Great Unclear Expectations: A Study on the Defectiveness of Autonomous Systems

Abstract

Once autonomous systems become more widespread, the product liability is expected to be called for help to compensate damages caused by autonomous systems. The defectiveness of autonomous systems as a condition of the product liability is difficult to assess. This difficulty becomes more evident in the case of design defects because the consumer expectations test adopted by the EU Product Liability Directive is quite vague. This paper aims to bring together the cases where there are clear safety expectations regarding autonomous systems. In cases where safety expectations are unclear, this paper suggests that the defectiveness should be assessed based on the objective duty of care. The distinguishing feature of autonomous systems compared to conventional products is autonomy. Hence the producers' objective duty of care is to consider the main implications of autonomy that are human-machine interaction and testing the dependability and to try to eliminate the harmful consequences hereof.

Keywords: autonomous systems, product liability, defect, human-machine interaction, testing

JEL Classification: K10, K20