He argues that models should not be seen primarily as predictive tools, but as instruments for exploring how assumptions and uncertainties shape conclusions.

Saltelli’s starting point is a familiar tension: models are often used where knowledge is incomplete, yet their results can appear highly authoritative. GSA addresses this by systematically shaking the model, exploring how uncertainty in inputs propagates through the analysis and testing whether policy-relevant conclusions remain robust under variation. Crucially, this shifts the focus from model outputs to the decisions models are intended to support.

The seminar also raises broader questions about the role of model-based assessment in policymaking. From one perspective, such practices risk becoming ritualised forms of validation. Alternatively, the dependence of models on prior assumptions and infrastructures can be treated not as a flaw, but as a condition that should be made explicit, helping to map conditional inference spaces and clarify what depends on what. In this view, GSA becomes a key tool for reconnecting technical modelling with both its epistemic limits and its political implications.

At the EUI and the Robert Schuman Centre, we are dedicated to removing barriers and providing equal opportunities for everyone. Please indicate your accessibility needs, if any, on the registration form. Alternatively, you can contact the event logistics organiser.