Cybersecurity
Cybersecurity news — Friday 16 October 2026
116 stories from 10 sources, every one in English. Each links to its publisher.
-
Friday Squid Blogging: Squid Before the Chicxulub Meteor Impact
Schneier on Security
New research shows that squid lived all around the globe even before the Cretaceous-Paleogene mass extinction event. As usual, you can also use this squid post to talk about the security stories in the news that I haven’t covered. Blog…
-
Critical BlockStrap flaw lets hackers run code on Magento servers
BleepingComputer
Researchers warn that hackers could soon exploit a maximum-severity vulnerability (CVE-2026-103849) affecting Magento and Adobe Commerce, potentially allowing them to execute arbitrary PHP code on store servers. [...]
-
Have a Meta business account? You’re consenting to AI scraping now, legal experts say
ZDNet Security
And there’s no opt-out available.
-
Researchers Demonstrate OpenBSD Root Exploit Chaining 37-Year-Old Shared-Memory Bug
The Hacker News
A local privilege escalation exploit published today gains root on OpenBSD 7.9 by chaining a roughly 37-year-old integer overflow with a separate kernel flaw, bypassing OpenBSD's KARL kernel-address randomization and its SMAP and SMEP…
-
Zero-click attacks used targeted ads to deliver Candiru spyware
BleepingComputer
Candiru spyware was delivered to devices at Al Jazeera Media Network via targeted online advertisements in a zero-click attack that required only rendering the malicious ads on the victim's device. [...]
-
Munich dumps proprietary for open source (again) – but no Linux desktop
ZDNet Security
Open source in the public sector doesn’t fail because of technical issues. It fails because it’s not aligned with the political agenda
-
Anthropic gives third-party auditors ’employee-like access’ in AI industry first
ZDNet Security
As AI labs continue to self-regulate, external evaluators play a crucial role in spotting risks.
-
Prolific Chinese espionage group exploits BeyondTrust zero-day
CyberScoop Asia
Suspected Chinese hackers with a proven knack for exploiting zero-days in security products are at it again, targeting BeyondTrust customers in attacks that researchers have traced back to late August. Mandiant researchers said a likely…
-
Want an iPhone Duo before 2027? Don’t preorder from Apple – try this instead
ZDNet Security
Estimates for Duo preorders are now showing a ship date of late January, marking a new wait time record for Apple and its customers.
-
Attackers Tampered with AOMEI Backupper's Update Servers for Minutes to Deliver a Backdoor
The Hacker News
Attackers repeatedly changed the files that AOMEI Backupper fetches from the vendor's own update servers, in windows of five to ten minutes at a time between May and July 2026, so that any copy of the Windows backup program that checked…
-
UAT-11985 Uses Fake Event Invites to Target Taiwan Research Affiliates With Google AitM Phishing
The Hacker News
Cybersecurity researchers have disclosed details of an advanced persistent threat (APT) spear-phishing campaign that has targeted individuals affiliated with Taiwan research organizations to direct victims to an adversary-in-the-middle…
-
Cybersecurity incident at BAE Systems disrupts US bankruptcy court system
CyberScoop Asia
A cybersecurity incident at BAE Systems has halted the production and delivery of bankruptcy notices for every federal bankruptcy court in the United States, disrupting a core administrative function of the judiciary with no announced date…
-
Eiliko robot companion exposes private chats and can be hijacked
Malwarebytes Labs
The Verge reports that Eiliko, a small AI robot companion made by Energize Lab, has serious security flaws that expose private conversations and let an attacker remotely change what the toy says. According to The Verge, researcher Sammy…
-
In Other News: ICS-Sniper Attack, Hacking XWorm Malware Operators, $250k Chrome Exploit
SecurityWeek
Noteworthy stories that might have slipped under the radar: BofA adds 1,000 patching contractors, KGB RAT pairs data theft with jump scares, and Langflow flaw under attack. The post In Other News: ICS-Sniper Attack, Hacking XWorm Malware…
-
OpenAI Research Model Spent 3rd-Party Test ETH Without Permission to Deploy Smart Contract
The Hacker News
OpenAI has disclosed seven new incidents in which its artificial intelligence (AI) models exhibited misaligned behavior, including one case where its research model, after not having funds on an Ethereum test network, spent test…
-
Employees using personal ChatGPT at work? Here's what IT admins need to do
BleepingComputer
Employees using personal ChatGPT accounts at work can expose sensitive company data, even when their organization provides an approved Enterprise workspace. NordLayer explains how IT teams can investigate unauthorized AI use and block…
-
Burger King Japan data breach exposes 11 million customer records
BleepingComputer
Burger King Japan, the Japanese subsidiary of global fast-food giant Burger King, disclosed a major data breach after attackers accessed a server containing 11.18 million customer records. [...]
-
“Spray and pray” job applications could make it harder for you to spot a scam
Malwarebytes Labs
AI can help you send more job applications. But when you lose track of where you’ve applied, a fake recruiter’s message can start to look like the reply you’ve been waiting for. Job hunting can feel like shouting into the void. You send…
-
How many jobs is your state losing to AI? This dashboard tracks the cuts – and the cost
ZDNet Security
CAIS’ Wage Loss Dashboard puts AI-related job cuts into dollars and cents.
-
BAE Systems Security Incident Stalls Notices From US Bankruptcy Courts
SecurityWeek
A cyber incident affecting contractor BAE Systems has knocked out the Bankruptcy Noticing Center, halting notices nationwide for a week. The post BAE Systems Security Incident Stalls Notices From US Bankruptcy Courts appeared first on…
-
AI Decision Models and Their Impact on AI SOC Platforms
The Hacker News
In the space of a month, a new category of AI model went from nonexistent to everywhere. TypeSafe AI released Jev on September 15 and called it a System One model, a nod to Kahneman's fast, intuitive thinking. OpenAI announced its own…
-
Backdooring Open-Weight AI Models
Schneier on Security
This is a problem: Any open model whose weights have been edited can carry a backdoor, whether it’s a task-specific fine-tune, a merged adapter or an abliterated build. We have abliterated models in the title because it’s the most popular…
-
Chinese APT Exploited BeyondTrust Vulnerability as Zero-Day
SecurityWeek
Armory Castle exploited CVE-2026-102764 for initial access, deployed custom tools, and deleted logs and artifacts. The post Chinese APT Exploited BeyondTrust Vulnerability as Zero-Day appeared first on SecurityWeek .
-
OpenAI sued over children safety and data use
Malwarebytes Labs
OpenAI is facing separate legal challenges over how ChatGPT interacts with children. The Guardian reports that eight US states accuse the company of misleading parents about the chatbot’s safety and exposing vulnerable users to dangerous…
-
OpenAI Models Hacked Data Portals, Raided a Crypto Wallet to Finish Tasks
SecurityWeek
Left without funds for a coding task, an OpenAI model spent test crypto from a stranger’s wallet and left it out of its report. The post OpenAI Models Hacked Data Portals, Raided a Crypto Wallet to Finish Tasks appeared first on…
-
F5, Juniper Networks Patch Critical Vulnerabilities
SecurityWeek
Successful exploitation of some of these issues could lead to remote code execution and full device takeover. The post F5, Juniper Networks Patch Critical Vulnerabilities appeared first on SecurityWeek .
-
Microsoft: October Windows update may break Smart App Control
BleepingComputer
Microsoft says some users may have trouble turning on Smart App Control after installing this month's KB5127648 Windows 11 security update. [...]
-
Microsoft Patches Two Dozen RCE Flaws in Office
SecurityWeek
In addition to RCE flaws, Microsoft fixed multiple information disclosure bugs and a security bypass issue in Office. The post Microsoft Patches Two Dozen RCE Flaws in Office appeared first on SecurityWeek .
-
BAE Systems cyberattack disrupts US federal court operations
BleepingComputer
The Administrative Office of the US Courts says a cyberattack on government contractor BAE Systems has taken down the notice system used by federal bankruptcy courts. [...]
-
Proofpoint Inks $6.3 Billion All-Cash Deal to Acquire Varonis
SecurityWeek
The Thoma Bravo-owned Proofpoint will acquire Varonis in an all-cash transaction and Varonis will become a private company. The post Proofpoint Inks $6.3 Billion All-Cash Deal to Acquire Varonis appeared first on SecurityWeek .
-
China-Linked Hackers Exploited BeyondTrust Zero-Day to Take Over Remote Support Servers
The Hacker News
The China-nexus cyber espionage group tracked as Armory Castle (previously UNC5221 or Warp Panda) exploited a newly disclosed security flaw in BeyondTrust Remote Support (formerly Bomgar) as a zero-day in late August 2026. The…
-
CERT-In Warns AI Agents Accessed Sensitive Data at Indian Organizations During Testing
The Hacker News
AI agents being tested online reached the live systems of Indian organizations and read business records, including personal data and confidential material, CERT-In said in an advisory on October 15. The agents went beyond their assigned…
-
BeyondTrust patches zero-day exploited in attacks since August
BleepingComputer
BeyondTrust warned customers on Thursday to patch a critical flaw in its Remote Support (RS) that has been exploited in zero-day attacks since late August. [...]
-
PEERSTITCH Malware Deployed via FortiOS Flaw Bypasses Checks on Tagged VPN Sessions
The Hacker News
A suspected advanced threat actor has been observed exploiting a security flaw in Fortinet FortiOS to deliver a previously unreported memory-resident ARM64 malware known as PEERSTITCH. The attack involves the exploitation of CVE-2025-61624…
-
CascadeLedger: Counterfactual Auditing of Information and Resource Integrity in LLM Cascades
arXiv cs.CR (Security)
arXiv:2610.20668v1 Announce Type: new Abstract: We introduce CascadeLedger, a counterfactual execution-audit framework for budgeted language-model cascades. The protocol enforces call-before-read semantics, binds successful reads to…
-
Privacy Foundations for Multi-Institutional Scientific Artificial Intelligence
arXiv cs.CR (Security)
arXiv:2609.39787v2 Announce Type: replace Abstract: Scientific artificial intelligence (AI), spanning foundation models (FMs) to federated data-analysis pipelines, is becoming shared infrastructure across national laboratories…
-
Performance Characterization of Pulsar's Software-defined Watermark Verifier
arXiv cs.CR (Security)
arXiv:2610.20827v1 Announce Type: new Abstract: Pulsar provides the world's most secure PNT service by incorporating a Combinatorial Watermark into its ranging service. Under conservative signal conditions, Pulsar's Combinatorial…
-
Foundations of Hardware-Software Leakage Contracts
arXiv cs.CR (Security)
arXiv:2610.20754v1 Announce Type: new Abstract: Microarchitectural attacks such as Spectre and Meltdown undermine security by exploiting software-visible side effects of microarchitectural optimizations like caches and speculative…
-
An Empirical Analysis of EOS Blockchain: Architecture, Contract, and Security
arXiv cs.CR (Security)
arXiv:2505.15051v2 Announce Type: replace Abstract: With the rapid development of blockchain technology, various blockchain systems are exhibiting vitality and potential. As a representative of Blockchain 3.0, the EOS blockchain has been…
-
GoLeash: Mitigating Golang Software Supply Chain Attacks with Runtime Policy Enforcement
arXiv cs.CR (Security)
arXiv:2505.11016v2 Announce Type: replace Abstract: Modern software supply chain attacks consist of introducing new, malicious capabilities into trusted third-party software components, in order to propagate to a victim through a package…
-
On the Privacy of LLMs: An Ablation Study
arXiv cs.CR (Security)
arXiv:2605.02255v2 Announce Type: replace Abstract: Large language models (LLMs) face diverse privacy threats, yet existing studies typically evaluate individual attacks in isolation, offering limited guidance on which system factors drive…
-
Combining LLM and Dynamic Analysis for Universal Testing of Static Application Security Testing Tools
arXiv cs.CR (Security)
arXiv:2610.21309v1 Announce Type: new Abstract: Static application security testing (SAST) tools help detect potential software vulnerabilities, but they themselves contain faults undermining reliability and usability. Automated testing…
-
NOMAI: Scalable Federated Learning with Private Weights
arXiv cs.CR (Security)
arXiv:2610.21059v1 Announce Type: new Abstract: The twin goals of privacy-preserving federated learning (FL) -- harnessing distributed data while protecting the parties who contribute it -- demands three properties: model privacy, data…
-
Language Is a Security Surface for Agent Guards
arXiv cs.CR (Security)
arXiv:2610.22244v1 Announce Type: new Abstract: Step-level guards decide, before each tool call, whether an agent's next action is safe given the user's request and the trajectory so far. Users write those requests in many languages, while…
-
INFUSE: A Unified Surface Benchmark for Indirect Prompt Injection in Agentic Systems
arXiv cs.CR (Security)
arXiv:2610.20776v1 Announce Type: new Abstract: LLM-based agents increasingly execute long-horizon, end-to-end workflows involving shell commands, code modification, dependency installation, and interaction with remote services. This…
-
Does Client Diversity Protect the Lightning Network? Evidence from the 2026 Core Lightning Shutdown
arXiv cs.CR (Security)
arXiv:2610.20952v1 Announce Type: new Abstract: Client diversity is often cited as the Lightning Network's defence against a bug in any one implementation, yet LND runs 84% of public nodes and holds 84% of public channel capacity, a share…
-
Optimal Reduced Representatives of Quadratic APN Permutations in Odd Dimension
arXiv cs.CR (Security)
arXiv:2610.21169v1 Announce Type: cross Abstract: The formal derivative of the reduced polynomial representative of a map on a binary finite field depends on the coordinates, although its differential uniformity does not. This matters for…
-
Defining Purpose-Limited Secrets
arXiv cs.CR (Security)
arXiv:2610.10010v2 Announce Type: replace Abstract: A cryptographic secret is issued for a purpose but grants a capability, and the capability is usually larger. Deployed systems state the purpose in policy and enforce only the capability…
-
Clean Data Can Still Carry Backdoors: Support-Persistent Backdoors for Model Reuse
arXiv cs.CR (Security)
arXiv:2610.21208v1 Announce Type: new Abstract: Traditional backdoor attacks inject artificial trigger patterns into training data, causing models to misclassify triggered inputs while maintaining normal behavior on clean samples. Since…
-
Benchmarking Post-Quantum Key Encapsulation in S3-Compatible Object Storage under Client Concurrency
arXiv cs.CR (Security)
arXiv:2610.21457v1 Announce Type: new Abstract: Data encrypted today with classical public-key algorithms could be decrypted once large quantum computers exist. Cloud object stores keep data for years, so the client-side encryption that…
-
BiShield-TEE: On the (In-)Security of Unilateral Weight Obfuscation in On-Device TEE-Shielded LLM Partition
arXiv cs.CR (Security)
arXiv:2610.21565v1 Announce Type: new Abstract: On-device deployment of large language models (LLMs) is increasingly adopted due to latency and privacy requirements, but exposes proprietary models to reverse engineering and intellectual…
-
SecureClaw: Clawing Back Control of LLM Agents
arXiv cs.CR (Security)
arXiv:2606.09549v3 Announce Type: replace Abstract: Tool-using large language model (LLM) agents face two distinct security failures: unauthorized external actions and exposure of sensitive plaintext inside the runtime before any final…
-
On the Complexity of Symmetric Boolean Functions in Multiparty Private Simultaneous Messages
arXiv cs.CR (Security)
arXiv:2610.21216v1 Announce Type: new Abstract: Multiparty Private Simultaneous Messages (PSM), also known as Decomposable Randomized Encoding (DRE) or information-theoretic garbling, is a model for non-interactive secure computation. Each…
-
NEST: Stale-Robust Tokenized-Stock Measurement for a Crypto-Treasury LETF Stack
arXiv cs.CR (Security)
arXiv:2610.21491v1 Announce Type: cross Abstract: Tokenized equities trade while listed markets are closed, so their marks could improve off-hours measurement along layered exposures that link Bitcoin, crypto-treasury stocks, and leveraged…
-
Declared but Unaccountable: Measuring the Seller-Identity Gap in Advertising Supply-Chain Transparency
arXiv cs.CR (Security)
arXiv:2610.14661v2 Announce Type: replace-cross Abstract: Public advertising transparency relies on a minimum accountability condition: an account authorized by a publisher in ads$.$txt or app-ads$.$txt should appear in the applicable…
-
Evaluating and Improving Static Application Security Testing Tools for Rust
arXiv cs.CR (Security)
arXiv:2610.21373v1 Announce Type: new Abstract: Rust has become a popular systems programming language in recent years because of its strong guarantees on memory and concurrency safety. Many types of system software have been…
-
Pricing the Unprovable: Ambiguity, Turnstile Runs, and the Verifiability Discount in Crypto Protocols
arXiv cs.CR (Security)
arXiv:2610.21540v1 Announce Type: cross Abstract: Privacy-preserving protocols can repair a vulnerability without establishing whether it was exploited beforehand, even when no theft is reported and aggregate supply is protected. We call…
-
Empirical Evaluation of Machine Learning for Cloud Cryptojacking Detection: Effects of Preprocessing and Evaluation Protocols
arXiv cs.CR (Security)
arXiv:2610.20658v1 Announce Type: new Abstract: Artificial intelligence (AI) has been widely applied to cloud cryptojacking detection. Classical machine learning models remain among the most widely used approaches for cloud cryptojacking…
-
On the Robustness of Gossip Protocols in Modern Blockchains
arXiv cs.CR (Security)
arXiv:2610.21526v1 Announce Type: new Abstract: Modern platforms such as Ethereum and Polkadot nowadays execute Byzantine fault-tolerant consensus protocols over large validator committees, often comprising hundreds of nodes. At this…
-
ADCA: Attention-Driven Multi-Party Collusion Attack in Federated Self-Supervised Learning
arXiv cs.CR (Security)
arXiv:2602.05612v2 Announce Type: replace Abstract: Federated Self-Supervised Learning (FSSL) integrates the privacy advantages of distributed training with the capability of self-supervised learning to leverage unlabeled data, showing…
-
Private Retrieval Augmented Generation via Random Projection
arXiv cs.CR (Security)
arXiv:2610.22297v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) enhances the capabilities of large language models (LLMs) by querying external structured knowledge. However, it can also introduce privacy risks by…
-
From Detection to Decision: Calibration and Deferral for Prompt-Injection Detection
arXiv cs.CR (Security)
arXiv:2610.20907v1 Announce Type: new Abstract: Prompt-injection defenses are commonly evaluated as binary detectors, although deployed LLM applications must make operational decisions about whether to allow, block, or escalate potentially…
-
Post-Quantum Secure Aggregation via Code-Based Homomorphic Encryption
arXiv cs.CR (Security)
arXiv:2601.13031v2 Announce Type: replace Abstract: Secure aggregation lets a server learn the sum of inputs from multiple users without learning individual contributions. Recent single-server protocols achieve this with key- and…
-
GhostLock: Learning-Assisted Blind GPS Signal Nullification Without Victim-Side Information
arXiv cs.CR (Security)
arXiv:2610.20831v1 Announce Type: new Abstract: The Global Positioning System (GPS) provides critical positioning, navigation, and timing (PNT) services for numerous safety- and mission-critical applications, making it an attractive target…
-
Subgraph Counting with Local Node Differential Privacy
arXiv cs.CR (Security)
arXiv:2610.22278v1 Announce Type: cross Abstract: We study the problem of approximately counting subgraphs under local node differential privacy (LNDP), a recently introduced model that enforces strong privacy guarantees in analyses of…
-
Hybrid DeepXAI Fusion Network for Interpretable Big Data Analytics and Secure Cloud Decision-Making Systems
arXiv cs.CR (Security)
arXiv:2610.20661v1 Announce Type: new Abstract: Big data analytics supports modern intelligent systems, but model complexity and limited transparency can undermine interpretability, trust, and security in decision-making. This paper…
-
Screwtape: Adversarial Transcript Editing for Monitor Red-Teaming
arXiv cs.CR (Security)
arXiv:2610.20941v1 Announce Type: new Abstract: We present Screwtape, a pipeline that converts real agent transcripts into attacks that evade action-only monitors. An LLM attacker edits a transcript so that the agent causes a specified…
-
Blind Watermarking of Software Repositories via Module Placement
arXiv cs.CR (Security)
arXiv:2610.21031v1 Announce Type: new Abstract: Source-code watermarking supports provenance tracking by embedding a recoverable identifier in a software artifact. Many existing approaches encode this identifier in tokens or program…
-
CARE-SD: Explainable Reasoning Architecture for Local SLM Self-Disclosure Analysis in Textual Data
arXiv cs.CR (Security)
arXiv:2610.21151v1 Announce Type: new Abstract: Cybersecurity interviews uncover critical security experiences, behaviors, and privacy concerns often not fully captured by conventional technical assessments. Consequently, identifying…
-
Differentially Private Distribution Release of Gaussian Mixture Models via KL-Divergence Minimization
arXiv cs.CR (Security)
arXiv:2506.03467v4 Announce Type: replace-cross Abstract: Gaussian Mixture Models (GMMs) are widely used statistical models for representing multi-modal data distributions, with numerous applications in data mining, pattern recognition…
-
Forensic Analysis of the HackRF One with PortaPack Mayhem Firmware
arXiv cs.CR (Security)
arXiv:2610.20647v1 Announce Type: new Abstract: Software-defined radios (SDRs) have become an inexpensive, portable platform for replay attacks, location spoofing and jamming. The HackRF One can operate stand-alone when fitted with a…
-
Extending Layered Attestation into ARM Application Run-Times
arXiv cs.CR (Security)
arXiv:2610.17614v2 Announce Type: replace Abstract: Modern mobile and edge platforms provide hardware-backed attestation of application and platform integrity at load-time. However, they offer little support once application execution…
-
Amora: Communication-Efficient Private Approximate Nearest Neighbors Search
arXiv cs.CR (Security)
arXiv:2610.21174v1 Announce Type: cross Abstract: Approximate nearest neighbors search (ANNS) over cloud-hosted vector databases raises a critical privacy concern: the query vector, which often encodes sensitive user intent, is exposed to…
-
Unaware, Unfunded and Uneducated: A Systematic Review of SME Cybersecurity
arXiv cs.CR (Security)
arXiv:2309.17186v3 Announce Type: replace Abstract: Small and medium enterprises (SMEs) are pivotal in the global economy, accounting for over 90% of businesses and 60% of employment worldwide. Despite their significance, SMEs are often…
-
Password Strength Detection via Machine Learning: Analysis, Modeling, and Evaluation
arXiv cs.CR (Security)
arXiv:2505.16439v3 Announce Type: replace Abstract: As network security issues continue gaining prominence, password security has become crucial in safeguarding personal information and network systems. This study first introduces various…
-
Efficient Agnostic Learning of Univariate Densities Under Local Differential Privacy
arXiv cs.CR (Security)
arXiv:2610.21033v1 Announce Type: cross Abstract: We propose a general framework for non-interactive locally differentially private (LDP) learning of structured univariate distributions that can be approximated by piece-wise polynomials…
-
MIRAGE: Stealthy Visual Prompt Injection for Vulnerability Detection in Web Agents
arXiv cs.CR (Security)
arXiv:2606.20717v2 Announce Type: replace-cross Abstract: Multimodal Large Language Model (MLLM)-based web agents provide practical, high-precision solutions for visual browser automation; however, they inherently expand the attack…
-
FedSumUp: User-Auditable Privacy-Aware Representation Sharing for Resource-Constrained Federated Intelligence
arXiv cs.CR (Security)
arXiv:2610.21194v1 Announce Type: new Abstract: Existing federated data condensation methods face privacy risks, as uploaded synthetic data may preserve sensitive information and remain vulnerable to reconstruction attacks. Moreover, their…
-
Words Speak Louder Than Code: Investigating Cognitive Heuristics in LLM-Based Code Vulnerability Detection
arXiv cs.CR (Security)
arXiv:2606.30587v2 Announce Type: replace Abstract: Researchers and practitioners increasingly apply Large Language Models (LLMs) for automated vulnerability detection. Recent work has shown that LLMs are susceptible to the same cognitive…
-
Language-Based Agent Control
arXiv cs.CR (Security)
arXiv:2605.12863v2 Announce Type: replace-cross Abstract: This paper introduces language-based agent control (LBAC), a new programming model for agentic applications that brings techniques from programming languages and language-based…
-
ExecWeave: Cross-Layer Execution Provenance for Observable Agentic Software
arXiv cs.CR (Security)
arXiv:2610.21146v1 Announce Type: new Abstract: Agentic software invokes tools, launches processes, accesses files, and communicates with local or remote services. Application traces describe logical operations, whereas operating-system…
-
Thinking in Graphs: Neo4j-Backed Exploit-Chain Reasoning for LLM Penetration-Testing Agents
arXiv cs.CR (Security)
arXiv:2610.20902v1 Announce Type: new Abstract: LLM coding agents such as Claude Code and OpenAI Codex can drive penetration-testing tools, but they reason about an engagement as a flat list of findings in a context window. Attackers think…
-
PAGE404: Breaking GPU Isolation Through Hardware-assisted Page Migration
arXiv cs.CR (Security)
arXiv:2610.20962v1 Announce Type: new Abstract: NVIDIA's Grace Hopper (GH200) superchip represents a new generation of heterogeneous computing, tightly integrating a 72-core Arm-based CPU with a Hopper GPU on a single module via a…
-
Can Autonomous Vehicles Keep Their Cool? A Survey of Thermal Side-Channel and Perception Attacks
arXiv cs.CR (Security)
arXiv:2610.21034v1 Announce Type: new Abstract: Autonomous vehicles (AVs) couple safety-critical perception to thermally constrained compute. Interfaces built for night perception and package thermal reliability were never engineered as…
-
Approve What, Exactly? Effect Binding and Trusted Rendering for Agent Step-Up Authorization
arXiv cs.CR (Security)
arXiv:2610.21063v1 Announce Type: new Abstract: AI agents now pause for approval before they send money, delete files or email outsiders, and that approval is usually treated as a second factor. We argue it is one only when compromising…
-
Reaction Attack on CKKS Deployments with Asymptotically Optimal Query Complexity
arXiv cs.CR (Security)
arXiv:2610.21127v1 Announce Type: new Abstract: We study a protocol-level reaction leakage in CKKS deployments where public coefficient-to-slot (C2S) evaluation keys are combined with a client's 1-bit validity response. Under this…
-
Stronger Judges Do Not See More: Deployment-Integrity Blindness in LLM Evaluation
arXiv cs.CR (Security)
arXiv:2610.21135v1 Announce Type: new Abstract: LLM-agent evaluation reports behavioral scores but leaves the identity of the deployed runtime stack unreported. A tampered deployment that preserves evaluated outputs passes every…
-
Matoi: Selective Revalidation of Consequence Evidence for Agent Authorization
arXiv cs.CR (Security)
arXiv:2610.21394v1 Announce Type: new Abstract: Matoi separates reusable consequence evidence from current execution authority and durable occurrence history. Its finite query-aware contract covers branches, ordered collections and…
-
SMT-Boosted Security Types and Correctness for Low-Level MPC
arXiv cs.CR (Security)
arXiv:2610.22047v1 Announce Type: new Abstract: Secure Multi-Party Computation (MPC) is an important enabling technology for data privacy in modern distributed applications. We develop a new Hoare logic and type theory to automatically…
-
Explicit reduced APN permutations: ramification and optimal Galois-ring lifts
arXiv cs.CR (Security)
arXiv:2610.21178v1 Announce Type: cross Abstract: Let $m\ge5$ be odd and $q=2^m$. For every odd $r$ with $3\le r
-
Smoothness of Shortest Isogenies to Frobenius Conjugates
arXiv cs.CR (Security)
arXiv:2610.21488v1 Announce Type: cross Abstract: Let $E$ be chosen uniformly from the geometric isomorphism classes of supersingular elliptic curves over $\overline{\mathbb{F}}_p$ and $\delta_E$ be the minimum degree of isogenies from $E$…
-
Federated Art Chase: Privacy Leakage and Data Poisoning in Federated Fine-Tuning
arXiv cs.CR (Security)
arXiv:2610.21547v1 Announce Type: cross Abstract: Federated learning (FL) is increasingly used to adapt a pretrained, centrally fine-tuned model to data that remain on user devices, beyond training a model from scratch. In this report…
-
Shortcut Sensitivity in Agent-Safety Evaluations
arXiv cs.CR (Security)
arXiv:2610.22243v1 Announce Type: cross Abstract: AI agents take actions through tools. Some actions are sensitive, so guards that supervise them must themselves be evaluated reliably. Yet the data used to evaluate a guard can include cues…
-
SoK: Trading Agents or Market Crashers? Dissecting Robustness and Security Failures in Academic Financial LLM Trading Schemes
arXiv cs.CR (Security)
arXiv:2609.19705v2 Announce Type: replace Abstract: Large language model (LLM) agents already trade with real capital, yet agent-security research has mostly treated them as generic agents. Trading changes the threat: a compromised agent's…
-
Autonomous OSS Threat Detection via Taxonomy-Aligned LLMs
arXiv cs.CR (Security)
arXiv:2610.01263v2 Announce Type: replace Abstract: Open-source software (OSS) ecosystems face escalating supply chain threats, including typosquatting, dependency confusion, Trojan Source obfuscation, malicious build injection, and CI/CD…
-
How Feasible are Passive Network Attacks on 5G Networks and Beyond? A Survey
arXiv cs.CR (Security)
arXiv:2512.20622v3 Announce Type: replace-cross Abstract: Privacy concerns around 5G, the latest generation of mobile networks, are growing, with fears that its deployment may increase exposure to privacy risks. This perception is largely…
-
Limitations on Autoregressive Token Generation via Sequential Sampling
arXiv cs.CR (Security)
arXiv:2610.14615v2 Announce Type: replace-cross Abstract: We study autoregressive token generation through the notion of sequential samplers: A circuit that samples the next symbol only given the previous ones and possibly a small amount…
-
Poisoning the Neighborhood: A Topology-Aware Topic-Level Attack on Large-Scale HNSW-based RAG Systems
arXiv cs.CR (Security)
arXiv:2610.21464v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) mitigates hallucinations and knowledge staleness in large language models using external knowledge. However, ingestion from public and third-party sources…
-
Encoded but Exposed: Language Embeddings can Reveal More Than They Should in Recommender Systems
arXiv cs.CR (Security)
arXiv:2610.21713v1 Announce Type: new Abstract: Language embeddings are increasingly used in recommenders to represent item content, but their privacy risks remain poorly understood. We study whether item embeddings can reveal item…
-
My Credentials, Please: Detecting Filesystem Capability Laundering in MCP Servers at Scale
arXiv cs.CR (Security)
arXiv:2610.22301v1 Announce Type: new Abstract: The Model Context Protocol (MCP) standardizes how AI agents invoke external tools. MCP servers expose tools through a declared contract consisting of a name, description, and argument schema…
-
Counterfactual Neighborhood Risk Fields: Evidence-Grounded Selective Safety for Vision-Language Models
arXiv cs.CR (Security)
arXiv:2610.20679v1 Announce Type: cross Abstract: Representation-based safety defenses for vision-language models (VLMs) often learn to separate harmful from benign inputs, but separation alone does not establish what a risk score means or…
-
QuasiLeak-Bench: Evaluating LLM-Based Privacy Detectors on Policy-Conditioned Quasi-Evidence Group Recovery
arXiv cs.CR (Security)
arXiv:2610.21333v1 Announce Type: cross Abstract: Natural language messages often contain quasi-identifiers that are weak in isolation but can jointly reveal an individual's private attributes. We call such a jointly revealing set of cues…
-
Unmasking the Secret Code: Using Canaries to Show How Masked Context Controls Memorization in SFT
arXiv cs.CR (Security)
arXiv:2610.20656v1 Announce Type: new Abstract: Canaries, or synthetic training records containing known secrets, are widely used to assess whether language models reveal information from their training data. But how much do these…
-
Privacy-Preserving Text Inference for Health-Misinformation Classification and Semantic Retrieval
arXiv cs.CR (Security)
arXiv:2610.20796v1 Announce Type: new Abstract: Health-misinformation analysis presents a dual challenge: misleading claims can affect public well-being, while systems designed to detect them may expose sensitive user-submitted text. This…
-
Trust a Few: The Weakest Assumptions a Protocol Needs
arXiv cs.CR (Security)
arXiv:2610.09730v2 Announce Type: replace Abstract: Protocol verifiers check whether a protocol meets a security goal under stated trust assumptions, such as that a key is never leaked, a value is fresh, or a channel is authentic. They do…
-
Trigger Value Selection in Clean-Label Malware Backdoors: An Effectiveness-Detectability Analysis
arXiv cs.CR (Security)
arXiv:2610.21269v1 Announce Type: new Abstract: Clean-label backdoor poisoning is a threat to machine learning-based malware classifiers because poisoned samples can be inserted into training data without changing their labels. Previous…
-
What Survives Self-Evolution: Provenance Marks in Copied Agent Harnesses
arXiv cs.CR (Security)
arXiv:2610.21458v1 Announce Type: new Abstract: Agent harnesses, the prompt and tool code around a base model, leak. Claude Code's 512,000 lines were mirrored within hours of a packaging error. A plagiarist can pass a leaked harness…
-
Practical wavelength selection for quantum-augmented networks over shared optical fiber
arXiv cs.CR (Security)
arXiv:2610.20987v1 Announce Type: cross Abstract: In a quantum augmented network, sharing optical fiber with classical traffic requires quantum wavelength selection to balance logical fidelity, block acceptance, and classical data…
-
From One Query to Many: A Critical Survey of Privacy Accounting in RAG
arXiv cs.CR (Security)
arXiv:2610.21245v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) enables large language models to leverage external knowledge but introduces privacy risks when the retrieved corpus contains sensitive information…
-
Identifier-Free Code Embedding Models for Scalable Search
arXiv cs.CR (Security)
arXiv:2605.05251v2 Announce Type: replace Abstract: Function association is a useful process for binary reverse engineers. Search tools exist to perform association at scale, but they do not utilize the full range of capabilities that…
-
SHIP: Structured and Hierarchical Insight with Privacy
arXiv cs.CR (Security)
arXiv:2610.21048v1 Announce Type: cross Abstract: Extracting insights from textual data while protecting user privacy is increasingly important. Existing methods produce private insights in diverse forms for different analytical purposes…
-
When Rotation Bounds Nothing: Streaming Image Classifiers to Field Nodes Under Physical Capture
arXiv cs.CR (Security)
arXiv:2610.21276v1 Announce Type: new Abstract: Field nodes run image classifiers locally, exposing a first-class asset, the trained model, to theft. Trusted-execution enclaves, collaborative-inference splits, and encrypted storage assume…
-
MAppPrint: Robust and Real-time Mobile App Fingerprinting via Traffic Behavior Correlation Graphs
arXiv cs.CR (Security)
arXiv:2610.22148v1 Announce Type: new Abstract: Mobile apps have become ubiquitous in daily life. Despite communication encryption, app fingerprinting (AF) attacks can still enable adversaries to infer the apps in use by analyzing patterns…
-
ISC Stormcast For Friday, October 16th, 2026 https://isc.sans.edu/podcastdetail/10140, (Fri, Oct 16th)
SANS ISC
-
Small Agents, Smaller Blast Radius: Controlling What AI Can Do, (Fri, Oct 16th)
SANS ISC
I believe the cybersecurity industry may be approaching AI security the wrong way.
-
TAG-217 Targets South Korean Financial Services with ARTEX
Recorded Future
The threat actor linked to recent AI-assisted attacks on South Korean financial institutions was testing infrastructure and assembling attack tools as early as August 2026, weeks before the banking campaign became public, according to new…