Welcome to our series on how to facilitate trust between chatbots and their users! The aim of this series is to give you a brief introduction to what trust means for human-agent interactions and why it is important. We will introduce you to four different aspects of trust relationships, but first: What even is trust?
There are many different definitions of trust, but for the purposes of this series, we will follow the Lee and See definition which define human trust in automation as the attitude that: “an agent will help achieve an individual’s goal in a situation characterised by uncertainty and vulnerability” [1].
When looking at trust, we can categorise the relationship into four different stages. First, we have the build phase. Here, the goal is to build a good initial trust relationship between the chatbots and its users. Once this has been established, we move on to maintaining said trust. Here, the goal is to keep the trust at an ideal level to avoid misuse and disuse.
Once the users interact with the chatbot, some form of failure is bound to happen. The break stage occurs when the chatbot makes a perceived failure and could impact the user's trust to different degrees. Once the trust has been broken, we enter the next stage, the goal here is similar to the build phase. Here we want to repair the breaks of trust through different repair strategies dependent on the failure and the situation. Once the trust has been restored, the cycle continues, always aimed at maintaining a good trust relationship between the user and the Chatbot.
The figure shows how all four stages might take place. With the goal being to always find the way back to a stage where we only need to maintain the users’ trust.
We will, in this series, introduce each of these stages in their own posts, and highlight some key aspects of how Kindly’s values align with this research.
This was part 1 of our series on building trust between chatbots and users. In the next post, we’ll take a closer look at how to build trust in the initial phase of the user journey. Read part 2 here.
[1] Lee JD, See KA. Trust in automation: designing for appropriate reliance. Hum Factors. 2004 Spring;46(1):50-80. doi: 10.1518/hfes.46.1.50_30392. PMID: 15151155.
