Towards an Active Predictive Relation by Reconceptualizing a Vacuum Robot: Research on the Transparency and Acceptance of the Predictive Behaviors

Abstract

With the development of Artificial intelligence, the connected objects are extended with the predictive capabilities and the character of things can change to “things that predict”. If a connected device is able to embrace a predictive system that not only profiles for scripted behavior but could also use the knowledge co-created by all the other similar devices and their users that encounter similar situations, the predictions can be generated based on that. In this case, a new type of interplay between humans and things called “predictive relation” is created. However, before this future takes place, it is required to find out appropriate patterns to address challenges such as the transparency and users’ acceptance of predictive behaviors of connected products. The research in this article takes a vacuum robot as a reference product for the study. The research starts by collecting users’ daily practice with vacuum robots through 4-day diary booklets. And then the booklets serve as sensitizing tools to envision the possible predictive capabilities and lead the discussion on the acceptance and transparency of general predicting things. From the creative sessions we propose 1) design qualities for the acceptance of the predicting things, and 2) a model of generating predictive behavior that enhances the transparency. Eventually, we also propose the idea of “Designers as the facilitators of the human-robot collaboration”.