Algorithms in threat management

In recent years, various Swiss cantons have set up a threat management system in order to be able to react preventively to potential perpetrators of violence. Prof. Dr. Nora Markwalder and Prof. Dr. Monika Simmler, together with their team from the Competence Centre for Criminal Law and Criminology (SK-HSG) at the University of St.Gallen, have for the first time investigated which groups of people the authorities have registered as so-called Gefährder:innen:

• in 93.8% of the cases men
• 48% between the ages of 26 and 45
• 54% Swiss and 46% foreign nationals
• one third each of unemployed persons (38%) and persons with mental illness (33%).

46.5% of the cases investigated concern domestic or sexual violence and incidents in intimate relationships, 25.5% violent activities threatening public officials. Extremism and radicalisation were associated with 5.2% of the cases.

The study also investigated the role of algorithms in assessing the risks of these individuals and the measures taken by the relevant agencies to mitigate these risks. The research also highlights that many of the cases are in a legal grey area, which is why the researchers advocate for the creation of a clear legal framework for threat management. 

Download full study