Davy Preuveneers

The trend to support healthcare outside of the premises and control of health institutions has given rise to a diversity of devices (e.g. wearables), services, and players contributing to the ecosystem. In consequence, health-related and medical data are frequently dispersed across multiple institutions and devices, posing a significant challenge to comprehensive data analysis and exploration for AI-driven research on new medicines or treatments. Centralizing such sensitive data is often unachievable due to the rigorous security and privacy demands imposed by various stakeholders, including patients, general practitioners, and pharmaceutical researchers. The decentralized and dispersed nature of healthcare data, in combination with stringent regulatory constraints such as HIPAA and GDPR, necessitate innovative approaches to collaborative healthcare research. Federated learning is a promising class of solutions to harness the collective intelligence of diverse datasets by allowing collaborative model training across decentralized data sources. Despite the many privacy-enhancing techniques proposed to ensure the confidentiality of data and intellectual property, this chapter discusses and illustrates the challenges of leveraging shared representations and knowledge transfer between smart health applications while addressing the intricacies of data fragmentation, security, and privacy.

The success of machine learning has significantly impacted biometric authentication systems, such as facial recognition, but it has also exposed them to substantial risks through adversarial examples. This work investigates physically realizable adversarial examples against RGB-D facial recognition systems, which combine traditional RGB images with depth data. While showing increased performance in a benign setting, we demonstrate that RGB-D systems remain vulnerable to adversarial examples in white and black-box settings. Furthermore, we show that attacks targeting a single modality, such as only making perturbations on the RGB color space without changing the depth data, remain highly successful, even against RGB-D models. Our research compares the robustness of RGB and RGB-D systems to identify the benefits and limitations of incorporating depth data into face recognition systems. We quantify how different fusion architectures of RGB-D systems influence their resistance to attack and evaluate how the size and positioning of adversarial patches impact attack success rates. Our findings highlight that while multi-modal systems offer potential advantages, they still face challenges in mitigating adversarial vulnerabilities.

This paper presents a comprehensive analysis of web bot activity, exploring both offensive and defensive perspectives within the context of modern web infrastructure. As bots play a dual role—enabling malicious activities like credential stuffing and scraping while also facilitating benign automation—distinguishing between humans, good bots, and bad bots has become increasingly critical. We examine the technical challenges of detecting web bots amidst large volumes of benign traffic, highlighting the privacy risks involved in monitoring users at scale. Additionally, the study dives into the use of Privacy Enhancing Technologies (PETs) to strike a balance between bot detection and user privacy. These technologies provide innovative approaches to minimising data exposure while maintaining the effectiveness of bot-detection mechanisms. Furthermore, we explore the legal and ethical considerations associated with bot detection, mapping the technical solutions to the regulatory frameworks set forth by the EU General Data Protection Regulation (GDPR) and the Artificial Intelligence Act (AI Act). By analysing these regulatory constraints, we provide insights into how organisations can ensure compliance while maintaining robust bot defence strategies, fostering a responsible approach to cybersecurity in a privacy-conscious world.

This paper presents a comprehensive analysis of web bot activity, exploring both offensive and defensive perspectives within the context of modern web infrastructure. As bots play a dual role—enabling malicious activities like credential stuffing and scraping while also facilitating benign automation—distinguishing between humans, good bots, and bad bots has become increasingly critical. We examine the technical challenges of detecting web bots amidst large volumes of benign traffic, highlighting the privacy risks involved in monitoring users at scale. Additionally, the study dives into the use of Privacy Enhancing Technologies (PETs) to strike a balance between bot detection and user privacy. These technologies provide innovative approaches to minimising data exposure while maintaining the effectiveness of bot-detection mechanisms. Furthermore, we explore the legal and ethical considerations associated with bot detection, mapping the technical solutions to the regulatory frameworks set forth by the EU General Data Protection Regulation (GDPR) and the Artificial Intelligence Act (AI Act). By analysing these regulatory constraints, we provide insights into how organisations can ensure compliance while maintaining robust bot defence strategies, fostering a responsible approach to cybersecurity in a privacy-conscious world.

Adversarial training, a widely used technique for fortifying the robustness of machine learning models, has seen its effectiveness further bolstered by modifying loss functions or incorporating additional terms into the training objective. While these adaptations are validated through empirical studies, they lack a solid theoretical basis to explain the models’ secure and robust behavior. In this paper, we investigate the integration of adversarial triplets within the adversarial training framework, a method previously shown to enhance robustness. However, the reasons behind this increased robustness are poorly understood, and the impact of different adversarial triplet configurations remains unclear. To address this gap, we utilize the robust and non-robust features framework to analyze how various adversarial triplet compositions influence robustness, providing deeper insights into the robustness guarantees of this approach. Specifically, we introduce a novel framework that explains how different compositions of adversarial triplets lead to distinct training dynamics, thereby affecting the model’s adversarial robustness. We validate our theoretical findings through empirical analysis, demonstrating that our framework accurately characterizes the effects of adversarial triplets on the training process. Our results offer a comprehensive explanation of how adversarial triplets influence the security and robustness of models, providing a theoretical foundation for methods that employ adversarial triplets to improve robustness. This research not only enhances our theoretical understanding but also has practical implications for developing more robust machine learning models.

This study critically examines the trustworthiness of widely used AI applications, focusing on their integration into daily life, often without users fully understanding the risks or how these threats might affect them. As AI apps become more accessible, users tend to trust them due to their convenience and usability, frequently overlooking critical issues such as security, privacy, and ethics. To address this gap, we introduce a user-centric framework that enables individuals to assess the trustworthiness of AI applications based on their own experiences and perceptions. The framework evaluates several dimensions—transparency, security, privacy, ethics, and compliance—while also aiming to raise awareness and bring the topic of AI trustworthiness into public dialogue. By analyzing AI threats, real-world incidents, and strategies for mitigating the risks posed by AI apps, this study contributes to the ongoing discussions on AI safety and trust.