Deterrence or prevention? The challenges of algorithmic crimeprediction in urban spaces

Predictive policing emerged as a promise to prevent crime before it occurs.Despite the waning hype, predictive policing remains an active practice,especially in urban contexts. Drawing on ethnographic research ona predictive policing system of a German state police force, this paperexamines the territorial and organizational conditions shaping the imple-mentation and effectiveness of algorithmic crime forecasts. We developa comparative analysis of their application in both urban and rural contexts.Our findings show that predictive policing is technically and conceptuallytailored to urban areas, with limited potential for effective implementation inrural areas. Its practical implementation in urban areas, however, is oftenhindered by resource constraints, organizational hurdles, and the short-termlogic of deterrence. As a result, the effectiveness of predictions is oftenlimited even in the urban settings for which they were designed. We contrastthis deterrence-oriented model with emerging preventive approaches basedon micro-segmented spatial analysis, which seek to transform the conditionsenabling crime rather than merely forecasting its occurrence. By distinguish-ing between deterrence and prevention, we argue that sustainable algorith-mic crime prevention in urban spaces requires reconfiguring predictivetechnologies as tools for collaborative, long-term interventions rather thanshort-term operational targeting.