MINNEAPOLIS - Artificial intelligence (AI)-powered chatbots frequently provide customer service, training, sales, marketing, and other business functions. They are a popular and attractive business tool as they help lower operating costs, boost efficiency, and enhance customer satisfaction and thus retention. Their irresponsible use, however, leads to serious problems - Air Canada’s recent bad experience in this regard offers a useful cautionary tale.

This saga began with Air Canada’s implementation of an AI chatbot as part of its customer service program. A customer asked the chatbot about the airline’s bereavement airfare policy - a bereavement flight in the United States and Canada is a flight purchased when a near relation has died or is dying. The chatbot advised him: “If you need to travel immediately or have already travelled and would like to submit your ticket for a reduced bereavement rate, kindly do so within 90 days of the date your ticket was issued by completing our Ticket Refund Application.” By all measures, this appeared to be a straightforward answer - the customer relied on it and bought his ticket. 

There was one major problem, however: The information the chatbot provided was incorrect. Therefore, when the customer submitted his request for reimbursement - exactly as instructed by its chatbot - Air Canada denied it. The airline’s human representative explained to the customer that this reimbursement was contrary to the airline’s policy. Understandably, the customer was upset and subsequently filed a complaint before the Civil Resolution Tribunal of British Columbia (CRTBC).