arXiv subjects

Ali Dehghantanha

Publications and source records attributed to Ali Dehghantanha.

At least 19 recordsLinked to original sources

arXiv2026

Beyond the Hype: Evaluating LLM Integration and Practical Limitations in Security Operation Centers

Large Language Models (LLMs) are increasingly being explored within Security Operation Centers (SOCs) to support text-heavy analytical work such as alert contextualization, incident summarization, and drafting investigative artifacts. Despite this interest, practitioners describe critical operational concerns, most notably hallucinations (plausible but incorrect outputs), opaque reasoning, and the verification effort required to safely use model-generated content in security workflows. In this paper, we present findings from semi-structured interviews with 20 SOC practitioners spanning frontline analysts, SOC managers, and tool developers. Participants report perceived time savings for low-stakes tasks that are quickly verifiable (e.g., summarizing logs or drafting initial investigative leads), but they consistently frame LLM outputs as preliminary drafts and suggestions rather than decision-grade conclusions. Participants also describe limited trust in LLMs for high-stakes security decisions due to unreliable outputs and unclear model reasoning, and they report relying primarily on ad-hoc verification norms and continuous human oversight rather than standardized mitigation procedures. Based on these interview-grounded accounts, we introduce a maturity rubric to characterize readiness for LLM integration and outline a research agenda emphasizing auditability and transparent explanation mechanisms to support safer adoption in SOC workflows.

cs.CR↗

arXiv2026

SoK: The Attack Surface of Agentic AI - Tools and Autonomy

Recent AI systems combine large language models with tools, external knowledge via retrieval-augmented generation (RAG), and even autonomous multi-agent decision loops. This agentic AI paradigm greatly expands capabilities - but also vastly enlarges the attack surface. In this systematization, we map out the trust boundaries and security risks of agentic LLM-based systems. We develop a comprehensive taxonomy of attacks spanning prompt-level injections, knowledge-base poisoning, tool/plug-in exploits, and multi-agent emergent threats. Through a detailed literature review, we synthesize evidence from 2023-2025, including more than 20 peer-reviewed and archival studies, industry reports, and standards. We find that agentic systems introduce new vectors for indirect prompt injection, code execution exploits, RAG index poisoning, and cross-agent manipulation that go beyond traditional AI threats. We define attacker models and threat scenarios, and propose metrics (e.g., Unsafe Action Rate, Privilege Escalation Distance) to evaluate security posture. Our survey examines defenses such as input sanitization, retrieval filters, sandboxes, access control, and "AI guardrails," assessing their effectiveness and pointing out the areas where protection is still lacking. To assist practitioners, we outline defensive controls and provide a phased security checklist for deploying agentic AI (covering design-time hardening, runtime monitoring, and incident response). Finally, we outline open research challenges in secure autonomous AI (robust tool APIs, verifiable agent behavior, supply-chain safeguards) and discuss ethical and responsible disclosure practices. We systematize recent findings to help researchers and engineers understand and mitigate security risks in agentic AI.

cs.CR↗

arXiv2026

AI Forensics Across White-, Grey-, and Black-Box Access: A Process Model and Research Agenda for Post-Incident Investigation of AI Systems

AI systems are increasingly involved in decisions and actions that may later require investigation. When an AI related incident occurs, investigators need to reconstruct what the system did, why it behaved that way, and which part of the system or supply chain contributed to the outcome. Existing work on AI forensics remains fragmented, often focusing on a specific system type, artifact, or analysis technique. This paper argues that investigator access is a useful starting point for organizing the field. We distinguish white box, grey box, and black box access and show how each access level changes what can be collected, preserved, analyzed, and reported. Based on this distinction, we propose a process model matrix for AI forensics across four phases: collection, preservation, analysis, and reporting. We also introduce an order of volatility for AI systems, covering runtime state, context windows, logs, retrieval stores, model artifacts, and training lineage. From this matrix, we derive an access conditioned examination framework and identify open research problems, including black box preservation, model version attestation, uncertainty quantification for surrogate based analysis, and chain of custody for mutable AI artifacts.

cs.CR↗

arXiv2026

Diagnosing and Repairing Factual Errors in RAG under Budget Constraints

Retrieval-Augmented Generation (RAG) improves the factuality of large language models by grounding responses in external evidence, yet real-world deployments remain fragile. Failures often stem from missing or weakly relevant evidence, as well as from generation that does not faithfully reflect the retrieved context. Many existing approaches rely on fine-tuning, privileged access to internal model signals, or resource-insensitive escalation strategies, which limits their practicality in black-box and budget-constrained settings. We propose D2R-RAG (Diagnose-to-Repair RAG), a model-agnostic and resource-aware framework that combines lightweight failure diagnosis with adaptive repair. D2R-RAG derives interpretable failure signatures from observable signals in the query, retrieved evidence, and generated response, and then selects from a small set of corrective actions under explicit latency and VRAM constraints. Experiments on FEVER and HotpotQA show that D2R-RAG improves reliability over recent baselines and achieves better accuracy--efficiency trade-offs across multiple compute budgets. The code is available at https://github.com/CyberScienceLab/D2R-RAG/.

cs.AI↗

arXiv2025

Uncovering the Persuasive Fingerprint of LLMs in Jailbreaking Attacks

Despite recent advances, Large Language Models remain vulnerable to jailbreak attacks that bypass alignment safeguards and elicit harmful outputs. While prior research has proposed various attack strategies differing in human readability and transferability, little attention has been paid to the linguistic and psychological mechanisms that may influence a model's susceptibility to such attacks. In this paper, we examine an interdisciplinary line of research that leverages foundational theories of persuasion from the social sciences to craft adversarial prompts capable of circumventing alignment constraints in LLMs. Drawing on well-established persuasive strategies, we hypothesize that LLMs, having been trained on large-scale human-generated text, may respond more compliantly to prompts with persuasive structures. Furthermore, we investigate whether LLMs themselves exhibit distinct persuasive fingerprints that emerge in their jailbreak responses. Empirical evaluations across multiple aligned LLMs reveal that persuasion-aware prompts significantly bypass safeguards, demonstrating their potential to induce jailbreak behaviors. This work underscores the importance of cross-disciplinary insight in addressing the evolving challenges of LLM safety. The code and data are available.

cs.CL↗

arXiv2025

Quantifying Security Vulnerabilities: A Metric-Driven Security Analysis of Gaps in Current AI Standards

As AI systems integrate into critical infrastructure, security gaps in AI compliance frameworks demand urgent attention. This paper audits and quantifies security risks in three major AI governance standards: NIST AI RMF 1.0, UK's AI and Data Protection Risk Toolkit, and the EU's ALTAI. Using a novel risk assessment methodology, we develop four key metrics: Risk Severity Index (RSI), Attack Potential Index (AVPI), Compliance-Security Gap Percentage (CSGP), and Root Cause Vulnerability Score (RCVS). Our analysis identifies 136 concerns across the frameworks, exposing significant gaps. NIST fails to address 69.23 percent of identified risks, ALTAI has the highest attack vector vulnerability (AVPI = 0.51) and the ICO Toolkit has the largest compliance-security gap, with 80.00 percent of high-risk concerns remaining unresolved. Root cause analysis highlights under-defined processes (ALTAI RCVS = 033) and weak implementation guidance (NIST and ICO RCVS = 0.25) as critical weaknesses. These findings emphasize the need for stronger, enforceable security controls in AI compliance. We offer targeted recommendations to enhance security posture and bridge the gap between compliance and real-world AI risks.

cs.CR↗

arXiv2024

P3GNN: A Privacy-Preserving Provenance Graph-Based Model for APT Detection in Software Defined Networking

Software Defined Networking (SDN) has brought significant advancements in network management and programmability. However, this evolution has also heightened vulnerability to Advanced Persistent Threats (APTs), sophisticated and stealthy cyberattacks that traditional detection methods often fail to counter, especially in the face of zero-day exploits. A prevalent issue is the inadequacy of existing strategies to detect novel threats while addressing data privacy concerns in collaborative learning scenarios. This paper presents P3GNN (privacy-preserving provenance graph-based graph neural network model), a novel model that synergizes Federated Learning (FL) with Graph Convolutional Networks (GCN) for effective APT detection in SDN environments. P3GNN utilizes unsupervised learning to analyze operational patterns within provenance graphs, identifying deviations indicative of security breaches. Its core feature is the integration of FL with homomorphic encryption, which fortifies data confidentiality and gradient integrity during collaborative learning. This approach addresses the critical challenge of data privacy in shared learning contexts. Key innovations of P3GNN include its ability to detect anomalies at the node level within provenance graphs, offering a detailed view of attack trajectories and enhancing security analysis. Furthermore, the models unsupervised learning capability enables it to identify zero-day attacks by learning standard operational patterns. Empirical evaluation using the DARPA TCE3 dataset demonstrates P3GNNs exceptional performance, achieving an accuracy of 0.93 and a low false positive rate of 0.06.

cs.CR↗

arXiv2024

A Federated Learning Approach for Multi-stage Threat Analysis in Advanced Persistent Threat Campaigns

Multi-stage threats like advanced persistent threats (APT) pose severe risks by stealing data and destroying infrastructure, with detection being challenging. APTs use novel attack vectors and evade signature-based detection by obfuscating their network presence, often going unnoticed due to their novelty. Although machine learning models offer high accuracy, they still struggle to identify true APT behavior, overwhelming analysts with excessive data. Effective detection requires training on multiple datasets from various clients, which introduces privacy issues under regulations like GDPR. To address these challenges, this paper proposes a novel 3-phase unsupervised federated learning (FL) framework to detect APTs. It identifies unique log event types, extracts suspicious patterns from related log events, and orders them by complexity and frequency. The framework ensures privacy through a federated approach and enhances security using Paillier's partial homomorphic encryption. Tested on the SoTM 34 dataset, our framework compares favorably against traditional methods, demonstrating efficient pattern extraction and analysis from log files, reducing analyst workload, and maintaining stringent data privacy. This approach addresses significant gaps in current methodologies, offering a robust solution to APT detection in compliance with privacy laws.

cs.CR↗

arXiv2023

Unscrambling the Rectification of Adversarial Attacks Transferability across Computer Networks

Convolutional neural networks (CNNs) models play a vital role in achieving state-of-the-art performances in various technological fields. CNNs are not limited to Natural Language Processing (NLP) or Computer Vision (CV) but also have substantial applications in other technological domains, particularly in cybersecurity. The reliability of CNN's models can be compromised because of their susceptibility to adversarial attacks, which can be generated effortlessly, easily applied, and transferred in real-world scenarios. In this paper, we present a novel and comprehensive method to improve the strength of attacks and assess the transferability of adversarial examples in CNNs when such strength changes, as well as whether the transferability property issue exists in computer network applications. In the context of our study, we initially examined six distinct modes of attack: the Carlini and Wagner (C&W), Fast Gradient Sign Method (FGSM), Iterative Fast Gradient Sign Method (I-FGSM), Jacobian-based Saliency Map (JSMA), Limited-memory Broyden fletcher Goldfarb Shanno (L-BFGS), and Projected Gradient Descent (PGD) attack. We applied these attack techniques on two popular datasets: the CIC and UNSW datasets. The outcomes of our experiment demonstrate that an improvement in transferability occurs in the targeted scenarios for FGSM, JSMA, LBFGS, and other attacks. Our findings further indicate that the threats to security posed by adversarial examples, even in computer network applications, necessitate the development of novel defense mechanisms to enhance the security of DL-based techniques.

cs.CR↗

arXiv2023

Systemization of Knowledge (SoK)- Cross Impact of Transfer Learning in Cybersecurity: Offensive, Defensive and Threat Intelligence Perspectives

Recent literature highlights a significant cross-impact between transfer learning and cybersecurity. Many studies have been conducted on using transfer learning to enhance security, leading to various applications in different cybersecurity tasks. However, previous research is focused on specific areas of cybersecurity. This paper presents a comprehensive survey of transfer learning applications in cybersecurity by covering a wide range of domains, identifying current trends, and shedding light on under-explored areas. The survey highlights the significance of transfer learning in addressing critical issues in cybersecurity, such as improving detection accuracy, reducing training time, handling data imbalance, and enhancing privacy preservation. Additional insights are provided on the common problems solved using transfer learning, such as the lack of labeled data, different data distributions, and privacy concerns. The paper identifies future research directions and challenges that require community attention, including the need for privacy-preserving models, automatic tools for knowledge transfer, metrics for measuring domain relatedness, and enhanced privacy preservation mechanisms. The insights and roadmap presented in this paper will guide researchers in further advancing transfer learning in cybersecurity, fostering the development of robust and efficient cybersecurity systems to counter emerging threats and protect sensitive information. To the best of our knowledge, this paper is the first of its kind to present a comprehensive taxonomy of all areas of cybersecurity that benefited from transfer learning and propose a detailed future roadmap to shape the possible research direction in this area.

cs.CR↗

arXiv2023

Machine Learning in Digital Forensics: A Systematic Literature Review

Development and exploitation of technology have led to the further expansion and complexity of digital crimes. On the other hand, the growing volume of data and, subsequently, evidence is a severe challenge in digital forensics. In recent years, the application of machine learning techniques to identify and analyze evidence has been on the rise in different digital forensics domains. This paper offers a systematic literature review of the research published in major academic databases from January 2010 to December 2021 on the application of machine learning in digital forensics, which was not presented yet to the best of our knowledge as comprehensive as this. The review also identifies the domains of digital forensics and machine learning methods that have received the most attention in the previous papers and finally introduces remaining research gaps. Our findings demonstrate that image forensics has obtained the greatest benefit from using machine learning methods, compared to other forensic domains. Moreover, CNN-based models are the most important machine learning methods that are increasingly being used in digital forensics. We present a comprehensive mind map to provide a proper perspective for valuable analytical results. Furthermore, visual analysis has been conducted based on the keywords of the papers, providing different thematic relevance topics. This research will give digital forensics investigators, machine learning developers, security researchers, and enthusiasts a broad view of the application of machine learning in digital forensics.

cs.CR↗

arXiv2022

Deep Fake Detection, Deterrence and Response: Challenges and Opportunities

According to the 2020 cyber threat defence report, 78% of Canadian organizations experienced at least one successful cyberattack in 2020. The consequences of such attacks vary from privacy compromises to immersing damage costs for individuals, companies, and countries. Specialists predict that the global loss from cybercrime will reach 10.5 trillion US dollars annually by 2025. Given such alarming statistics, the need to prevent and predict cyberattacks is as high as ever. Our increasing reliance on Machine Learning(ML)-based systems raises serious concerns about the security and safety of these systems. Especially the emergence of powerful ML techniques to generate fake visual, textual, or audio content with a high potential to deceive humans raised serious ethical concerns. These artificially crafted deceiving videos, images, audio, or texts are known as Deepfakes garnered attention for their potential use in creating fake news, hoaxes, revenge porn, and financial fraud. Diversity and the widespread of deepfakes made their timely detection a significant challenge. In this paper, we first offer background information and a review of previous works on the detection and deterrence of deepfakes. Afterward, we offer a solution that is capable of 1) making our AI systems robust against deepfakes during development and deployment phases; 2) detecting video, image, audio, and textual deepfakes; 3) identifying deepfakes that bypass detection (deepfake hunting); 4) leveraging available intelligence for timely identification of deepfake campaigns launched by state-sponsored hacking teams; 5) conducting in-depth forensic analysis of identified deepfake payloads. Our solution would address important elements of the Canada National Cyber Security Action Plan(2019-2024) in increasing the trustworthiness of our critical services.

cs.CR↗

arXiv2022

The Dichotomy of Cloud and IoT: Cloud-Assisted IoT From a Security Perspective

In recent years, the existence of a significant cross-impact between Cloud computing and Internet of Things (IoT) has lead to a dichotomy that gives raise to Cloud-Assisted IoT (CAIoT) and IoT-Based Cloud (IoTBC). Although it is pertinent to study both technologies, this paper focuses on CAIoT, and especially its security issues, which are inherited from both Cloud computing and IoT. This study starts with reviewing existing relevant surveys, noting their shortcomings, which motivate a comprehensive survey in this area. We proceed to highlight existing approaches towards the design of Secure CAIoT (SCAIoT) along with related security challenges and controls. We develop a layered architecture for SCAIoT. Furthermore, we take a look at what the future may hold for SCAIoT with a focus on the role of Artificial Intelligence(AI).

cs.CR↗

arXiv2022

Crypto Makes AI Evolve

Adopting cryptography has given rise to a significant evolution in Artificial Intelligence (AI). This paper studies the path and stages of this evolution. We start with reviewing existing relevant surveys, noting their shortcomings, especially the lack of a close look at the evolution process and solid future roadmap. These shortcomings justify the work of this paper. Next, we identify, define and discuss five consequent stages in the evolution path, including Crypto-Sensitive AI, Crypto-Adapted AI, Crypto-Friendly AI, Crypto-Enabled AI, Crypto-Protected AI. Then, we establish a future roadmap for further research in this area, focusing on the role of quantum-inspired and bio-inspired AI.

cs.IT↗

arXiv2022

Block Hunter: Federated Learning for Cyber Threat Hunting in Blockchain-based IIoT Networks

Nowadays, blockchain-based technologies are being developed in various industries to improve data security. In the context of the Industrial Internet of Things (IIoT), a chain-based network is one of the most notable applications of blockchain technology. IIoT devices have become increasingly prevalent in our digital world, especially in support of developing smart factories. Although blockchain is a powerful tool, it is vulnerable to cyber attacks. Detecting anomalies in blockchain-based IIoT networks in smart factories is crucial in protecting networks and systems from unexpected attacks. In this paper, we use Federated Learning (FL) to build a threat hunting framework called Block Hunter to automatically hunt for attacks in blockchain-based IIoT networks. Block Hunter utilizes a cluster-based architecture for anomaly detection combined with several machine learning models in a federated environment. To the best of our knowledge, Block Hunter is the first federated threat hunting model in IIoT networks that identifies anomalous behavior while preserving privacy. Our results prove the efficiency of the Block Hunter in detecting anomalous activities with high accuracy and minimum required bandwidth.

cs.CR↗

arXiv2022

Communication Layer Security in Smart Farming: A Survey on Wireless Technologies

Human population growth has driven rising demand for food that has, in turn, imposed huge impacts on the environment. In an effort to reconcile our need to produce more sustenance while also protecting the ecosystems of the world, farming is becoming more reliant on smart tools and communication technologies. Developing a smart farming framework allows farmers to make more efficient use of inputs, thus protecting water quality and biodiversity habitat. Internet of Things (IoT), which has revolutionized every sphere of the economy, is being applied to agriculture by connecting on-farm devices and providing real-time monitoring of everything from environmental conditions to market signals through to animal health data. However, utilizing IoT means farming networks are now vulnerable to malicious activities, mostly when wireless communications are highly employed. With that in mind, this research aims to review different utilized communication technologies in smart farming. Moreover, possible cyber attacks are investigated to discover the vulnerabilities of communication technologies considering the most frequent cyber-attacks that have been happened.

cs.CR↗

arXiv2021

Cybersecurity in Smart Farming: Canada Market Research

The Cyber Science Lab (CSL) and Smart Cyber-Physical System (SCPS) Lab at the University of Guelph conduct a market study of cybersecurity technology adoption and requirements for smart and precision farming in Canada. We conducted 17 stakeholder/key opinion leader interviews in Canada and the USA, as well as conducting extensive secondary research, to complete this study. Each interview generally required 15-20 minutes to complete. Interviews were conducted using a client-approved interview guide. Secondary and primary research focussed on the following areas of investigation: Market size and segmentation Market forecast and growth rate Competitive landscape Market challenges/barriers to entry Market trends/growth drivers Adoption/commercialization of the technology

cs.CR↗

arXiv2020

A Survey of Machine Learning Techniques in Adversarial Image Forensics

Image forensic plays a crucial role in both criminal investigations (e.g., dissemination of fake images to spread racial hate or false narratives about specific ethnicity groups) and civil litigation (e.g., defamation). Increasingly, machine learning approaches are also utilized in image forensics. However, there are also a number of limitations and vulnerabilities associated with machine learning-based approaches, for example how to detect adversarial (image) examples, with real-world consequences (e.g., inadmissible evidence, or wrongful conviction). Therefore, with a focus on image forensics, this paper surveys techniques that can be used to enhance the robustness of machine learning-based binary manipulation detectors in various adversarial scenarios.

cs.CR↗