Social eXplainable Artificial Intelligence



The research community (and society in general) has already realised that the current centralised approach to AI is not an acceptable and sustainable model in the long run. The vision of SAI is towards a decentralised “collective” of local machine-learning-based AI components, interpreting data and interacting according to human-centric design principles, where explainability is guaranteed both at the local and collective level.

Read More
The SAI approach

Our Pillars

A network of PAIVs

In SAI each individual is associated with their own “Personal AI Valet” (PAIV), which acts as the individual’s proxy in a complex ecosystem of interacting PAIVs. PAIVs process individuals’ data via explainable AI models tailored to the specific characteristics of their human twins.

Explainable collective AI

PAIVs interact with each other, to build global decentralised AI models and come up with collective explainable decisions starting from the local (i.e., individual) models. These interactions and AI algorithms are human-centric, i.e., driven by quantifiable models of the individual and social behaviour of their human users.

Data-driven evaluation

The SAI design principles are validated on three concrete use cases: private traffic management, opinion diffusion/fake news detection in social media, and pandemic tracking and control. Real-life and synthetic datasets will be exploited.

Discover our publications



Consiglio Nazionale delle Ricerche


University of Pisa

Department of Computer Science

Central European University

Department of Network and Data Science

University of Sheffield

Department of Computer Science

Polish Academy of Sciences

Institute of Computer Science

University of Tartu

Institute of Computer Sciencei

Sofia University

GATE Institute

The next wave of ML-driven AI will be human-centric, explainable, and decentralised. SAI will help turn this vision into reality.

Andrea Passarella

Project PI

What we're up to

News from us

  • February 12, 2021

SAI kick-off meeting

The SAI project has been officially launched!

Read more

SAI is funded by the following bodies: