Amazon’s Amazing AI
Amazon has amazing AI. When I called to resolve a billing problem and realized that I was talking with an AI, I was really impressed. One of their goals is to be the “Earth’s most customer-centric company,” and to support this, they have developed a few flavours of AI - some for internal customer service and some for outside organizations like NASA through their AWS operations.
Not a Chatbot

Most text-based online customer service systems feature automated agents that can handle simple requests. Typically, these agents are governed by rules, rather like flow charts that specify responses to particular customer inputs. If the automated agent can’t handle a request, it refers the request to a human customer service representative. It can be a frustrating experience.
Amazon started phasing in automated agents that use neural networks rather than rules. These agents can handle a broader range of interactions with better results.
Deep Learning
Neural networks are the core of deep learning algorithms. Their name and structure are inspired by the human brain, mimicking the way that biological neurons signal to one another.
Deep learning uses multilayer neural network architecture which focuses on the use of data and algorithms to imitate the way that humans learn, gradually improving accuracy. It drives AI applications and services that improve automation, performing analytical and physical tasks without human intervention.
Examples of deep learning driven products and services include digital assistants, voice-enabled TV remotes, credit card fraud detection as well as emerging technologies such as self- driving cars.
Three Pillars
AI and machine learning powers three popular Amazon products: Alexa, the Amazon Go Store, and the Amazon recommendation engine.
Data from these three main pillars of the company work together to create a cohesive customer experience. A customer can visit the Amazon Go store to get a few items for dinner, ask Alexa to look up a recipe and the product recommendation engine can determine that the customer likely needs to purchase a certain type of sauce pan. This cooperation delivers a customized and cohesive customer experience.
My Experience
When I called Amazon customer service to resolve a billing error, I was expecting to wait quite some time. Thanks to their AI, the wait was not long.
The first thing I had noticed was that the bot’s voice was a lot smoother than I expected, not mechanical at all. By the way it responded to my answers to its questions, It felt like I was speaking with a real person.
Instead of asking me to choose from a list of options of why I might be calling, it simply asked me to describe my problem. It then asked follow-up questions to collect as much information as it could to try to help me. It provided me some useful information in an attempt to resolve my problem., but it wasn’t enough in my case. When it realized that it could not fix my issue, it transferred me to a customer representative and sent along all of the information that it had collected. It was nice to not have to provide all of my information over again.
If you’re going to be involved in designing an AI-assisted customer service centre, I suggest that you check out the Amazon approach.
If you’d like to comment on this article or explore these ideas further, contact me at roban.
This article was published in the
November 2022
edition of The TMC Advisor
- ISSN 2369-663X Volume:9 Issue:6
©2022 TMC Consulting