Founded in 2018 in collaboration with prestigious scientific partners, Intesa Sanpaolo Innovation Center's AI Lab aims to promote an open innovation model in applied research. The main goal of the lab is to develop and industrialize new technological assets, including advanced algorithms, mathematical models, and innovative humanoid robotics and service solutions, to actively support business. These efforts not only strengthen the center's intellectual property but also contribute to the production of numerous scientific publications that attest to the level of novelty and impact of the work being done.
The AI Lab is committed to knowledge sharing, disseminating the results of its research not only among colleagues but also to a wider audience through popular seminars, thus emphasizing its commitment to scientific dissemination and interdisciplinary collaboration.
Main Areas
Finance, Insurance & Wealth Management
We develop solutions to support in risk management and optimize financial strategies, minimizing the impact of market variations and simplifying complex processes
Anti Financial Crime & Fraud Detection
We support the fight against financial crime, facilitating the establishment of the AntiFinancialCrime Digital Hub, a consortium company that uses AI to counter AML, terrorism, financial sanctions and corruption
Climate & Biodiversity
We assess indirect climate risks in areas prone to extreme events, estimating the impact on property value also to support risk management related to retail mortgage portfolios
Health & Safety and Wellbeing
Since March 2020, we have been supporting Sacco Hospital in genomic sequencing and analysis of SARS-CoV-2, studying origin and spread in Italy with informatics, statistics and data science expertise
Activities of the Artificial Intelligence Lab
Artificial intelligence (AI) is the technological core of the AI Lab at Intesa Sanpaolo Innovation Center. It is used to develop systems capable of simulating typically human tasks, such as visual recognition, natural language understanding, complex problem solving, decision making and autonomous learning. These capabilities enable AI to act as a powerful support tool in the areas of productivity, creativity and decision-making capabilities, transforming the way business activities can be conducted and optimized. The AI Lab leverages these technologies to innovate and improve processes, leveraging artificial intelligence to address and solve operational and strategic challenges in ways that were previously unimaginable.
Collaborative model and PNRR
The model adopted by Intesa Sanpaolo Innovation Center's AI Lab is based on close collaboration among various actors, an approach also strongly emphasized in the National Plan for Recovery and Resilience (PNRR). This strategy promotes innovation as a result of a constant exchange of ideas and resources between the public and private sectors. Thanks to this collaborative model, significant projects such as the Anti Financial Crime Digital Hub have been launched, with the involvement of institutional and academic partners such as Intesa Sanpaolo, Intesa Sanpaolo Innovation Center, Politecnico di Torino (PoliTO), University of Turin (UniTO) and CENTAI. This synergy between different entities facilitates the creation of innovative solutions, which are essential for tackling complex challenges.
Claudia Berloco, Gianmarco De Francisci Morales, Daniele Frassineti, Greta Greco, Hashani Kumarasinghe, Marco Lamieri, Emanuele Massaro, Arianna Miola, Shuyi Yang
Michele Starnini, Charalampos E. Tsourakakis, Maryam Zamanipour, André Panisson, Walter Allasia, Marco Fornasiero, Laura Li Puma, Valeria Ricci, Silvia Ronchiadin, Angela Ugrinoska, Marco Varetto, Dario Moncalvo
ECML PKDD - European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (2021)
Carlo Abrate, Alessio Angius, Gianmarco De Francisci Morales, Stefano Cozzini, Francesca Iadanza, Laura Li Puma, Simone Pavanelli, Alan Perotti, Stefano Pignataro, Silvia Ronchiadin
ECML PKDD - European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (2021)
Predicting corporate credit risk: Network contagion via trade credit
Claudia Berloco, Gianmarco De Francisci Morales, Daniele Frassineti, Greta Greco, Hashani Kumarasinghe, Marco Lamieri, Emanuele Massaro, Arianna Miola, Shuyi Yang
Smurf-based Anti-Money Laundering in Time-Evolving Transaction Networks
Michele Starnini, Charalampos E. Tsourakakis, Maryam Zamanipour, André Panisson, Walter Allasia, Marco Fornasiero, Laura Li Puma, Valeria Ricci, Silvia Ronchiadin, Angela Ugrinoska, Marco Varetto, Dario Moncalvo
ECML PKDD - European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (2021)
Continuous-Action Reinforcement Learning for Portfolio Allocation of a Life Insurance Company
Carlo Abrate, Alessio Angius, Gianmarco De Francisci Morales, Stefano Cozzini, Francesca Iadanza, Laura Li Puma, Simone Pavanelli, Alan Perotti, Stefano Pignataro, Silvia Ronchiadin
ECML PKDD - European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (2021)
We apply GenAI to assist territorial legal consulting in analyzing the guarantee texts presented by clients, ensuring they comply with both internal and external regulations, and identifying deficiencies or significant misalignments. We are also exploring the possibility of drafting opinions based on similar cases. The "Human in the Loop" approach will ease the workload of colleagues, enabling them to focus more on more complex tasks, thus improving the overall efficiency of the process.
Risk Sensitivity
We support the Chief Risk Officer Area in defining analytical tools based on guidelines for measuring and monitoring exposure to respective risk factors (interest rates, equity, commodities) of financial instrument portfolios (debt securities, stocks, indices, derivatives) of the bank. These tools are designed to minimize the economic impacts of adverse market fluctuations through the identification of a specific algorithm. The algorithm should allow the grouping of individual risk factors (e.g., specific points on interest rate curves for different maturities) into “homogeneous” classes and the corresponding exposures to these factors through AI techniques, particularly using unsupervised learning methods.
Risk Overlay
We have developed solutions to support the front office in defining hedging strategies for investment portfolios. The project has improved efficiency in terms of time and performance, optimizing the definition of risk drivers, the creation of hedging portfolios, and their dynamic management. This has increased the available options for hedging, allowing the exploration of scenarios that were previously unconsidered.
Our Artificial Intelligence seminars
Register to get in touch with our network and tell us
the solutions you are interested in
