“Get nothing wrong”: Perspectives on the functions and fallibilities of professionals and algorithmic technologies in law and justice

peer reviewed ; 38th EGOS Colloquium 2022 Sub-theme 31: Imperfect Knowledge: Re-examining the Role of Experts and Expertise 7-9th July 2022, Vienna, Austria “Get nothing wrong”: perspectives on the functions and fallibilities of professionals and algorithmic technologies in law and justice Christophe Dubois (University of Liège) c.dubois@uliege.be James Faulconbridge (Lancaster University) j.faulconbridge@lancaster.ac.uk Frida Pemer (Stockholm School of Economics) frida.pemer@hhs.se Caroline Ruiner (University of Hohenheim) caroline.ruiner@uni-hohenheim.de Aline P. Seepma (University of Groningen) a.p.seepma@rug.nl Tale Skjølsvik (Oslo Metropolitan University) talesk@oslomet.no Martin Spring (Lancaster University) m.spring@lancaster.ac.uk Introduction In this paper, we explore questions about the definition and constitution of expertise and experts as algorithmic technologies impact professional work. It is the aim of this paper to analyse the effects the implementation of Artificial Intelligence (AI) on professions, focusing on law and justice. We draw on Eyal’s (2019: 26) typology of different conceptions of expertise, and analyses that disaggregate expert work at the level of tasks (Sampson, 2020), to identify “what experts do”. This allows us to examine differing degrees and forms of expertise in different facets of expert work (Dreyfus and Dreyfus, 2005). Part of a professional logic is to “get nothing wrong”, yet the use of algorithmic technologies introduces new sources of imperfection, as well as revealing existing (human) ones. Based on the introduction of cases in law using algorithmic technologies, we propose a framework for understanding the different ways algorithmic technologies do and do not reconstitute the different roles and practices of professional experts.