By Peter J. Scott and Roman V. Yampolskiy

In this paper we examine historical failures of artificial intelligence (AI) and propose a classification scheme for categorising future failures. By doing so we hope that (a) the responses to future failures can be improved through applying a systematic classification that can be used to simplify the choice of response and (b) future failures can be reduced through augmenting development lifecycles with targeted risk assessments.

This article was published in Delphi Journal’s Special Issue on AI Governance. Please follow this link to read more.