Cybersecurity
Cybersecurity news — Friday 02 October 2026
138 stories from 8 sources, every one in English. Each links to its publisher.
-
Friday Squid Blogging: EU is Trying to Fight Unregulated Squid Fishing
Schneier on Security
The EU is recommending import controls to combat unregulated squid fishing in the Southwest Atlantic. I’m not optimistic. As usual, you can also use this squid post to talk about the security stories in the news that I haven’t covered…
-
iPhone 18 Pro Max stuck in SOS mode? Try AT＆T’s urgent fix ASAP
ZDNet Security
Some iPhone 18 Pro Max phones on AT&T are stuck in SOS mode. Two updates can prevent it, but affected phones need replacing.
-
Frontline Education breach exposes school district employee data
BleepingComputer
Frontline Education is notifying school districts of a data breach after attackers exploited a vulnerability in third-party software to gain unauthorized access to its systems and steal employee information, including Social Security…
-
Warlock ransomware breach SharePoint in water, telecom operator attacks
BleepingComputer
The China-linked ransomware group Warlock targeted a water utility, a telecom provider, a regional government body, and a university by exploiting SharePoint vulnerabilities to gain initial access. [...]
-
This new ChatGPT scam tricks you into installing malware – how to spot the trap
ZDNet Security
This authentic-looking scam starts with a sponsored Google link – and it’s catching people off-guard.
-
The legal questions raised by agentic AI hacks
CyberScoop Asia
As AI agents escaping testing sandboxes and hacking organizations have gone from unprecedented to seemingly routine in a matter of weeks, policymakers, regulators and cybersecurity attorneys largely agree on one thing: Something must be…
-
GitLab Patches Critical 9.9 AI Gateway Flaw Allowing Command Execution on Self-Hosted Servers
The Hacker News
A critical flaw in GitLab's AI Gateway could let a logged-in user with Duo Agent Platform access run commands on the gateway under certain conditions, GitLab said in an advisory. The gateway is the service that connects a GitLab instance…
-
Antino Backdoor Uses Outlook and OneDrive for C2 in China-Nexus Espionage Campaign
The Hacker News
Government and policy organizations across Asia have become the target of a new campaign orchestrated by a China-nexus threat actor. The activity, which has targeted government and policy organizations in Taiwan, India, the Philippines…
-
Dell CSM Flaws Enable Unauthenticated Admin Access and Root on Kubernetes Nodes
The Hacker News
Dell has released security updates to address multiple critical security flaws in Dell Container Storage Modules (CSM) that could be exploited by bad actors to take over susceptible systems. The vulnerabilities are listed below …
-
GitLab warns of critical RCE vulnerability in AI Gateway service
BleepingComputer
GitLab warned customers today to immediately patch a critical AI Gateway vulnerability that could let attackers run arbitrary commands on vulnerable instances. [...]
-
US sanctions Tren de Aragua gang members in ATM hacks crackdown
BleepingComputer
The U.S. Treasury Department has sanctioned eight members of the Venezuelan gang Tren de Aragua (TdA) for their role in the theft of millions of dollars in ATM jackpotting attacks across the United States. [...]
-
Google Wallet not working on your Pixel? 4 ways to fix tap-to-pay
ZDNet Security
Some Pixel users are having tap-to-pay problems, and there’s no fix from Google yet. But here a few things that helped others.
-
Unidentified Flock Cameras in Florida
Schneier on Security
St. Lucie County in Florida discovered ( alt link ) a dozen Flock cameras whose ownership it can’t identify, and that the county government had not permitted. I am reminded of the decade-old story of StingRay cell phone surveillance…
-
In Other News: $15K iCloud Spoofing Bugs, AI Policy Experts Phished, Adblocker Spies on AI Chats
SecurityWeek
Noteworthy stories that might have slipped under the radar: Kiteworks patches over 100 vulnerabilities, Microsoft publishes 2026 Digital Defense Report, AI finds 24 Android app flaws. The post In Other News: $15K iCloud Spoofing Bugs, AI…
-
The EDR blind spot: 3 ways browser attacks evade endpoint telemetry
BleepingComputer
Browser-based attacks can steal sessions, abuse extensions, or manipulate users without creating the endpoint artifacts EDR is designed to detect. NordLayer explains three ways attacks can evade endpoint telemetry and why browser-level…
-
macOS Users Targeted by Fake Zoom Installer Carrying CloudSyncD Backdoor
SecurityWeek
The dropper “carries a complete universal Mach-O inside itself, roughly 756 KB in the development build, and extracts it at runtime. The post macOS Users Targeted by Fake Zoom Installer Carrying CloudSyncD Backdoor appeared first on…
-
Dell asks admins to patch max severity CSM flaws as soon as possible
BleepingComputer
Dell has patched two maximum severity vulnerabilities in the Container Storage Modules (CSM) that connect Dell enterprise storage arrays to Kubernetes environments. [...]
-
OpenAI Parts Ways With Three Safety Researchers Over Sensitive Information Mishandling
The Hacker News
OpenAI has parted ways with three members of its safety team after they leaked private information in violation of company policies, The Wall Street Journal reported. "We have parted ways with three individuals for violating our policies…
-
Crypto Scammers Hijack Microsoft’s Official X Account
SecurityWeek
Hackers used the account, which has 13 million followers, to amplify a Clippy-themed cryptocurrency account. The post Crypto Scammers Hijack Microsoft’s Official X Account appeared first on SecurityWeek .
-
Why CISOs Struggle to Answer the Board's Three Hardest Questions, and How to Fix the Report
The Hacker News
The quarterly board meeting is two weeks out. The security team is pulling exports from the identity provider, the cloud posture tool, the vulnerability scanner, the SIEM and the EDR console. Someone is building a spreadsheet to reconcile…
-
In Rare Move, Alleged Iranian State Hacker Extradited to US
SecurityWeek
Amir Barati, an alleged member of the Mabna Institute, was indicted for targeting universities, private organizations, and government entities in the US and abroad. The post In Rare Move, Alleged Iranian State Hacker Extradited to US…
-
How American Political Campaigns Are Using AI—and What They’re Spending on the Tools
Schneier on Security
This essay was written with Nathan E. Sanders, and originally appeared in The Guardian . New campaign finance disclosure data shines a light on which US political campaigns are using AI tools and how much they are spending on them…
-
Warlock Expands SharePoint Exploitation in Critical Infrastructure Attacks
SecurityWeek
The China-based hacking group has been exploiting SharePoint vulnerabilities since July 2025. The post Warlock Expands SharePoint Exploitation in Critical Infrastructure Attacks appeared first on SecurityWeek .
-
Microsoft’s X account hacked in crypto pump-and-dump scheme
BleepingComputer
On Thursday, unknown attackers hijacked the official Microsoft account on X, which has over 13 million followers, in what appeared to be a pump-and-dump scheme promoting a crypto token. [...]
-
AI Agents Aimed SQL Injection at US and Canadian Government Sites
SecurityWeek
The attacks targeted the US Department of Education and Library and Archives Canada, and researchers linked some agents to OpenAI. The post AI Agents Aimed SQL Injection at US and Canadian Government Sites appeared first on SecurityWeek .
-
Exploited Fortinet FortiMail Zero-Day Calls for Urgent Action
SecurityWeek
CVE-2026-104286 is a critical-severity path traversal vulnerability that could allow attackers to write arbitrary files to the system. The post Exploited Fortinet FortiMail Zero-Day Calls for Urgent Action appeared first on SecurityWeek .
-
Android 17 Advanced Protection Locks Accessibility Services to Verified Accessibility Tools
The Hacker News
Google has announced a new security measure that limits access to Android's accessibility services to verified applications classified as Accessibility Tools when Advanced Protection is enabled. With malicious Android applications abusing…
-
Critical FortiMail Zero-Day Flaw Exploited in Attacks Allows Unauthenticated Arbitrary File Writes
The Hacker News
The U.S. Cybersecurity and Infrastructure Security Agency (CISA), on Thursday, added a critical security flaw impacting Fortinet FortiMail to its Known Exploited Vulnerabilities (KEV) catalog, following reports of active exploitation. The…
-
Walking the Embedding Space: Datastore Extraction from Multimodal RAG
arXiv cs.CR (Security)
arXiv:2610.01871v1 Announce Type: new Abstract: Multimodal Retrieval-Augmented Generation (MRAG) has emerged as a reliable and cost-effective technique of grounding the generative capabilities of Multimodal Large Language Models (MLLMs)…
-
A Hybrid Approach to Malware Detection: Integrating Few-Shot Model-Agnostic Meta-Learning with Autoencoders
arXiv cs.CR (Security)
arXiv:2610.01949v1 Announce Type: new Abstract: Ransomware has emerged as a major cybersecurity threat, with incidents increasing in frequency and impact across critical sectors. These attacks are typically launched through phishing…
-
Fault-Tolerant Budget Conservation in Distributed Multi-Agent Delegation
arXiv cs.CR (Security)
arXiv:2610.00349v1 Announce Type: cross Abstract: Resource limits are becoming an authorization boundary for AI agents that delegate work across concurrent and failure-prone workers. Parent-child allocation constraints, affine objects, and…
-
DeBERTa-ConPara: Attack-Aware and Deployment-Realistic Detection of AI-Generated Text
arXiv cs.CR (Security)
arXiv:2610.00883v1 Announce Type: cross Abstract: Robust detection of AI-generated text under deployment conditions is challenging: distribution shifts across domains and generators, adversarial perturbations of the input surface, and the…
-
Learnt Attacks on Quantum Key Distribution under Channel Noise and Device Drift
arXiv cs.CR (Security)
arXiv:2610.01792v1 Announce Type: cross Abstract: Quantum key distribution (QKD) links are provisioned from security analyses of stationary channels, whereas the devices that determine the channel drift between recalibrations. Whether an…
-
Rethinking Anonymity Claims in Synthetic Data Generation: A Model-Centric Privacy Attack Perspective
arXiv cs.CR (Security)
arXiv:2601.22434v2 Announce Type: replace Abstract: Training generative machine learning models to produce synthetic tabular data has become a popular approach for enhancing privacy in data sharing. As this typically involves processing…
-
TensorCommitments: A Lightweight Verifiable Inference for Language Models
arXiv cs.CR (Security)
arXiv:2602.12630v2 Announce Type: replace Abstract: Most large language models (LLMs) run on external clouds: users send a prompt, pay for inference, and must trust that the remote GPU executes the LLM without any adversarial tampering. We…
-
HarnessAgent: Scaling Automatic Fuzzing Harness Construction with Tool-Augmented LLM Pipelines
arXiv cs.CR (Security)
arXiv:2512.03420v4 Announce Type: replace Abstract: Large language model (LLM)-based techniques have achieved notable progress in generating harnesses for program fuzzing. However, applying them to arbitrary functions (especially internal…
-
The Next Challenge for Agentic Cybersecurity: A Realistic, Contamination-Free Reverse Engineering Benchmark
arXiv cs.CR (Security)
arXiv:2608.11469v2 Announce Type: replace Abstract: AI agents are rapidly improving in cybersecurity when source code is available, yet much of the software most consequential to security, including malware, firmware, and proprietary…
-
VideoSTF: Stress-Testing Output Repetition in Video Large Language Models
arXiv cs.CR (Security)
arXiv:2602.10639v2 Announce Type: replace-cross Abstract: Video Large Language Models (VideoLLMs) have achieved strong performance on video understanding tasks, yet existing benchmarks evaluate only what models predict, leaving the…
-
Sleeping Secrets: How Fine-Tuning Reawakens Privacy Risks in Language Models
arXiv cs.CR (Security)
arXiv:2610.01365v1 Announce Type: new Abstract: Beyond adapting Large Language Models (LLMs) to specialized applications, fine-tuning has recently been shown to recover private information that is no longer accessible through direct…
-
System-Level Optimization Beyond Cryptographic Kernels: An ML-KEM Case Study on Arm Cortex-M7
arXiv cs.CR (Security)
arXiv:2610.01960v1 Announce Type: new Abstract: Recent work on embedded post-quantum cryptography has focused primarily on instruction-level optimization, including arithmetic-kernel improvements, assembly tuning, register allocation, and…
-
Supersingularity and Superspeciality Verification of Abelian Surfaces
arXiv cs.CR (Security)
arXiv:2610.01924v1 Announce Type: cross Abstract: Supersingular abelian surfaces are essential in isogeny-based cryptography. Despite this, we have no efficient algorithm to verify if a given abelian surface is supersingular. In this work…
-
KaliBench: A Fine-Grained Benchmark for Cybersecurity Tool Use on Kali Linux with Runtime-Free Verifiable Rewards
arXiv cs.CR (Security)
arXiv:2610.02206v1 Announce Type: cross Abstract: LLMs are increasingly applied to cybersecurity workflows, where they are expected to translate analysts' intent into tool invocations. However, existing evaluations focus on knowledge-based…
-
Actions with Receipts: Jointly Binding Claims, Evidence, and Execution for Replayable Tool-Agent Auditing
arXiv cs.CR (Security)
arXiv:2610.00327v1 Announce Type: new Abstract: Tool-using agents can expose citations and execution logs while leaving a critical association unaudited: whether the claim shown to a user is the claim emitted by the committed execution and…
-
High-quality Data Do not Mean Safe! Poisoning LLMs after Data Selection
arXiv cs.CR (Security)
arXiv:2610.01367v1 Announce Type: new Abstract: Safety-aligned Large Language Models remain vulnerable to fine-tuning on small sets of harmful or benign-looking samples. However, prior studies typically assume that poisoned samples…
-
False Floors: LLM Safety Routing Evaluations Break Under Distribution Shift
arXiv cs.CR (Security)
arXiv:2610.01535v1 Announce Type: new Abstract: Safety routers send each request to one of several models and are judged against the best single model. A major routing benchmark picks that comparator on the evaluation data. In the…
-
Refusal Localizes, the Damage Relocates: Safety Layers Under Few-Sample Fine-Tuning
arXiv cs.CR (Security)
arXiv:2610.00320v1 Announce Type: cross Abstract: Fine-tuning adapts aligned large language models (LLMs) to downstream tasks, but a few dozen harmful examples can remove their refusal of harmful requests. Prior work localizes…
-
Multi-Jurisdictional Legal Identity Assurance for Capability Gating: A Design-Science Proposal for Tiered, Reusable Identity Assurance of Natural, Juridical, and Machine Entities
arXiv cs.CR (Security)
arXiv:2610.00287v1 Announce Type: cross Abstract: Identity assurance is the cost a digital system pays for dishonesty and uncertainty: it exists to make acts attributable when not everyone can be trusted at their word. A common way to pay…
-
PSR2: A Phase-based Semantic Reasoning Framework for Atomicity Violation Detection via Contract Refinement
arXiv cs.CR (Security)
arXiv:2604.06975v3 Announce Type: replace Abstract: With the rapid advancement of decentralized applications, smart contract security faces severe challenges, particularly regarding atomicity violations in complex logic such as Oracle and…
-
Analyzing Defensive Misdirection Against Model-Guided Automated Attacks on Agentic AI Systems
arXiv cs.CR (Security)
arXiv:2606.20470v4 Announce Type: replace Abstract: Agentic AI systems increasingly rely on language-model components to interpret instructions, process external data, invoke tools, and coordinate with other agents. These capabilities make…
-
Evasion Attacks: How Adversarial Noise Bypasses ML Classifiers
arXiv cs.CR (Security)
arXiv:2610.00136v1 Announce Type: new Abstract: This paper presents a reproducible, educational study of evasion attacks in image classification and text classification. A compact convolutional network trained on MNIST reached 98.63% clean…
-
Coherence Rather Than Error Rate Governs Privacy in Multi-Tenant Quantum Computing
arXiv cs.CR (Security)
arXiv:2609.34411v2 Announce Type: replace-cross Abstract: Multi-tenant computing enables providers of commercial cloud quantum processors to rent disjoint sectors of a device to independent users. Average gate error, which cloud quantum…
-
Tight Post-Quantum Parallel Repetition for Private-Coin Arguments
arXiv cs.CR (Security)
arXiv:2609.38941v2 Announce Type: replace-cross Abstract: We show that assuming the existence of homomorphic encryption, parallel repetition of all interactive arguments (after being run under homomorphic encryption) reduces the soundness…
-
Alignment via Training Against Probes Without Losing Monitorability
arXiv cs.CR (Security)
arXiv:2609.38645v2 Announce Type: replace-cross Abstract: Models are usually aligned based on their observed outputs, using demonstrations, preference data, or reward signals. These objectives reward responses that look aligned. More…
-
Quantum Advantage for Two-Party Differential Privacy
arXiv cs.CR (Security)
arXiv:2610.02113v1 Announce Type: cross Abstract: We introduce information-theoretically private quantum protocols for two-party Hamming distance when both parties must output the same estimate. Classically, for input length $n$…
-
Homomorphic Advantage Operator: Stabilizing Reinforcement Learning Under Fully Homomorphic Encryption Constraints
arXiv cs.CR (Security)
arXiv:2610.02074v1 Announce Type: cross Abstract: Privacy-preserving machine learning presents significant deployment challenges on the cloud for intelligent systems with confidential data. Fully Homomorphic Encryption (FHE) offers a…
-
In Vino Veritas and Vulnerabilities: Examining LLM Safety via Drunk Language Inducement
arXiv cs.CR (Security)
arXiv:2601.22169v2 Announce Type: replace-cross Abstract: Humans are susceptible to undesirable behaviours and privacy leaks under the influence of alcohol. This paper investigates drunk language, i.e., text written under the influence of…
-
Evidence Coverage for Intent-Bound Execution: Scope, Obligations, and Cutoff Reasoning
arXiv cs.CR (Security)
arXiv:2610.01386v1 Announce Type: new Abstract: A verifier may authenticate every available record and still lack grounds to call an execution account complete. Such a claim requires a justified account of which records were due for the…
-
SoK: Decentralized Agent Economic Infrastructure
arXiv cs.CR (Security)
arXiv:2610.01756v1 Announce Type: new Abstract: Decentralized agent economies increasingly build a single task from protocols that were designed and secured separately. This creates a simple problem: a workflow can look correct at each…
-
Federated Detection of Open Charge Point Protocol 1.6 Cyberattacks
arXiv cs.CR (Security)
arXiv:2502.01569v2 Announce Type: replace Abstract: The ongoing electrification of the transportation sector requires the deployment of multiple Electric Vehicle (EV) charging stations across multiple locations. However, the EV charging…
-
Geometric Data Perturbation with Noisy-Anchor Alignment for Privacy-Preserving Collaborative Learning
arXiv cs.CR (Security)
arXiv:2608.18749v2 Announce Type: replace-cross Abstract: Geometric data perturbation enables one-shot representation sharing for privacy-preserving collaborative learning: each participant applies a secret distance-preserving…
-
A Safe Prototype Is Not a Safety Direction: Reference Dependence and Prompt Confounds in Response-Safety Embeddings
arXiv cs.CR (Security)
arXiv:2610.01801v1 Announce Type: cross Abstract: Can response safety be scored by cosine similarity to the mean embedding of known-safe responses? A recent sleeper-agent detector proposes exactly this score, yet the raw positive-centroid…
-
Backdoor Containment via Expert Quarantine and Shutdown in LLMs
arXiv cs.CR (Security)
arXiv:2610.00663v1 Announce Type: cross Abstract: Backdoored large language models (LLMs) can behave normally on benign inputs while producing attacker-specified outputs under hidden triggers. Existing defenses span four…
-
Characterizing and Codifying Malware Sophistication
arXiv cs.CR (Security)
arXiv:2610.00098v1 Announce Type: new Abstract: 'Sophisticated' is widely used to describe malware, yet it lacks a consistent definition within academic literature. While existing software quality and complexity metrics offer some insight…
-
UnifiedAttack: Evaluating the Safety of Large Multimodal Models in Synergistic Harmful Image-Text Generation
arXiv cs.CR (Security)
arXiv:2610.00341v1 Announce Type: new Abstract: As Large Multimodal Models (LMMs) transition toward natively unified architectures, evaluating their safety in synergistic harmful image-text generation tasks becomes a critical challenge…
-
SAGE: Similarity-Based Cleaning of Poisoned Training Data from Verified Examples
arXiv cs.CR (Security)
arXiv:2610.01788v1 Announce Type: cross Abstract: As machine learning increasingly relies on public, untrusted data sources, data poisoning attacks, which inject malicious examples into training data to induce misclassification of a chosen…
-
GNSS Spoofing in Mobile Devices: A Survey on Impact and Countermeasures
arXiv cs.CR (Security)
arXiv:2610.01294v1 Announce Type: new Abstract: Smartphones rely on Global Navigation Satellite System (GNSS)-based positioning for many of the functions they execute everyday. The GNSS receivers embedded in smartphones are susceptible to…
-
Runtime Assurance Under Measurement Attack: Necessary and Sufficient Observability Conditions for Learned Control in Radio Access Networks
arXiv cs.CR (Security)
arXiv:2610.00285v1 Announce Type: cross Abstract: Runtime assurance pairs a verified fallback with an untrusted controller and a switching monitor, and is the leading route to admitting learned policies into safety-relevant network…
-
SAFE: Spatially-Aware Feedback Enhancement for Fault-Tolerant Trust Management in Event-Based VANETs
arXiv cs.CR (Security)
arXiv:2604.07552v2 Announce Type: replace-cross Abstract: In event-based trust management for vehicular ad hoc networks (VANETs), vehicles that witness a road event broadcast its state, and vehicles that later witness the same event score…
-
Can AI Oversight Be Zero Knowledge?
arXiv cs.CR (Security)
arXiv:2610.01995v1 Announce Type: cross Abstract: AI systems increasingly produce outputs from confidential data, such as a fitness-for-duty assessment from medical records or the predicted properties of a drug candidate from its secret…
-
Intrusion Detection for Agentic Processes: Evidence-Based Runtime Monitoring
arXiv cs.CR (Security)
arXiv:2610.00151v1 Announce Type: new Abstract: Agent deployments increasingly combine language-model inference with retrieval, delegation, tool execution, external-system access, and human approval. Security-relevant deviations can…
-
Do Defenses Against LLM Extraction Work Across Attacks? A Lifecycle Benchmark of Black-Box Model Extraction
arXiv cs.CR (Security)
arXiv:2610.00839v1 Announce Type: new Abstract: Large language models (LLMs) deployed through text-only APIs face model extraction risks, as adversaries can collect their responses to train surrogates that reproduce their capabilities…
-
Jev-IDS: System One Models for Network Intrusion Detection
arXiv cs.CR (Security)
arXiv:2610.01079v1 Announce Type: new Abstract: Machine-learning Network Intrusion Detection Systems (IDS) depend on substantial labeled datasets and task-specific training, whereas Large Language Models (LLMs) detection can analyze flow…
-
A Systematization of Knowledge on DeFi Vaults: Architectures, Curation Mechanisms, and Strategy Design
arXiv cs.CR (Security)
arXiv:2610.01300v1 Announce Type: new Abstract: Decentralized finance (DeFi) vaults are smart-contract-based asset management systems that pool deposits, execute programmable strategies, and mint tokenized shares representing claims on…
-
The Innocent Courier: Covert Exfiltration Through Legitimate LLM Web Fetching
arXiv cs.CR (Security)
arXiv:2610.01768v1 Announce Type: new Abstract: With the increasing capabilities of Large-Language-Models (LLMs) and LLM-based agents, users are increasingly using them to solve everyday problems, such as answering e-mails or providing…
-
Backdoor Purification for LoRA-Tuned LLMs via Null-Space Projection
arXiv cs.CR (Security)
arXiv:2610.00685v1 Announce Type: cross Abstract: With the rapid adoption of large language models (LLMs) and parameter-efficient fine-tuning (PEFT) methods, the risk of backdoor attacks has become more severe. Existing backdoor…
-
Device-Independent Conference Keys from Parity-Extended Games
arXiv cs.CR (Security)
arXiv:2610.01025v1 Announce Type: cross Abstract: Device-independent conference key agreement (DI-CKA) lets a group of parties establish a shared secret key from untrusted quantum devices, with security certified by non-locality. Existing…
-
Evaluating Neural Decompilation of Dart AOT Binaries: Fine-Tuning, Metric Validity, Specification Leakage, and Reliability
arXiv cs.CR (Security)
arXiv:2607.06125v2 Announce Type: replace-cross Abstract: We present an execution-based evaluation of neural decompilation for Dart ahead-of-time binaries and an audit of what its scores measure. Across six archived adapter-baseline…
-
ZoneClaw: Mitigating Persistent Memory Attacks by Establishing Memory-Zoning in OpenClaw-Style Computer-Use Agents
arXiv cs.CR (Security)
arXiv:2610.00450v1 Announce Type: new Abstract: Computer-use agents increasingly operate as long-running assistants through persistent workspace memory, which OpenClaw-style CUAs realize as automatically reloaded files that hold user…
-
From Network Intrusion Detection to Blockchain-Backed Endpoint Detection and Response: Mapping the Landscape of Decentralized Detection-and-Response Architectures
arXiv cs.CR (Security)
arXiv:2610.01872v1 Announce Type: new Abstract: While the literature on blockchain-assisted intrusion detection and prevention systems (IDS/IPS) for Internet of Things (IoT) and Industrial Internet of Things (IIoT) networks is mature…
-
A Resource-Aware Behavior Reconstruction and Hierarchical Semantic Learning Framework for Host Intrusion Detection
arXiv cs.CR (Security)
arXiv:2610.01250v1 Announce Type: new Abstract: System calls (syscalls) record key interactions between running programs and the operating system kernel, providing fine-grained and minimally intrusive data for host-based intrusion…
-
Detection and Resolution of Periodic Artifacts in OpenDP's Discrete Laplace Sampler
arXiv cs.CR (Security)
arXiv:2610.01907v1 Announce Type: new Abstract: Differential privacy implementations rely on precise sampling from noise distributions to provide formal privacy guarantees. We report the discovery of systematic artifacts in OpenDP's…
-
Pseudoentanglement in constant depth: How trivial states can have non-trivial entanglement structure
arXiv cs.CR (Security)
arXiv:2605.31448v2 Announce Type: replace-cross Abstract: We construct a family of 2D-local constant-depth quantum circuits that output states whose entanglement entropy across a specified cut cannot be estimated in quantum polynomial…
-
AuraForge: Scaling Security Supervision for Training Coding Agents
arXiv cs.CR (Security)
arXiv:2610.00850v1 Announce Type: new Abstract: Coding agents are now proficient enough to generate complex software applications from a single prompt. As their capabilities have grown, human oversight has increasingly shifted from…
-
Autonomous OSS Threat Detection via Taxonomy-Aligned LLMs
arXiv cs.CR (Security)
arXiv:2610.01263v1 Announce Type: new Abstract: Open source software (OSS) ecosystems face growing threats from sophisticated supply chain attacks including typosquatting, dependency confusion, Trojan Source obfuscation, malicious build…
-
Is it Possible to Generate Irreversible PolyProtected Templates from Face Embeddings using System-Specific Keys?
arXiv cs.CR (Security)
arXiv:2610.01385v1 Announce Type: new Abstract: This work aims to answer the question of whether it is possible to generate irreversible protected templates when the PolyProtect biometric template protection method is applied to face…
-
SafeDepth: Safety-Aware Token-Level Adaptive Computation
arXiv cs.CR (Security)
arXiv:2610.00815v1 Announce Type: new Abstract: Recent studies suggest that not every token needs to pass through all Transformer layers, motivating token-level adaptive models that selectively skip layers to reduce computation. Our…
-
Quantum Secret Sharing and Error Correction vs No-Cloning
arXiv cs.CR (Security)
arXiv:2610.00441v1 Announce Type: cross Abstract: Secret sharing is ubiquitous throughout cryptography. All possible access structures are classically feasible, and in the case of threshold access structures, the protocols are even very…
-
PACE: Provenance-Aware Capability Enforcement for Tool-Using LLM Agents
arXiv cs.CR (Security)
arXiv:2610.01349v1 Announce Type: new Abstract: Tool-using large language model (LLM) agents turn generated text into real side effects, so poisoned tool metadata, retrieved pages, memory, and reusable skills can steer the next call…
-
Key-Reuse Vulnerability of Phase-Keyed Fourier-Curve Modulation: Relation Leakage and Key-Refresh Cost on Coded Links
arXiv cs.CR (Security)
arXiv:2610.01484v1 Announce Type: new Abstract: The security of keyed modulation is often argued from the key-space size and the error rate of a key-less receiver. This evidence fails when the key is reused and the waveform is harmonically…
-
Chaining Skills to Hijack LLM Agents
arXiv cs.CR (Security)
arXiv:2610.01564v1 Announce Type: new Abstract: LLM agents use skills to improve performance on specialized tasks. To complete a user request, an agent may invoke several skills in sequence, allowing information produced under one skill to…
-
Verifiable quantum advantage in extremely low depth
arXiv cs.CR (Security)
arXiv:2609.01448v2 Announce Type: replace-cross Abstract: We give a sampling problem that is solvable by shallow quantum circuits, hard for polynomial-time classical algorithms under lattice-based assumptions, and efficiently verifiable by…
-
Robust and leakage-resilient device-independent oblivious transfer in MiniQCryp
arXiv cs.CR (Security)
arXiv:2610.01421v1 Announce Type: cross Abstract: Assuming post-quantum one-way functions, we construct device-independent (DI) oblivious transfer (OT) and bit commitment: honest parties use only trusted classical computation to operate…
-
A Structured State Space Sequence Model for Multi-Class Classification of Malware
arXiv cs.CR (Security)
arXiv:2610.01893v1 Announce Type: new Abstract: By 2030, Internet of Things (IoT) devices are projected to reach 40 billion, with fast-paced technological advancements in fields such as industry, healthcare, agriculture, automobiles, and…
-
On the pseudorandomness of simple quantum processes
arXiv cs.CR (Security)
arXiv:2610.02100v1 Announce Type: cross Abstract: Can simple processes appear highly complex? Gowers (Comb. Prob. Comp. '96) conjectured that repeatedly composing local random reversible operations can yield global permutations that are…
-
Time-space lower bounds for breaking quantum cryptography
arXiv cs.CR (Security)
arXiv:2610.02101v1 Announce Type: cross Abstract: We prove near-optimal time-space lower bounds for breaking quantum cryptography in the random oracle model. Specifically, we show that a $T$-query adversary with $S$ qubits of non-uniform…
-
Safety Must Survive Self-Improvement: Why Failures Persist and How Agents Recover
arXiv cs.CR (Security)
arXiv:2610.01073v1 Announce Type: cross Abstract: Recursive self-improvement (RSI) allows agents to carry useful changes across generations. Maintaining safety across these generations involves both preventing unsafe behavior from…
-
SyzHarness: Patch-Based Kernel Bug Reproduction with LLM-Synthesized Fuzzing Harnesses
arXiv cs.CR (Security)
arXiv:2609.23889v3 Announce Type: replace Abstract: Automated kernel vulnerability reproduction is essential for bug triage, patch validation, and regression testing, but still lacks an effective and efficient solution. The core challenge…
-
The Achilles' Heel of Partial Reconfiguration: Optical Side-Channel Leakage on the 7-Series ICAP
arXiv cs.CR (Security)
arXiv:2610.01736v1 Announce Type: new Abstract: Major FPGA manufacturers have incorporated bitstream encryption to protect sensitive configuration data. However, for the most widely used FPGA families, multiple attacks against unpatchable…
-
On Reliability of Membership Inference Vulnerability Evaluation
arXiv cs.CR (Security)
arXiv:2605.25819v2 Announce Type: replace-cross Abstract: Membership inference attacks (MIAs) are popular methods for empirically assessing the leakage of sensitive information in the training data through models or statistics learned from…
-
Helol Tunnel: Covert Channel Exploitation of TLS Extensibility ＆ Privacy Features
arXiv cs.CR (Security)
arXiv:2610.01009v1 Announce Type: new Abstract: Covert channels exploiting network protocols for data exfiltration and command-and-control (C2) are integral parts of modern cyberattacks. In search of a significant covert channel within the…
-
Identity-Bound Governance Under Execution Uncertainty: An Accountability Proof Block for LLM Agent Persistent Halts, with Cryptographic Implementation and Cross-Model Calibration
arXiv cs.CR (Security)
arXiv:2610.00787v1 Announce Type: new Abstract: A correctly governed LLM agent can reach a state in which neither continuing execution nor automatically halting is admissible: the system has detected a persistent failure of its…
-
The Cognitive Continuity Test: Verifying Governed State Transitions in Persistent AI Agents
arXiv cs.CR (Security)
arXiv:2610.00132v1 Announce Type: new Abstract: Persistent AI agents revise beliefs, consolidate memory, and replace execution substrates. Similar successor states can accompany differently authorized transition claims, while legitimate…
-
Stability Buys Time: A Re-Keying Game for Encrypted Multi-Agent Control
arXiv cs.CR (Security)
arXiv:2607.12742v2 Announce Type: replace Abstract: Encrypted control lets a cloud coordinate a fleet of agents on fully homomorphically encrypted state, keeping their positions and commands private. The approximate scheme for real-valued…
-
On the Relationship between Model Quantization and Model Inversion Attacks
arXiv cs.CR (Security)
arXiv:2610.00382v1 Announce Type: new Abstract: Model quantization reduces the numerical precision of neural network weights and activations to lower storage and computational costs. Model inversion attacks recover or reconstruct sensitive…
-
QUFIG: GNN-Based Prediction of Quantum Fault Injection Vulnerabilities with Gate-Level Precision
arXiv cs.CR (Security)
arXiv:2610.01777v1 Announce Type: cross Abstract: The growing scale and accessibility of quantum hardware exposed new reliability and security challenges in the quantum computing workflow, such as the run-time fault injection attacks in…
-
Towards the Impossibility of Imperfectly Complete Key Agreement in the QROM
arXiv cs.CR (Security)
arXiv:2608.17610v2 Announce Type: replace-cross Abstract: We make progress towards the impossibility of quantum-computation, classical-communication (QCCC) key agreement by giving the first unconditional polynomial-query attacks that…
-
From A2A Attacks to Envelope-Layer Defense: Red-Teaming Evaluation of LLM Agents and a Three-Layer Isomorphic Attack-Defense Model
arXiv cs.CR (Security)
arXiv:2610.00392v1 Announce Type: new Abstract: Agent interaction protocols such as ACP and A2A have moved LLM-based agents toward multi-agent collaboration, introducing new security threats. A task sent by a remote peer over A2A is…
-
Authorization for Self-Modifying AI Agent Populations: Conserving Authority across Replacement, Forking, and Rollback
arXiv cs.CR (Security)
arXiv:2610.00347v1 Announce Type: new Abstract: Self-modifying AI agents can replace, fork, and roll back identity-bearing software while descendants remain executable. Per-successor authorization does not constrain the resulting…
-
Crossing the Cyber Divide: Sim-to-Sim and Sim-to-Real Transfer for RL Agents
arXiv cs.CR (Security)
arXiv:2610.00759v1 Announce Type: new Abstract: Cyber attack agents are typically trained and evaluated within a single simulator, making it unclear whether learned policies transfer beyond the environments in which they were developed…
-
Trapdoored Clifford Operators and Applications
arXiv cs.CR (Security)
arXiv:2610.01848v1 Announce Type: cross Abstract: Random Clifford operators have numerous applications in quantum computing, including randomized benchmarking, classical shadows, and quantum authentication. However, sampling and…
-
False Prophets: On the Security of World Models in Agentic Systems
arXiv cs.CR (Security)
arXiv:2607.23147v2 Announce Type: replace Abstract: Large language models now power autonomous agents capable of complex, multi-step tasks in different environments. Accurate and reliable execution of these tasks requires the agent to…
-
Representation Transitions Reveal Emerging Safety Risks in Multi-Turn LLM Agents
arXiv cs.CR (Security)
arXiv:2610.00400v1 Announce Type: cross Abstract: Multi-turn attacks on agentic systems can compose individually permissible actions into harmful outcomes, challenging defenses that assess actions or states in isolation. We show that such…
-
Towards Hierarchical Cyber Defense with Large Language Models: From Planning to Execution
arXiv cs.CR (Security)
arXiv:2610.00590v1 Announce Type: new Abstract: An autonomous cyber defender trained with reinforcement learning (RL) is typically tied to the network on which it was trained, limiting its ability to generalize as network scale changes…
-
Blind Unforgeability implies Plus-One Unforgeability
arXiv cs.CR (Security)
arXiv:2610.00517v1 Announce Type: cross Abstract: We prove that quantum blind-unforgeability implies quantum plus-one unforgeability, settling an open question in the literature. Since plus-one unforgeability is known not to imply…
-
TESLA-for-5G: Broadcast Authentication for 5G Networks Using TESLA
arXiv cs.CR (Security)
arXiv:2606.26528v2 Announce Type: replace Abstract: 5G base stations broadcast unauthenticated system information (SI) that every user equipment (UE) reads during cell selection. This enables attackers to broadcast forged SI from a fake…
-
Dithered Gaussian Mechanism for Randomness-Efficient Differential Privacy
arXiv cs.CR (Security)
arXiv:2607.06320v3 Announce Type: replace Abstract: We present the dithered Gaussian mechanism, an alternative to the discrete Gaussian mechanism for differential privacy that discretizes the private output rather than the noise…
-
Protocol Integration of Physical Layer Deception into EAP-TEAP Wi-Fi Authentication
arXiv cs.CR (Security)
arXiv:2610.01580v1 Announce Type: new Abstract: Credential-based Extensible Authentication Protocol (EAP) authentication cannot distinguish a legitimate credential holder from an adversary using compromised credentials. Physical Layer…
-
OverAct: Measuring and Mitigating Proactive Over-Authorization in LLM Tool-Calling Agents
arXiv cs.CR (Security)
arXiv:2610.01508v1 Announce Type: new Abstract: LLM agents with tool-calling capabilities can access external services and private user data, but they may retrieve more information than a user's request explicitly requires. We study this…
-
Progressive-Resolution Secure Aggregation for Federated Learning
arXiv cs.CR (Security)
arXiv:2610.00695v1 Announce Type: new Abstract: Secure aggregation lets a server recover an aggregate of client updates without observing any individual update, but conventional protocols fix the aggregate precision when clients upload. We…
-
Made to Measure: Designing Image Watermarks to Specification
arXiv cs.CR (Security)
arXiv:2610.00780v1 Announce Type: new Abstract: Image watermarking supports provenance and attribution by embedding verifiable identity information into images. Practical deployments, however, must jointly satisfy requirements for attack…
-
Decisional Monogamy-of-Entanglement for Coset States and Applications to Unclonable Cryptography with Correlated Challenges
arXiv cs.CR (Security)
arXiv:2608.18841v2 Announce Type: replace-cross Abstract: A main application of quantum information in cryptography is copy-protection, where we encode a functionality (such as a decryption key or software) into a reusable quantum state so…
-
No One Architecture Fits All: A Cross-Environment Evaluation of Hierarchical Red Team Agents
arXiv cs.CR (Security)
arXiv:2610.00557v1 Announce Type: new Abstract: Autonomous red team agents increasingly stress-test AI-enabled cyber defenses by planning strategy and executing multistage attacks. Reinforcement learning (RL) and large language models…
-
SURE: Framework for Safety to Construct Trustworthy AI
arXiv cs.CR (Security)
arXiv:2609.38249v2 Announce Type: replace Abstract: Warning: This paper contains harmful and offensive text. Recently, large language models such as GPT-4, and Claude have revolutionized tasks in various domains. As the use of these large…
-
Incentives and Outcomes in Bug Bounties
arXiv cs.CR (Security)
arXiv:2509.16655v2 Announce Type: replace-cross Abstract: Bug bounty programs have contributed significantly to security in technology firms in the last decade, but little is known about the role of reward incentives in producing useful…
-
Deny Without Disabling: Authorization-Paired Evaluation and Control for Multi-Agent Systems
arXiv cs.CR (Security)
arXiv:2610.00371v1 Announce Type: cross Abstract: Multi-agent systems derive their capabilities from sharing evidence, delegating tasks, and combining information across agents. The same process creates a safety problem: contributions that…
-
ReCast: Contract-Preserving Protection for Fixed-Interface Multimodal Reasoning
arXiv cs.CR (Security)
arXiv:2610.01184v1 Announce Type: new Abstract: Remote multimodal models offer strong numerical reasoning capabilities over charts and speech, but sending private inputs risks exposing sensitive content. Text-only sanitization cannot…
-
Sapien: A Stateful Policy Engine for Autonomous AI Agents
arXiv cs.CR (Security)
arXiv:2610.00797v1 Announce Type: cross Abstract: Contextual security defenses prevent AI agents from taking rogue actions by synthesizing a task-specific policy and enforcing it on the agent's tool calls. In multi-step tasks, however…
-
Combining Homomorphic Encryption and Differential Privacy in Federated Learning for Model Inspection and Availability
arXiv cs.CR (Security)
arXiv:2610.01650v1 Announce Type: new Abstract: The increasing prevalence of decentralized data has led to a growing interest in federated learning, which enables collaborative model training without clients sharing their sensitive local…
-
Harbormaster: Evidence-Gated, Replay-Safe Maritime Anomaly Detection on AWS
arXiv cs.CR (Security)
arXiv:2610.00519v1 Announce Type: new Abstract: Ships broadcast their positions through the Automatic Identification System (AIS), and those reports can be false or missing. An operator who acts on an anomaly alert needs that alert to be…
-
Safety in Self-Evolving Agents: A Survey
arXiv cs.CR (Security)
arXiv:2610.00093v1 Announce Type: new Abstract: Large language models (LLMs) exhibit strong general capabilities, yet their parameters typically remain fixed after deployment, limiting learning from new interactions. In open-ended…
-
Tokenized Key-Gated Adapter Routing: A Secure Access Control Mechanism Against Private Data Leakage in LLMs
arXiv cs.CR (Security)
arXiv:2610.00309v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly deployed in privacy-critical domains (e.g., healthcare, finance, and government), but their propensity to memorize and disclose personally…
-
A Comprehensive Review of One-Pixel Attack: Research Status, Taxonomy, Applications, Regulation Policy and Future Directions
arXiv cs.CR (Security)
arXiv:2610.00125v1 Announce Type: new Abstract: One-Pixel Attacks (OPAs) represent one of the most extreme demonstrations of adversarial fragility in deep learning, where modifying a single pixel can reliably induce high-confidence…
-
A Verifier Can Leak the Answer: Diagnosability Before Optimization in Closed-Loop Agent Debugging
arXiv cs.CR (Security)
arXiv:2610.00126v1 Announce Type: new Abstract: Agent developers increasingly compare prompts, tools, policies, and diagnosis algorithms through simulator-grounded verifiers. A verifier can nevertheless make a solver comparison vacuous: if…
-
Removing the NEEDLE in the Haystack: Backdoor Removal in LLMs via Weight Orthogonalisation
arXiv cs.CR (Security)
arXiv:2610.00348v1 Announce Type: new Abstract: Backdoor attacks can be implanted in Large Language Models (LLMs) during training, causing unwanted behaviour when a trigger appears in the input. Existing backdoor defences for LLMs attempt…
-
Proof-Gated Signing: Solver-Checked Transaction Guards that Hold Under State Drift for Onchain AI Agents
arXiv cs.CR (Security)
arXiv:2610.00354v1 Announce Type: new Abstract: AI agents that control wallets read attacker-reachable content, so they can be steered into proposing harmful transactions. The usual last line of defense is a pre-signing check: a static…
-
ABSENTIA: Detecting Broken Access Control Vulnerabilities in Web Applications
arXiv cs.CR (Security)
arXiv:2610.00977v1 Announce Type: new Abstract: Broken access control, the failure of authorization, is one of the most prevalent web security risks. Unlike injection, a flow of untrusted input into a dangerous operation, authorization is…
-
MOMAT: Mixture of Multiple Atlases for Low-Power Jailbreak Defense of Quantized LLMs
arXiv cs.CR (Security)
arXiv:2610.01058v1 Announce Type: new Abstract: Quantized large language models are increasingly deployed on edge devices for their low latency and energy efficiency. However, model quantization weakens alignment safeguards, leaving qLLMs…
-
ISC Stormcast For Friday, October 2nd, 2026 https://isc.sans.edu/podcastdetail/10120, (Fri, Oct 2nd)
SANS ISC