Protecting On-Device AI Inference: A Systematic Review of Attacks and Defence Mechanisms

The need for secure and private Artificial Intelligence AI and Machine Learning ML on edge and mobile devices has increased the necessity of protecting the architecture of these systems from threats to both security and privacy. With an ever-increasing number of pre-trained AI models being used o...

Automatically Attacking Software Reverse Engineering AI Agents

Software tools for reverse engineering executable binary files, such as Ghidra, enable malware analysts to safely conduct robust static analysis without having access to original source code. Coupled with the analytic power of large language models LLM, agentic systems enabled with tools, such as...

Investigating Detection and Obfuscation of Prompt Injection Attacks against Software Reverse Engineering AI Agents

Agentic software reverse engineering systems are vulnerable to prompt injection attacks placed into the source code of executable binary files. This research demonstrates defensive tactics for detecting the presences of prompt injection strings in the decompiler output of adversarial example...

MIRAGE: Context-Aware Prompt Injection against Mobile GUI Agents Via User-Generated Content

Mobile graphical user interface GUI agents driven by vision-language models VLMs perceive the screen as rendered pixels and choose actions from what they see, so they cannot reliably separate trusted interface elements from user-generated content. We present MIRAGE Mobile Injection of Realistic...

A Surveillance Evasion Game with Continuous Sensor Redeployment Via Bilevel Optimization

Uncrewed Aerial Systems UASs have become a growing threat to the security of critical infrastructure, exploiting spatiotemporal gaps in sensor perimeters to infiltrate restricted airspace undetected. We formulate this interaction as a two-player zero-sum differential game between an adversarial U...

Disentangling Adversarial Prompts: A Semantic-Graph Defense for Robust LLM Security

Large Language Models LLMs are increasingly vulnerable to adversarial prompts that exploit semantic ambiguities to bypass safety mechanisms, resulting in harmful or inappropriate outputs. Such attacks, including jailbreaking and prompt injection, pose significant risks to the integrity and...

"What Is the Problem Space?" Defining Host-Space Adversarial Perturbations against Network Intrusion Detection Systems

Network Intrusion Detection Systems NIDS are now increasingly leveraging Machine Learning ML techniques to detect malicious network activities. Numerous papers have scrutinized the security of ML-based NIDS ML-NIDS by testing them against various attacks involving adversarial perturbations. The...

Building an Adversarial Malware Dataset by Family and Type: Generation, Evasion, and Poisoning Evaluation

We present a dataset of adversarial malware samples derived from the public RawMal-TF collection of real-world malware binaries. Using a suite of adversarial malware generators, we construct two sets of adversarial PE files: 44,347 family-labelled samples and 33,596 type-labelled samples, achievi...

Adversarial Vulnerability under Temporal Concept Drift: A Longitudinal Study of Android Malware Detection

We present a longitudinal, drift-aware evaluation of adversarial robustness across more than a decade of Android applications using static and dynamic feature representations extracted from emulator and real-device executions. The dataset is organized into yearly slices and evaluated under three...

Microsoft Open-Sources RAMPART and Clarity to Secure AI Agents During Development

Microsoft has unveiled two new open-source tools called RAMPART and Clarity to assist developers in better testing the security of artificial intelligence AI agents. RAMPART, short for Risk Assessment and Measurement Platform for Agentic Red Teaming, functions as a Pytest-native safety and securi...

Agent Security Is a Systems Problem

We take the position that agent security must be approached as a systems problem: the AI model powering the agent must be treated as an untrusted component, and security invariants must be enforced at the system level. Through this lens, efforts to increase model robustness the dominant viewpoint...

Devilray: A Systematic Adversarial Model Revealing Blind Spots in Fake Base Station Detection

Fake Base Station FBS detection has been a critical focus of cellular security research for over two decades. However, significant financial and regulatory barriers to accessing commercial FBS C-FBS devices have limited direct visibility into real-world operations, forcing detection systems to be...

Not What You Asked For: Typographic Attacks in Household Robot Manipulation

Open-vocabulary embodied AI agents increasingly rely on vision-language models such as CLIP for object perception and task grounding. However, the shared embedding space that enables this flexibility introduces a structural vulnerability to typographic attacks, where printed text in a physical...

A No-Defense Defense against Gradient-Based Adversarial Attacks on ML-NIDS: Is Less More?

Gradient-based adversarial attacks subtly manipulate inputs of Machine Learning ML models to induce incorrect predictions. This paper investigates whether careful architectural choices alone can yield an inherently robust Deep Neural Network DNN-based Network Intrusion Detection Systems NIDS,...

DarkLLM: Learning Language-Driven Adversarial Attacks with Large Language Models

While vision and multimodal foundation models underpin critical tasks from perception to complex reasoning, they remain highly vulnerable to adversarial attacks. However, traditional adversarial attacks are typically limited to single, predefined objectives, tightly coupling each attack to a...

Hidden in Memory: Sleeper Memory Poisoning in LLM Agents

Large language models are increasingly augmented with persistent memory, allowing assistants to store user-specific information across sessions for personalization and continuity. This statefulness introduces a new security risk: adversarial content can corrupt what an assistant remembers and...

WARD: Adversarially Robust Defense of Web Agents against Prompt Injections

Web agents can autonomously complete online tasks by interacting with websites, but their exposure to open web environments makes them vulnerable to prompt injection attacks embedded in HTML content or visual interfaces. Existing guard models still suffer from limited generalization to unseen...

Deserialization of Untrusted Data

Overview adversarial-robustness-toolbox is a Toolbox for adversarial machine learning. Affected versions of this package are vulnerable to Deserialization of Untrusted Data in the model loading process. An attacker can execute arbitrary code by uploading a maliciously crafted model file to an...

EUVD-2026-29553

The Adversarial Robustness Toolbox ART thru 1.20.1 contains a command-line argument injection vulnerability in its Kubeflow component robustnessevaluationfgsmpytorch.py. The script uses the unsafe eval function to parse string values provided via the --clipvalues and --inputshape command-line...

EUVD-2026-29552

The Adversarial Robustness Toolbox ART thru 1.20.1 contains an insecure deserialization vulnerability CWE-502 in its Kubeflow component's model loading functionality. When loading model weights from a file e.g., model.pt during robustness evaluation, the code uses torch.load without the...