Clothing is more than fabric: it's how we choose to feel and express ourselves. In that way it is a highly personal endeavor.

Endless aisles in-store and online make shopping time-consuming and often stressful. But fret not, AI is looking into helping us. AI can use our aesthetic and preferences, and tailor the options we are presented.

The evolution of online retail

“It’s been 25 years since Net-a-Porter convinced fashion brands to sell products “without a store”, as the company framed it at the time. The Web1 era — or the first era of the internet — brought radical change to how we shop. Ten years later, more disruption followed the dawn of Web2, driven by social media and smartphones.

Now, as we enter the era of Web3, the next major transition in how we shop is underway. Similar to the first two phases of the internet and e-commerce, it brings new cultural and technical norms.”

- Vogue Business

We are now building the third wave of the internet. This will disrupt how data is created and which data is individually consumed. AI will enable hyper-personalisation and less friction connecting the most aligned products and data with its consumers. This new age will allow us to live more authentically and intentionally, giving us greater control over our daily experience. (Yes, I am an optimist about it :)

A podcast that follows changing retail trend closely: The Future of Shopping

Economic motivation : How much is AI worth to the retail business ?

According to McKinsey analysis, AI could add, conservatively, $150 billion and up to $275 billion to the apparel, fashion, and luxury sectors’ operating profits in the next three to five years.

Investors are looking for companies using AI to further profitability in these sectors.

“Our focus has been identifying solutions that amplify design capabilities and maximise efficiency gains at every step of the design and sales process“, Dana Settle, co-founder and managing partner at venture capital firm Greycroft.

Consumer Experience

1. Personalised Virtual Shopping

The endless aisle problem: When shopping online, you often sift through 20+ pages to find one or two items you like, then you jump between sites or scroll endlessly on social media. How can we make this experience better?

Learned preferences can tailor our shopping experience by showing us relevant products. This can also enable us to discover new brands that match our aesthetic and company values.

This not only enables us to filter out uninteresting options but also gives us frictionless access to discovering new ones. Instead of only seeing products whose marketing push reaches us, we may also discover high quality smaller brands organically, as AI will pull them towards us by understanding our wants. Some experiences we can expect to see in the nearer future:

• A focused, curated webpage generated by AI following each customers individual preference and previous purchase patterns. Even better to walk around a customised virtual store in an Apple Vision Pro shopping app?

• A startup MirrorMirrorAI, learns your aesthetic from your submitted images and recommends clothes for you based on query prompts like “elevated look for work”. The client will receive a personalized catalogue of looks featuring the client themselves wearing the suggested clothes.
  

2. Virtual Try-on

Returns are a big problem for apparel brands : a $218 billion per year problem. In 2020, returned inventory created an estimated 5.8 billion pounds of landfill waste and return shipping emitted 16 million metric tons of carbon dioxide.

Customers often struggle to envision how products will look on their bodies before buying online. An AI-powered virtual try-on and styling tool can significantly improve the shopping experience.

• An estimated 85 per cent of US apparel brands and retailers either use or plan to use virtual try-on tools.

• (2017) Amazon acquired Body Labs, which makes 3D models for virtual try-on.

• (2022) Augmented reality try-on company Wannaby (acq. by Farfetch) launched a pilot with Maison Valentino ready-to-wear.

• (June 2023) Google piloted a tool that enables people to see items of clothing on a range of models and body types

• (November 2023) Balmain partnered with BODS to begin testing virtual fitting and styling
  

3. Product Search by prompt or image

Generative AI tools are evolving to interpret both text and images for more intuitive search results. As such, we can search for specific products using more human-like natural language prompts or even provide reference images from a magazine or social media post when looking for an exact or similar item.

• (Product lookup by prompt) Startup Daydream is working on a platform to enable product discovery using natural language prompts such as “pastel-coloured dresses for a garden party in Charleston” or “high-end suits for a business conference in NY”

• (Product lookup by image) Upload an image from an editorial or social media post asking for the exact dress or for a look inspired by that image. Pinterest in a way offers this, when clicking on an image it can often help you find additional images of the outfit and this sometimes leads to an image with the product link. They also do a great job of finding outfits with similar aesthetic and fabric but varied pattern cuts for more diversified options to select from.

4. Virtual Stylist

An AI powered stylist can learn from the most popular trending looks, learn design principles, styling techniques, visual merchandising, color theories and recommend options to a customer given body shape, skin tone for best colors to wear and any mentioned preferences.

Brand Operation

1. Sustainability

AI can analyze fashion trends, consumer behavior, and social media data to predict future trends and help make informed decisions about inventory needs by area and source product development close to projected markets. These optimisations can even lead to more efficient distribution processes, reducing environmental footprint.

2. Product description

Text descriptions can be generated by using images as a prompt, automating the text for product page and saving time and cost esp for small businesses. For existing businesses these description styles can be learnt by training on descriptions of past products.

3. Campaign Development

AI can rapidly generate campaign content and virtual avatars for various marketing channels - improving efficiency and customization. This can also enable personalized campaigns targeting groups or individuals based on their interests and preferences to increase engagement
.

4. Counterfeit detection

The luxury resale marketplace faces authenticity challenges from counterfiet products. A company Entrupy uses AI to authenticate designer vintage handbags and sneakers in the resale market. Their AI compares these photos against a database to verify authenticity with a claimed accuracy of 99.1%.

Overall, AI has the potential to drive innovation in the retail industry.