arXiv subjects

George Loukas

Publications and source records attributed to George Loukas.

7 recordsLinked to original sources

arXiv2026

LAAF: A Layered Accountability Architecture Framework for LLM Applications

Large Language Models (LLMs) operate in hospitals, courtrooms, banks, and public service desks, where fluent, confident outputs are treated as authoritative even when ungrounded or incorrect. When such an output contributes to harm, who is answerable, and through what mechanisms can responsibility be traced, explained, and acted upon? Following PRISMA guidance, five databases were searched from January 2022 to March 2026 against four review questions; of 4,512 records identified, 122 primary studies were included, together with 12 regulatory and standards documents analysed as primary sources. The review consolidates a sociotechnical account of accountability as an actor-forum relation resolved into five dimensions, and synthesises mechanisms across four families: technical controls, human oversight, organisational governance, and documentation and traceability, each with a maturity assessment. The corpus is read through a four-layer classification device spanning provenance, application logic, human oversight, and governance and redress, cross-cut by traceability, role clarity, and continuous monitoring. Both are mapped onto the EU AI Act, whose high-risk obligations have applied since 2 August 2026, the NIST AI RMF with its Generative AI Profile, ISO/IEC 42001, and sectoral guidance in healthcare, consumer finance, education, and the public sector. Four persistent gaps emerge: under-specification of human oversight, absence of shared accountability metrics, disciplinary disconnection, and limited empirical evaluation, alongside five structural tensions that no surveyed instrument resolves. The review closes by consolidating the classification device into an integrated accountability architecture, LAAF, with cybersecurity aligned to the OWASP LLM Top 10 (2025); it is a synthesis of the surveyed evidence rather than a validated artefact.

cs.AI↗

arXiv2026

Dorsal Hand Images for Immersive (XR) and Privacy-preserving Age Assurance and Child Safety

Ensuring that Extended Reality (XR) environments are age-appropriate is an important regulatory and safety challenge. However, current age assurance operates only at registration and cannot verify the age of the active user during a session. Face-based approaches, the dominant solution in social media and adult platforms, are impractical in XR, because they require removing the headset and taking a self-captured image, often on a mobile app. This both breaks immersion and introduces the privacy risk of sharing face pictures with third parties, which leaves XR platforms without a viable path to continuous, in-session and privacy-preserving age assurance. We propose the dorsal part of the hand as an alternative to the face, by exploiting the egocentric cameras that XR headsets inherently and naturally use to capture gesture interactions. To evaluate this, we collect an age- and sex-stratified, ethnodiverse dataset of 436 participants spanning the minor--adult boundary, captured under unconstrained lighting and orientation conditions. To characterise what is achievable with off-the-shelf methods at the minor--adult boundary, we evaluate standard neural network architectures for age assurance at the legally critical 18-year threshold. Analysis confirms performance is robust to skin-tone variation. On this dataset, the challenge-31 operating point achieves zero minor admission, making the system a viable first-stage filter for age assurance. These findings position dorsal hand morphometrics as an effective and more privacy-preserving biometric modality for in-session age assurance in XR.

cs.CV↗

arXiv2024

A qualitative field study on explainable AI for lay users subjected to AI cyberattacks

In this paper we present results from a qualitative field study on explainable AI (XAI) for lay users (n = 18) who were subjected to AI cyberattacks. The study was based on a custom-built smart heating application called Squid and was conducted over seven weeks in early 2023. Squid combined a smart radiator valve installed in participant homes with a web application that implemented an AI feature known as setpoint learning, which is commonly available in consumer smart thermostats. Development of Squid followed the XAI principle of interpretability-by-design where the AI feature was implemented using a simple glass-box machine learning model with the model subsequently exposed to users via the web interface (e.g. as interactive visualisations). AI attacks on users were simulated by injecting malicious training data and by manipulating data used for model predictions. Research data consisted of semi-structured interviews, researcher field notes, participant diaries, and application logs. In our analysis we reflect on the impact of XAI on user satisfaction and user comprehension as well as its use as a tool for diagnosing AI attacks. Our results show only limited engagement with XAI features and suggest that, for Squid users, common assumptions found in the XAI literature were not aligned to reality. On the positive side, users appear to have developed better mental models of the AI feature compared to previous work, and there is evidence that users did make some use of XAI as a diagnostic tool.

cs.HC↗

arXiv2023

SoK: The MITRE ATT&CK Framework in Research and Practice

The MITRE ATT&CK framework, a comprehensive knowledge base of adversary tactics and techniques, has been widely adopted by the cybersecurity industry as well as by academic researchers. Its broad range of industry applications include threat intelligence, threat detection, and incident response, some of which go beyond what it was originally designed for. Despite its popularity, there is a lack of a systematic review of the applications and the research on ATT&CK. This systematization of work aims to fill this gap. To this end, it introduces the first taxonomic systematization of the research literature on ATT&CK, studies its degree of usefulness in different applications, and identifies important gaps and discrepancies in the literature to identify key directions for future work. The results of this work provide valuable insights for academics and practitioners alike, highlighting the need for more research on the practical implementation and evaluation of ATT&CK.

cs.CR↗

arXiv2021

HoneyCar: A Framework to Configure Honeypot Vulnerabilities on the Internet of Vehicles

The Internet of Vehicles (IoV), whereby interconnected vehicles communicate with each other and with road infrastructure on a common network, has promising socio-economic benefits but also poses new cyber-physical threats. Data on vehicular attackers can be realistically gathered through cyber threat intelligence using systems like honeypots. Admittedly, configuring honeypots introduces a trade-off between the level of honeypot-attacker interactions and any incurred overheads and costs for implementing and monitoring these honeypots. We argue that effective deception can be achieved through strategically configuring the honeypots to represent components of the IoV and engage attackers to collect cyber threat intelligence. In this paper, we present HoneyCar, a novel decision support framework for honeypot deception in IoV. HoneyCar builds upon a repository of known vulnerabilities of the autonomous and connected vehicles found in the Common Vulnerabilities and Exposure (CVE) data within the National Vulnerability Database (NVD) to compute optimal honeypot configuration strategies. By taking a game-theoretic approach, we model the adversarial interaction as a repeated imperfect-information zero-sum game in which the IoV network administrator chooses a set of vulnerabilities to offer in a honeypot and a strategic attacker chooses a vulnerability of the IoV to exploit under uncertainty. Our investigation is substantiated by examining two different versions of the game, with and without the re-configuration cost to empower the network administrator to determine optimal honeypot configurations. We evaluate HoneyCar in a realistic use case to support decision makers with determining optimal honeypot configuration strategies for strategic deployment in IoV.

cs.CR↗

arXiv2020

Optimizing Investments in Cyber Hygiene for Protecting Healthcare Users

Cyber hygiene measures are often recommended for strengthening an organization's security posture, especially for protecting against social engineering attacks that target the human element. However, the related recommendations are typically the same for all organizations and their employees, regardless of the nature and the level of risk for different groups of users. Building upon an existing cybersecurity investment model, this paper presents a tool for optimal selection of cyber hygiene safeguards, which we refer as the Optimal Safeguards Tool. The model combines game theory and combinatorial optimization taking into account the probability of each user group to being attacked, the value of assets accessible by each group, and the efficacy of each control for a particular group. The model considers indirect cost as the time employees could require for learning and training against an implemented control. Utilizing a game-theoretic framework to support the Knapsack optimization problem permits us to optimally select safeguards' application levels minimizing the aggregated expected damage within a security investment budget. We evaluate OST in a healthcare domain use case. The Critical Internet Security Control group 17 for implementing security awareness and training programs for employees belonging to the ICT, clinical and administration personnel of a hospital. We compare the strategies implemented by OST against alternative common-sense defending approaches for three different types of attackers: Nash, Weighted and Opportunistic. Nash defending strategies are consistently better than the competing strategies for all attacker types with a minor exception where the Nash defending strategy performs at least as good as other common-sense approaches.

cs.CR↗

arXiv2019

The spreading of computer viruses on time-varying networks

Social networks are the prime channel for the spreading of computer viruses. Yet the study of their propagation neglects the temporal nature of social interactions and the heterogeneity of users' susceptibility. Here, we introduce a theoretical framework that captures both properties. We study two realistic types of viruses propagating on temporal networks featuring Q categories of susceptibility and derive analytically the invasion threshold. We found that the temporal coupling of categories might increase the fragility of the system to cyber threats. Our results show that networks' dynamics and their interplay with users features are crucial for the spreading of computer viruses.

physics.soc-ph↗