologies and application scenarios, maadaa.ai has developed a series of both open datasets and standard datasets, which can help industrial and academic customers accelerate AI innovations in Fashion and E-Commerce.
Please click the links below to find out!
maadaa.ai, founded in 2015, is a comprehensive AI data services company that provides professional data services such as text, voice, image, and video to customers in the AI industry.
From AI data collection, processing and labeling, to high-quality AI datasets and dataset management, maadaa.ai helps customers to efficiently collect, process, and manage data, and conduct model training for rapid and cost-effective adoption of AI technologies. [9]
Reference List:
[1] https://maadaa.ai/blog/chatgpt-for-fashion-industry-new-opportunities-and-challenges/
[2] https://ai.plainenglish.io/what-happens-when-an-ai-personal-stylist-meets-chatgpt-5bb0a38a91dc
[4] https://vue.ai/industries/ai-in-fashion-retail/
[5] https://aicontentfy.com/en/blog/chatgpt-and-beauty-industry-personalizing-product-recommendations
[6] https://internetretailing.net/zalando-virtual-fitting/
[7] https://www.theverge.com/2023/3/27/23658385/levis-ai-generated-clothing-model-diversity-denim
