top of page

Luigi Ruggerone

Senior Director, ISPIC (Intesa Sanpaolo Innovation Center)

After completing his postgraduate studies in Economics in the UK, Luigi Ruggerone joined Banca Commerciale Italiana in 1996, working in Milan for the Research Department and then the Risk Management Department.

In 2013, with his wife Alessandra and his kids Luca and Sara, Luigi moved to Washington DC to join the International Monetary Fund, where he contributed to writing several issues of the Global Financial Stability Report until 2015.

In October 2015 he joined Intesa Sanpaolo Group to open and manage its new Representative Office in Washington DC.

On October 1st 2018 he was appointed Head of Innovation Business Development with Intesa Sanpaolo Innovation Center, a new company of Intesa Sanpaolo Group based in Turin and entirely dedicated to scout, analyze accelerate and support innovative technologies.

In  July 1st 2019 Luigi was appointed Director of Applied Research with Intesa Sanpaolo Innovation Center.

As of February 2023, Luigi is Responsible for the Business and Innovation Research Area of Intesa Sanpaolo Innovation Center, reporting to the Board of Directors.



Preserving confidentiality and privacy in data-driven applications


Increasingly, in data-driven applications one finds oneself dealing with data from different sources and entities that may be subject to privacy or confidentiality. 

Data encryption and pseudonymization, although they are different tools, serve the same purpose, which is to obfuscate the data to make it unintelligible to those who are not authorized to see it and thus fulfill the guidelines of the General Data Protection Regulation (GDPR).


Cryptography has always been concerned with protecting the confidentiality of private information, often encrypted in the form of numeric data, during its communication through insecure channels or during its storage. In this scenario, when a piece of data is encrypted it is unusable to anyone who does not possess the "decryption key," while it becomes visible "in the clear" if that key is known.

However, these techniques require decryption that let the user use the data in the clear.

To avoid this constraint, there are several techniques that can be used and are becoming increasingly established. In particular, we believe that Multiparty Computation (MPC) and Federated Learning are particularly worth investigating.


MPC is an area of cryptography that allows multiple parties (counterparties, legal entities, etc.) to jointly compute a function on private inputs (data made available in encrypted mode by the same parties). 

The MPC thus allows different parties to process the same data without having to expose it in the clear, and it is already used in some contexts (e.g. tax fraud detection in Estonia; collision avoidance between satellites belonging to different nations without revealing trajectories; E-election) and is currently more suitable for real-world/commercial applications.

Since 2018 the Artificial Intelligence Lab of Intesa Sanpaolo Innovation Center, has been working on an applied research challenge to create a MPC system (protocol).

Throughout this research project, a prototype capable of handling different computational functions (linear, polynomial and random forest) in a computationally efficient manner has been developed. The innovative elements of what was developed enabled the Intesa Sanpaolo Group to obtain an Italian patent (ID 102019000021576).

Federated Learning is a collaborative machine learning technique that allows different entities that own data to create a federation in which the information exchanged consists solely of the parameters/weights learned from the models trained on the data owned by the individual entities, thus ensuring the data remains within the organizations that own it. Within the Anti Financial Crime Digital Hub, a consortium created with the aim of combating financial crime through the use of new technologies and artificial intelligence, an applied research project was activated aimed at defining a framework to leverage information distributed across different counterparts to improve the learning of models based on AI techniques to fight financial crime.

The use cases examined were the first examples to explore some of Privacy-enhancing technologies (technologies that embody fundamental data protection principles by minimizing personal data use and maximizing data security) but further insights could be of interest for the banking sector in general. 


bottom of page