Foundations of AI Security and Privacy
Workshop Details
Practical Information
No specific equipment required.
Basic awareness about AI and machine learning.
About this Workshop
Artificial Intelligence (AI) systems are increasingly embedded in critical infrastructures, autonomous agents, and physical devices, creating unprecedented opportunities but also new risks to security, privacy, and societal trust. As AI becomes embodied and agentic—integrated into healthcare diagnostics, autonomous robots, wireless networks, and edge devices—vulnerabilities at both the algorithmic and hardware levels expose individuals, organisations, and societies to threats that transcend traditional cybersecurity boundaries.
At large, we aim to provide a constructive forum for promoting international research collaborations to advance the foundations, technologies, and governance mechanisms required to design, deploy, and regulate trustworthy AI systems.
In this first edition of this workshop at AI Days 2026, we will invite prestigious researchers to talk about their ongoing work on developing verifiable methods for protecting AI systems against emerging adversarial, data-poisoning, and privacy attacks.
Workshop Program
13:00–13:05
Welcome & Introductory Remarks
13:05–13:30
Sayan Biswas, EPFL
Decentralized ML: Trust No One, Train Together
13:30–13:55
Yann Chevaleyre, Sorbonne University
Adversarial Attacks: From Image Classification Neural Networks to LLMs
13:55–14:05
Break
14:05–14:30
Anastasiia Kucherenko, HES-SO Valais-Wallis
Understanding LLM Training Data: Search, Attribution, and Security Implications
14:30–14:55
Leonardo F. Toso, Columbia University
Learning What to Share: Feature Learning for Adversarially Robust Federated Systems
14:55–15:00
Closing Remarks
Speakers & Organizers
Dr. Rafael Pinot
Junior Professor, Department of Mathematics (Organizer)
Sorbonne University
Rafael is a junior professor in the department of mathematics at Sorbonne University. He holds a chair on the mathematical foundation of computer and data science within the LPSM research unit. He is also an active member of the Responsible AI initiative within the Sorbonne Center for Artificial Intelligence (SCAI). From 2021 to 2023, he was a postdoctoral researcher at École Polytechnique Fédérale de Lausanne, where he worked with Pr. Rachid Guerraoui and Pr. Anne-Marie Kermarrec within the Ecocloud Research Center. From 2017 to 2020 he completed his PhD in Computer Science at PSL University (Paris Dauphine) and Université Paris Saclay (CEA LIST) where he was advised by Pr. Jamal Atif, Dr. Florian Yger, and Dr. Cédric Gouy-Pailler.
Dr. Nirupam Gupta
Tenure-Track Assistant Professor of Computer Science (Organizer)
University of Copenhagen
Nirupam is a Tenure-Track Assistant Professor in the ML Section of the Department of Computer Science at University of Copenhagen (DIKU). Before joining DIKU, he was a Postdoctoral Researcher in the School of Computer Science at EPFL (Switzerland) and the Department of Computer Science at Georgetown University (USA). He obtained his PhD in 2019 from the University of Maryland College Park (USA) and his Bachelor's degree in 2013 from the Indian Institute of Technology Delhi (India).
Prof. Yann Chevaleyre
Full Professor, Department of Mathematics
Sorbonne University (LAMSADE)
Yann Chevaleyre is a full professor in the department of mathematics at Sorbonne University. He holds a chair on the mathematical foundation of computer and data science within the LPSM research unit. He is also an active member of the Responsible AI initiative within the Sorbonne Center for Artificial Intelligence (SCAI). Since 2024, he is responsible for the Data Science team at LAMSADE. His research focuses on machine learning, reinforcement learning, game theory, and adversarial robustness in AI systems.
Dr. Sayan Biswas
Postdoctoral Researcher
EPFL - Scalable Computing Systems (SaCS) Lab
Sayan Biswas is a postdoctoral researcher at EPFL's Scalable Computing Systems Lab, supervised by Prof. Anne-Marie Kermarrec. He completed his PhD in Computer Science at INRIA and École Polytechnique in 2023. His research focuses on designing secure and trustworthy distributed systems for decentralized learning and training ML models with emphasis on privacy-preserving approaches, differential privacy, federated learning, fairness, and personalization. He has been recognized with the Best Paper Award at CADE 2022 and has published extensively at premier venues including ICML, PoPETs, and other top-tier conferences.
Dr. Anastasiia Kucherenko
Postdoctoral Researcher
Institute of Entrepreneurship and Management, HES-SO Valais-Wallis
Anastasiia Kucherenko is a computer science researcher with a strong mathematical background, passionate about solving real-world problems. Currently focused on AI safety, she specializes in training data attribution—the first step toward explainable AI. Her research expertise includes gossip and epidemic protocols, differential privacy, distributed algorithms, probability theory, graph theory, network science, cryptography, and complexity theory.
Leonardo F. Toso
Ph.D. Candidate
Columbia University, Department of Electrical Engineering
Leonardo F. Toso is a fourth-year Ph.D. candidate in Electrical Engineering at Columbia University, advised by Prof. James Anderson. He is a Presidential and CAIRFI (Center for AI and Responsible Financial Innovation) Fellow. His research focuses on the intersection of control theory, machine learning, and optimization, with particular emphasis on meta-learning, federated learning, and adaptive control. His work integrates safety, robustness, and learning in complex distributed systems. He has received the Best Paper Award at L4DC 2024 and the Outstanding Paper Award at CDC 2025, and has published extensively at top venues including ICLR, AAAI, and other premier conferences.