AI-Based Public Policy Making: A New holistic, Integrated and “AI by Design” Approach

Public policy development is becoming more and more challenging for public administrations and the recent events (first of all the Covid-19 Outbreak) demonstrated how an evidence-based approach is crucial for managing critical situations with short responses, fast adaptation, and citizens' support. The ultimate vision of data-driven policy making entails the use of Artificial Intelligence (AI) as a means of increasing the efficiency of the policy development and management process (i.e. going beyond development to adaptation and optimization) and boosting a more responsive, adaptive, intelligent, and citizen-centric governance. The cross-industry standard process for data mining, known as CRISP-DM, is the most widely-used open standard process model that describes common approaches used by AI experts. The paper presents how the H2020 project AI4PublicPolicy leverages this model to promote a new holistic, integrated, and “AI by design” approach for public policy making, providing the Virtualized Policy Management Environment (VPME) and the blueprints for organizational transformation. Moreover, the paper presents a practical use and validation of this approach in a real-life public policy making case.