How Amazon is using AI, from Rufus to movie recommendations

AI For Business


Amazon has announced an AI chatbot called Rufus.
Amazon

  • Amazon has started implementing AI into various aspects of its business.
  • AI can offer advice on everything from what size shoe to buy to determining how to minimise packaging waste.
  • Here are six ways Amazon is using AI today.

If you've ever received a parcel from Amazon or bought a sweater from the company's website, there's a good chance you've been impressed with the tech giant's AI efforts.

Chatbots like OpenAI's ChatGPT are one of the most visible forms of AI, but Amazon is incorporating this hot technology into many corners of its business — and many of its uses impact Amazon customers even if they aren't immediately visible in stores like Whole Foods or when browsing Amazon.com.

Here are six examples of how Amazon is incorporating AI into its consumer business.

Amazon's Rufus lets you ask shoppers for product recommendations

Amazon's Rufus chatbot.
Amazon/Screenshots

Sorting through hundreds of product listings on Amazon can be a daunting task, and in February, Amazon introduced Rufus, a chatbot assistant that aims to make shopping a little more user-friendly.

Instead of using a traditional search function, shoppers can ask Rufus questions about what products they need for a particular trip, or what qualities they should consider when making a big purchase.

A person familiar with the project told Business Insider last year that Rufus was meant to act like a savvy salesman, and BI reported that the project was code-named “Project Nile” within Amazon.

Last week, Amazon announced that Rufus was now available in its shopping app for all US customers.

Amazon says early users are not only asking Rufus for details about specific products, but also asking the chatbot broader questions like, “What's the difference between a gas and a wood-fired pizza oven?”

Amazon Fashion uses AI to make sure the clothes you buy will fit

Amazon in January rolled out several AI features aimed at making ordering clothing on its website easier, particularly reducing the chances of having to return new clothes because they don't fit.

One feature is an algorithm that takes into account several factors, such as whether a brand's sizes tend to be smaller or larger than normal.

The algorithm can also offer recommendations based on a shopper's past purchases: “For example, if you bought a certain size pair of kids' pants this month, the algorithm will consider that you may need a larger size in the coming months,” Amazon said.

Amazon is also using AI to create highlights about a garment's fit, fabric quality, and other aspects based on customer reviews. Size charts on a product's Amazon page also reflect AI data.

Meanwhile, apparel brands can use Amazon's Fit Insights Tool to get information about sizing issues based on returns data.

“Project PI” aims to reduce the number of damaged or incorrect products shipped by Amazon

Amazon warehouse.
Sean Gallup/Getty Images

Amazon's “Project PI” (short for “Private Investigator”) aims to use AI to reduce the chances of finding a ripped bag of pet food in your next order.

The strategy involves sending items through an image tunnel at an Amazon fulfillment center, where Amazon's AI detects if the item is damaged or incorrect and signals that it needs to be replaced. Warehouse workers then review the call and decide what to do with the damaged item.

“Project PI uses a combination of generative AI and computer vision techniques to spot defects, like damaged products or products that are the wrong color or size, before they reach customers,” Amazon said in a June blog post about the project.

Amazon's Packaging Decision Engine Reduces Waste

Amazon branded boxes and envelopes.
Marco Bertolello/AFP via Getty Images

Amazon uses AI to reduce packaging waste: Remember when there were small items inside giant boxes?

This approach allows Amazon to use cardboard and other packaging materials more efficiently and ship products in smaller quantities.

“Data scientists trained models to understand various product characteristics, such as product shape and durability, and to analyze customer feedback on the performance of different packaging options,” the company said in a February blog post.

AI makes recommendations on Amazon's Fire TV

Amazon isn't just using AI for shopping: this year it added an AI search feature to Fire TV. If you want to spend less time searching for something to watch, you can ask it for recommendations for movies and TV shows in your favorite genres.

The search option will also be useful for more specific requests, the May blog post said, such as “What movie has the line, 'Smalls, are you going to kill me?'”

In a post, Amazon said the AI ​​application allows users to “find streaming content recommendations the same way you would ask a friend who has an encyclopedic knowledge of TV shows and movies.”

Fire TV owners can also ask Alexa to create AI-generated images.

Amazon's Maestro creates playlists

In April, Amazon announced that it would be offering its AI playlist builder, Maestro, for beta testing to some Amazon Music users.

Amazon says Maestro can create playlists based on user input, such as “Myspace-era hip hop” or a single emoji.

Do you work in Amazon Flex, Whole Foods, Amazon Fresh, or any other part of Amazon's retail business and have a story idea you'd like to share? Contact this reporter. contact address



Source link

Leave a Reply

Your email address will not be published. Required fields are marked *