Home                 Projects                 People               Resources                  Papers               Workshops                Contact

 

The Social Consequences of Algorithmic Forecast
in Insurance, Medicine and Policing

Elena Esposito, Bielefeld University / University of Bologna

 

Project Summary

The algorithmic turn of prediction, connected with Big Data and Machine Learning, presents an exciting and urgent challenge for the social sciences. Recent advances in digital forecasting claim to provide a predictive score for individual persons or singular events, thereby introducing a new way to manage the uncertainty of the future. But knowing the future in advance is not only advantageous. In fact, for our society, uncertainty about the future is also a resource.

ERC Logo

Since modernity, with the support of probability calculus various social institutions in different domains have developed means of coping with ignorance of the future by starting with the one thing that we all share - uncertainty. What happens to the stabilized forms of management of the future when their first resource - shared uncertainty - is missing?

This project includes a set of theory-driven empirical studies of the transition from probabilistic forms of uncertainty management to the new algorithmic forms of prediction. We investigate three important social areas highlighting three fundamental dimensions with which digital forecast must deal.

  • In personalized insurance our key dimension is individualization of prediction, where the challenge is that such prediction could undermine the mutualization principle organized around actuarial practices.

  • Our second research area is precision medicine, highlighting the dimension of generalization where the challenge is the combination of algorithmic procedures with established statistical methods.

  • In the third field, predictive policing underscores the problem of bias while it challenges the distinction between prevention and repression.

How is society coping with the innovative and disruptive consequences of algorithmic prediction?

Exploring these transformation and its consequences, the project aims at developing a comprehensive approach to study the social, technical and theoretical aspects of prediction in digital society.

 

Social fieldDimensionChallengeData and methods

Personalized Insurance

 

Individualization

 

  • personalization of risk calculation
  • inversion of information asymmetry
  • proactivity
  • archives of personalized car insurance policies
  • online offers of collaborative insurance platforms; interviews
  • Open-F@b Call4Ideas for preventive insurance models

Precision Medicine

Generalization

  • n=all vs. n-of-1
  • non-statistical notion of health
  • right not-to-know
  • algorithmic semantic analysis of biomedical journals archives
  • HREB requests for approval
  • court cases; debates in medical and bioethical literature

Predictive Policing

Bias

  • distinction prevention and repression
  • intransparency of predictive algorithms
  • ethnographic research in police departments
  • citizen's petitions and administrative answers

 

ERC Logo

This project has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (Grant agreement No. 833749).

 

 

 

 

 

 

News

Recent Publications:

HEIMSTÄDT, M., EGBERT, S., & ESPOSITO, E. (2021). A Pandemic of Prediction: On the Circulation of Contagion Models between Public Health and Public Safety. Sociologica, Vol 14 n. 3, p. 1-24. Open Access

Recent Videos:

Elena Esposito. Teoria delle previsioni on RAI 3, Italy