A common assertion when discussing our future alongside AI is that we will need to learn new skills and adapt our practices in order to fully utilize the new technology. Research from the University of Surrey confirms this narrative and argues that our ability to interact with AI-based bots will play a key role in our experience with a range of commercial and public services.

The researchers conducted a study on how people handle issues during interactions with a chatbot, using simulated medical appointment bookings with a text-based chatbot created by Spryt, an AI scheduling assistant.

“While chatbots can be incredibly useful, one sticking point is that they sometimes need help understanding what you are saying,” they explain. “People often get very confused and frustrated with bots as a result of this.”

Working with bots

The study discovered that when people rely less on speaking to a bot as if it were human, they are more likely to resolve misunderstandings. For instance, rephrasing a question after a misunderstanding only resulted in a successful outcome in about half of all cases.

On the other hand, when people utilized strategies that are less common in human-to-human communication, such as restating the original purpose of the interaction or using the bot’s own wording, their success rate increased to 100% or over 60%.

“Users of chatbots and other forms of conversational AI can’t necessarily rely on the strategies they use from face-to-face interaction,” the researchers conclude. “This means that they have to carefully monitor whether the strategies they are using when interacting with bots are effective and get used to new ways of interacting.”