Reflections on the relationship between algorithmic transparency and the social acceptability of autonomous vehicles. - Naver Labs Europe
preloder

Abstract

This position paper discusses the relationship between the social acceptability of autonomous and semi-autonomous vehicles and the transparency of the underlying artificial intelligence algorithms that drive autopilot technology. I draw on existing, ethnographically informed approaches to the design of interface / interaction mechanisms and autopilot features that advocate for the need to make autonomous vehicles appear more competent with respect to the social and organizational properties of traffic, and argue that this approach may come at the cost of transparency, and the opportunity for users to usefully and successfully appropriate the technology.