Nobody Likes A Complainer

Whining about mistreatment by a corporation is annoying news, even when it is the honest truth. If you want to get your message out, it needs to be interesting, even shocking, and if possible, amusing!

I am sure that many of you are familiar with the tale of the emperor’s new clothes which was written in 1837. Some con men sold the emperor on the idea of a beautiful new suit and the emperor and his court went along with the fiction because no one wanted to be the person who would admit that there was no suit. Even the emperor went along with the fiction, perhaps because he was a trusting sort of person.
No one else wanted to admit they could not see the emperor’s new suit because no one wanted to be thought of as a simpleton or unfit for their job.

In today’s world, artificial intelligence (AI) is the beautiful new suit for corporations seeking to save money by replacing human employees with robots. AI is powering phone systems, chatbots, assembly lines, translation systems, driving automobiles, and performing customer service duties. AI does a remarkable job on routine tasks and saves thousands of hours of valuable human resources when it is applied to those tasks.

When AI replaces human beings in driving vehicles, there are many roads that this system cannot handle and a human driver has to take over.

When AI is used for customer service duties, it can be a catastrophe waiting to occur. Why? Because AI does not handle random inputs well. Customer data entry will always contain some errors whether voice-based or keyboard entries. If the AI-driven system does not provide the user access to a human supervisor, as is the case at Square, the system can make judgments that are costly to customers and ruinous to small businesses. If you are using Square to process funds, this could be you.

Man vs artificial intelligence. Pop art retro vector illustration drawing

If you search the internet for “Problems with AI customer support”, you will find many examples of scenarios where AI is ineffective. In financial transactions, when there is a problem customers want interaction with a real human being. A system that denies you interaction with human beings, as is the case at Square, generates more ill will than you can imagine. I am sure that this is not because the people at Square are evil, it is more a case that they have all bought the story that AI will allow them to replace human customer support with robots.

A robot cannot care for a customer and if it is set to detect errors in data input and freeze funds for 90 days, that is what it will do without fail. At Square, they have attempted to handle data entry errors by using another AI system to interview the business owner, instead of using a customer service person who could clear the matter up in one call. The robot interview contains some serious programming errors and as a result, it attempts to solve them by repeating the interview until you figure out the answer the system is looking for. The thousands of online complaints about this robot interview show that no human being is looking at the results. As a result, Square is building a reputation for freezing customer funds for no good reason and this will eventually result in many people realizing that the emperor has no clothes when it comes to customer support.

If your company is considering using AI for customer support, look at the suggestions given by companies who are successfully using AI in the customer interface. Use this search to help you with your system design: Pros and cons with AI customer support

Example of a Square customer support questionnaire

This is what you get when you make a data entry error or the AI system thinks you are a high risk customer because you have a transaction larger than what the system considers normal. They put you through a robot interrogation to determine if you are a high risk customer.

A high risk transfer according to Square, is an international purchase or a card not present transaction. If you are an online business with overseas customers, you should avoid Square. If you have a transaction larger than usual, you should use another service to handle it as Square will freeze your money based on their models. The risk evaluation is based on Square’s machine learning models which seem to be flawed if the customer complaints are to be believed.

Square customer support questionnaire:


This entry was posted in artificial intelligence. Bookmark the permalink.

Leave a Reply

Your email address will not be published.

ninety three − eighty six =

This site uses Akismet to reduce spam. Learn how your comment data is processed.