Dimitra Theofanou-Fülbier’s research while affiliated with Mercedes-Benz (Germany) and other places

In a fully autonomous driving situation, passengers hand over the steering control to a highly automated system. Autonomous driving behaviour may lead to confusion and negative user experience. When establishing such new technology, the user’s acceptance and understanding are crucial factors regarding success and failure. Using a driving simulator and a mobile application, we evaluated if system transparency during and after the interaction can increase the user experience and subjective feeling of safety and control. We contribute an initial guideline for autonomous driving experience design, bringing together the areas of user experience, explainable artificial intelligence and autonomous driving. The AVAM questionnaire, UEQ-S and interviews show that explanations during or after the ride help turn a negative user experience into a neutral one, which might be due to the increased feeling of control. However, we did not detect an effect for combining explanations during and after the ride.

Nowadays, intelligent assistants, such as Amazon's Alexa, are widely available. Unsurprisingly, intelligent assistants find their way into cars, in some cases as a major way to interact with the car. We conducted a user enactment exploring the impact of transparency on a possible future user experience with a digital AI assistant in the car. The focus is on whether tasks should be performed in an opaque way, only involving the user when it is necessary, or in a transparent way, always offering the user insights into what is being done and how. We present initial findings indicating a slight preference towards more transparency.

It is typically not transparent to end-users, how AI systems derive information or make decisions. This becomes crucial, the more pervasive AI systems enter human daily lives, the more they influence automated decision-making, and the more people rely on them. We present work in progress on explainability to support transparency in human AI interaction. In this paper, we discuss methods and research findings on categorizations of user types, system scope and limits, situational context, and changes over time. Based on these different dimensions and their range and combinations, we aim at individual facets of transparency that address a specific situation best. The approach is human-centered to provide adequate explanations with regard to their depth of detail and level of information, and we outline the different dimensions of this complex task.