No abstract

Autonomous vehicles have been in development for nearly thirty years and recently have begun to operate in real-world, uncontrolled settings. With such advances, more widespread research and evaluation of human interaction with autonomous vehicles (AV) is necessary. Here, we present an interview study of 32 pedestrians who have interacted with Uber AVs. Our findings are focused on understanding and trust of AVs, perceptions of AVs and artificial intelligence, and how the perception of a brand affects these constructs. We found an inherent relationship between favorable perceptions of technology and feelings of trust toward AVs. Trust in AVs was also influenced by a favorable interpretation of the company's brand and facilitated by knowledge about what AV technology is and how it might fit into everyday life. To our knowledge, this paper is the first to surface AV-related interview data from pedestrians in a natural, real-world setting.

No abstract

No abstract

We recommend using a model-centric, Boolean Satisfiability (SAT) formalism to obtain useful explanations of trained model behavior, different and complementary to what can be gleaned from LIME and SHAP, popular data-centric explanation tools in Artificial Intelligence (AI).We compare and contrast these methods, and show that data-centric methods may yield brittle explanations of limited practical utility.The model-centric framework, however, can offer actionable insights into risks of using AI models in practice. For critical applications of AI, split-second decision making is best informed by robust explanations that are invariant to properties of data, the capability offered by model-centric frameworks.