Ethical Plateaus in Danish Child Protection Services

Abstract

This paper analyses how controversies shape an emerging field of AI in Danish child protection services. In a context of high controversiality, we examine how algorithmic systems evolve in conjunction with changing ethical stakes. Empirically, we report a study comprising all Danish attempts (n=4) to develop algorithmic models for child protection services. These attempts were never fully implemented and have been either cancelled, paused or changed significantly since their outset. Combining Fischer’s (2004) notion of ‘ethical plateaus’ with insights from valuation studies, we propose that public controversies shape how organisations enact their algorithms as ethically ‘good’. Our findings demonstrate how valuations of ethically contestable algorithms involve the very distribution of agency across humans and algorithms, i.e., how much power and agency should be delegated to algorithmic models. In the case of Danish child protection services, this moves towards reducing their agency.

Algorithmic constructions of risk: Anticipating uncertain futures in child protection services

“…Empirically, we compare two Danish experiments developing algorithmic models to calculate children's probability of future maltreatment. While predictive algorithms are internationally on the rise in child protection services, they are highly controversial in a Danish context and are yet to make it into child protection services (Kristensen, 2022; Leslie et al, 2020; Ratner and Schrøder, in press; Redden, 2020). One of the projects, initiated by a Danish municipality, has been cancelled whereas the other, a research project, has been changed to the extent that the algorithmic model will not be tested on real cases.…”