AX·NEWS
Sign in

Cybersecurity

Cybersecurity news — Wednesday 28 October 2026

112 stories from 10 sources, every one in English. Each links to its publisher.

Open in the live feedRSS feedOn a TV

  1. Leak site publishes data on more than 41,000 DHS employees obtained by hacktivists
    USCyberScoop

    A leak website was published Wednesday, showcasing internal data on more than 41,000 Department of Homeland Security employees and contractors after it was obtained by hackers who claimed to have breached an identity service used by the…

  2. Police seize Virtualine bulletproof service tied to ransowmare, cybercrime
    USBleepingComputer

    U.S. law enforcement has seized the primary domain of Virtualine, a bulletproof hosting service advertised on cybercrime forums for facilitating illegal activities. [...]

  3. The 4 biggest Amazon smart home announcements from today
    USZDNet Security

    From a new smart alarm clock to an Alexa-powered 3D map of your home, here’s what’s new from Amazon, including prices and release dates.

  4. Lawsuit against Ex-Overstock CEO tied to 2020 election film falls apart over question of ‘actual malice’
    USCyberScoop

    A Florida judge ruled this week that ex-Overstock CEO Patrick Byrne is not liable in a lawsuit brought by a Dominion Voting Systems executive over a film that accuses the former electronic voting hardware and software company executive of…

  5. Apple’s new touchscreen MacBook Pros may be late, but they have one big advantage
    USZDNet Security

    Apple finally revealed its new MacBook M5 Pro and Max with a touchscreen OLED display, bringing the best features from the iPhone to its laptops.

  6. Your public Discord messages may soon appear in Google search results – how to opt out
    USZDNet Security

    Discord announced a new feature that can make your server messages discoverable on Google. Here’s how to opt out.

  7. Amazon’s new Alexa Photo Frame could have what I’ve been missing – here’s why
    USZDNet Security

    You can ask Amazon’s voice assistant to show you your favorite photos.

  8. Discord to let search engines index public server messages
    USBleepingComputer

    Discord will allow search engines to index discussions in Discoverable Servers and make them accessible on the web to users outside the platform without an account or invitation. [...]

  9. Volt Typhoon Used Covert Silent Train Network in Four Zero-Day Exploitation Campaigns
    INThe Hacker News

    The China-nexus cyber espionage group known as Volt Typhoon has been attributed to a now-disrupted covert access network dubbed Silent Train that has been put to use for zero-day exploitation and strategic access. "First identified in late…

  10. These Claude Enterprise features let you track AI ROI – what’s new
    USZDNet Security

    Is your pricey Claude Enterprise account worth it? Anthropic is trying to surface insights with three tools.

  11. Google to Allow One-Tap Android App Installs From Approved Third-Party Stores Outside U.S.
    INThe Hacker News

    Third-party Android app stores can now apply to Google for permission to install apps with one tap, without Android's usual confirmation screens, and to update them in the background. To qualify, a store must pass identity checks and send…

  12. Amazon’s Echo Show 16 makes Apple’s 7-inch HomeView look puny – is it worth $100+ extra?
    USZDNet Security

    Amazon just dropped new smart home gear in response to Apple. Here’s how the Echo Show 16 stacks up.

  13. Sony ISP subsidiary data breach exposes 2.5 million customer records
    USBleepingComputer

    Sony Network Communications has confirmed a data breach that compromised approximately 2.5 million records containing current and former customers' personal information. [...]

  14. Researcher Reveals How Logic Bugs Alone Cracked Microsoft Edge in $175,000 Pwn2Own Win
    INThe Hacker News

    A chain of logic flaws, with no memory corruption, can turn a malicious web page into code running on a Windows computer outside Microsoft Edge's sandbox, a security researcher has shown. Orange Tsai of DEVCORE published the full method on…

  15. Opera will run AI locally in your browser now – why that’s a gamechanger and how to try it
    USZDNet Security

    Thanks to the newly proposed WebAI web standard, the company will start shipping local AI models with its flagship browser.

  16. Yet Another ClickFix Campaign, (Wed, Oct 28th)
    USSANS ISC

    Introduction

  17. PHP2Shell Attack Roots 25 Servers, Breaches U.S. Government-Affiliated Network
    INThe Hacker News

    Cybersecurity researchers have disclosed details of what appears to be an artificial intelligence (AI)-assisted attack orchestrated by a suspected China-nexus threat actor that has leveraged the PHP2Shell vulnerabilities to breach exposed…

  18. Volt Typhoon hides zero-day attacks using 'Silent Train' framework
    USBleepingComputer

    Chinese state-sponsored hacking group Volt Typhoon has been using a framework that relies on rotating chains of connected devices compromised on demand, including for zero-day attacks targeting high-value systems. [...]

  19. Researchers uncover Volt Typhoon’s covert, ephemeral network
    USCyberScoop

    Volt Typhoon, a China-government backed hacking group that has infiltrated America’s critical infrastructure networks for years, has consistently built and dismantled ephemeral malicious infrastructure to burrow into hundreds of sensitive…

  20. Prosecutors charge Madagascar scammers in NFT-stealing scheme based on fake online dating profiles
    USCyberScoop

    Prosecutors announced charges against six alleged scammers based out of Africa who targeted Americans with fake non-fungible token (NFT) investment scams by posing as women on dating websites, netting hundreds of victims. According to an…

  21. Amazon’s new Echo Studio gives Alexa+ an AI camera to see inside your home
    USZDNet Security

    The Echo Studio with Home Camera essentially equips Alexa+ with eyes to enable ‘visual intelligence’ features – but there’s a kill switch.

  22. Samsung’s new clip-style Galaxy Buds On have a major advantage over Sony and Bose
    USZDNet Security

    Advanced audio features are great, but ecosystem integration is better. Here’s how Samsung’s newest clip-on Buds compare to its rivals.

  23. Ukrainian Hive and Nefilim Ransomware Affiliate Gets 70 Months in US Prison
    USSecurityWeek

    Artem Stryzhak was sentenced to 70 months in prison after pleading guilty to using Nefilim ransomware in attacks. The post Ukrainian Hive and Nefilim Ransomware Affiliate Gets 70 Months in US Prison appeared first on SecurityWeek .

  24. Microsoft urges Entra ID admins to migrate users to passkeys
    USBleepingComputer

    Microsoft again asked Entra ID administrators to migrate users to phishing-resistant methods to avoid sign-in disruptions before it retires SMS and voice authentication in February 2027. [...]

  25. Amazon’s new Echo Dot can turn into a Wi-Fi extender – how it works
    USZDNet Security

    Amazon’s new Echo Dot includes a built-in Eero Wi-Fi extender, along with other smart home features.

  26. InfraTrust report warns of exploited flaws in critical network infrastructure
    USBleepingComputer

    Attackers are actively exploiting vulnerabilities across critical network infrastructure, and PHP flaws known as "PHP2Shell" put the web management interfaces of firewalls, VPNs, application delivery controllers, and other appliances at…

  27. Upwind Security Raises $400 Million at $3.8 Billion Valuation
    USSecurityWeek

    The massive funding round comes just days after reports emerged that Upwind had attracted acquisition interest from several companies, including Datadog The post Upwind Security Raises $400 Million at $3.8 Billion Valuation appeared first…

  28. Even after PHP2Shell, blast radius matters more than a CVSS score
    USBleepingComputer

    PHP2Shell shows how quickly attackers can exploit critical vulnerabilities, often before organizations have time to patch. Token Security explains why limiting the access of service accounts, API keys, and AI agents is just as important as…

  29. Chrome and Firefox Updates Patch Over 200 Vulnerabilities
    USSecurityWeek

    The Chrome 156 update fixes 52 bugs, while Firefox 159 was rolled out with 167 security patches. The post Chrome and Firefox Updates Patch Over 200 Vulnerabilities appeared first on SecurityWeek .

  30. Banks are blocking AI agents from customer accounts
    USMalwarebytes Labs

    This week, Axios reported that major US banks were detecting third-party AI agents operating inside customer accounts. In many cases, customers have given the agents their usernames and passwords so they could act on their behalf. This…

  31. TLS Flaws Found in GoAhead Web Server Used in Hundreds of Millions of Devices
    USSecurityWeek

    FoxIO disclosed three GoAhead TLS bugs that can let revoked certificates through or expose some encrypted sessions to eavesdroppers. The post TLS Flaws Found in GoAhead Web Server Used in Hundreds of Millions of Devices appeared first on…

  32. Hackers Target Critical Oracle E-Business Suite RCE Flaw Days After Security Patch
    INThe Hacker News

    Threat actors are beginning to probe for a newly patched critical vulnerability in Oracle E-Business Suite (EBS), according to threat intelligence firm Defused Cyber. The vulnerability in question is CVE-2026-101767 (CVSS score: 9.8) , a…

  33. Hackers now scanning systems for recently-patched Oracle flaw
    USBleepingComputer

    Attackers have begun probing the Internet for systems unpatched against a recently disclosed critical Oracle E-Business Suite (EBS) vulnerability. [...]

  34. China-Linked Hackers Exploit PHP2Shell to Root Firewalls
    USSecurityWeek

    The attackers chained PHP2Shell and Fragnesia exploits and hacked two servers of a US government-affiliated organization. The post China-Linked Hackers Exploit PHP2Shell to Root Firewalls appeared first on SecurityWeek .

  35. Critical 16-Year-Old UEFI Flaw Can Run Code Before Secure Boot on Network-Booting Servers
    INThe Hacker News

    A critical 16-year-old flaw in EDK II, the open-source code that most hardware makers use to build UEFI firmware, can allow an attacker to run code on a machine before its operating system loads, but only while the machine is booting over…

  36. Wazuh and AI for Enhanced SOC Workflows
    INThe Hacker News

    Artificial Intelligence (AI) has become one of this decade's defining technologies. From healthcare and finance to manufacturing and education, organizations increasingly rely on AI to automate repetitive tasks, uncover patterns hidden…

  37. Adobe Fixes 8 Critical Campaign Classic Flaws Allowing Code Execution and Security Bypass
    INThe Hacker News

    Adobe has released patches to address multiple critical security flaws in Campaign Classic that, if successfully exploited, could result in arbitrary code execution, security feature bypass, and privilege escalation. The list of eight…

  38. Chrome and Firefox updates fix serious security flaws. Install them now
    USMalwarebytes Labs

    On October 27, Google issued a new Stable channel update for Chrome. Most users are on Chrome’s Stable channel. The desktop update brings Chrome to versions 156.0.8078.73/.74 for Windows and Mac and 156.0.8078.73 for Linux . The update…

  39. Studying Fraudulent North Korean IT Workers
    USSchneier on Security

    Researchers set up a fake start-up company to study fraudulent North Korean IT workers.

  40. Long-Hidden Flaws Patched in EPICS Research Facility Control Software
    USSecurityWeek

    Four EPICS Base flaws found by an autonomous system could let attackers crash processes at particle accelerators and fusion labs. The post Long-Hidden Flaws Patched in EPICS Research Facility Control Software appeared first on SecurityWeek…

  41. CISA warns of hackers now exploiting critical Node2Shell flaw
    USBleepingComputer

    On Tuesday, the US Cybersecurity and Infrastructure Security Agency (CISA) confirmed that hackers have begun exploiting a critical Brotli vulnerability known as Node2Shell, Broccoli2Shell, and HTTP2Shell. [...]

  42. American Organizations Under Attack: Skyw4lker Device Code Phishing Campaign Exposed
    INThe Hacker News

    Modern phishing campaigns can compromise American businesses without relying on traditional credential theft. Attackers can abuse legitimate cloud services, authentication mechanisms, and trusted business workflows to gain access to…

  43. Nobody Trusts the Security Questionnaire. We Rely on It Anyway.
    USSecurityWeek

    The process that gates nearly every B2B deal is a self-assessment the sender distrusts and the receiver accepts. There is finally a better basis for trust between companies. The post Nobody Trusts the Security Questionnaire. We Rely on It…

  44. Secrets Management Firm Infisical Valued at $600 Million After $40 Million Raise
    USSecurityWeek

    The company will use the Series B funds to expand its platform for securing AI agents’ access to sensitive systems. The post Secrets Management Firm Infisical Valued at $600 Million After $40 Million Raise appeared first on SecurityWeek .

  45. Nefilim and Hive ransomware affiliate gets 70 months in prison
    USBleepingComputer

    A Ukrainian national was sentenced to 70 months in prison for his role in ransomware attacks targeting hospitals, government entities, and high-revenue businesses worldwide between 2018 and 2021. [...]

  46. Hackers Start Probing Fresh Oracle EBS Vulnerability
    USSecurityWeek

    CVE-2026-101767 is an unauthenticated SQL injection that can be exploited for remote code execution. The post Hackers Start Probing Fresh Oracle EBS Vulnerability appeared first on SecurityWeek .

  47. CISA Warns of Node2Shell Exploitation as Honeypots Log Attack Attempts
    USSecurityWeek

    CISA added the Brotli flaw to its KEV catalog, and two honeypot operators logged attempts, including one that used a broken exploit. The post CISA Warns of Node2Shell Exploitation as Honeypots Log Attack Attempts appeared first on…

  48. Anthropic's Unreleased AI Model Chains Flaws Into Working Chrome, Safari, Windows and Linux Exploits
    INThe Hacker News

    An unreleased Anthropic AI model found security flaws in Chrome, Safari, Android, Windows, and Linux and chained them into working proof-of-concept exploits, the company said in its latest report. The exploits were built in the lab after…

  49. FBI Seizes Virtualine Domain Over Alleged Criminal Hosting and Bitcoin Money Laundering
    INThe Hacker News

    The U.S. Department of Justice (DoJ) on Tuesday announced the seizure of a domain associated with a bulletproof hosting provider (BPH) named Virtualine. The move was in response to the service being used to commit and facilitate prolific…

  50. Japan's GMO Offers $1 Million Bounty to Catch Hackers Behind FX Trading Cloud Breach
    INThe Hacker News

    Japan's GMO Internet Group is offering up to $1 million in total for tips that help catch the attackers who breached its Onamae.com Desktop Cloud service, after refusing to pay an extortion demand. The reward program opened on October 28…

  51. Tailscale Patches SSH Flaws That Could Let Hacked Devices Gain Root Access on Other Servers
    INThe Hacker News

    Tailscale has released version 1.104.4 to fix flaws in its built-in SSH server that could let an attacker reach services and files on other servers that only the root user should be able to touch. The flaws are in how Tailscale SSH…

  52. ShinyHunters-Exploited Oracle PeopleSoft Zero-Day Turns Out to Be Three Chained Flaws
    INThe Hacker News

    The critical Oracle PeopleSoft bug that ShinyHunters has exploited since May is not one flaw but three chained together, the researchers who reported it said this week. Oracle patched all three under a single identifier, CVE-2026-35273, in…

  53. Adobe Commerce PolyShell Attacks Breach Hundreds of Stores as Brotli Exploit Traffic Emerges
    INThe Hacker News

    The U.S. Cybersecurity and Infrastructure Security Agency (CISA) on Tuesday added two vulnerabilities impacting Adobe Commerce and Magento, and Google's Brotli decoder library to its Known Exploited Vulnerabilities (KEV) catalog, citing…

  54. CrowdStrike 2026 Legal Threat Landscape Report: eCrime Leads Attacks, China Activity Intensifies
    USCrowdstrike Blog
  55. When AI Guardrails Expose Vulnerabilities
    USCrowdstrike Blog
  56. Janus Vehicles: Entity-Centric Threat Modelling of Protocol-Conformant Vehicles under a Malicious-Manufacturer Assumption
    USarXiv cs.CR (Security)

    arXiv:2610.37059v1 Announce Type: new Abstract: Automotive cybersecurity commonly assumes benign end-product manufacturers and design authorities, focusing on supply-chain threats, external attacks, and compromised network participants…

  57. BadTime: An Effective Backdoor Attack on Multivariate Long-Term Time Series Forecasting
    USarXiv cs.CR (Security)

    arXiv:2508.04189v3 Announce Type: replace Abstract: Multivariate long-term time series forecasting (MLTSF) models are increasingly deployed in critical domains such as climate, finance, and transportation. Despite their growing importance…

  58. A Simple Presentation of Robust Membership Inference Attacks
    USarXiv cs.CR (Security)

    arXiv:2610.36458v1 Announce Type: new Abstract: In a membership inference attack, an attacker is given the data of a particular individual, and tries to determine if that individual was included in a particular dataset, given some kind of…

  59. Silent Blindness: Fault Vulnerability of Hardware Malware Detectors
    USarXiv cs.CR (Security)

    arXiv:2610.36612v1 Announce Type: new Abstract: Hardware malware detectors (HMDs) classify hardware performance-counter measurements to identify malicious execution. Faults in their decision logic or parameters can weaken detection and…

  60. Measuring Users' Security and Privacy Concerns About Third-Party Community Bots on Discord
    USarXiv cs.CR (Security)

    arXiv:2610.36626v1 Announce Type: new Abstract: Third-party community-based bots are central to how social platforms such as Discord work. They moderate content, manage roles, play music, and support AI-assisted conversations across…

  61. RHINO: Guided Reasoning for Mapping Network Logs to Adversarial Tactics and Techniques with Large Language Models
    USarXiv cs.CR (Security)

    arXiv:2510.14233v2 Announce Type: replace Abstract: Modern Network Intrusion Detection Systems (NIDS) generate vast volumes of low-level alerts that remain semantically fragmented, requiring labor-intensive manual correlation with…

  62. Beyond LWE: Post-Quantum Assumptions under a Counterfactual Break of the Dihedral Coset Problem
    USarXiv cs.CR (Security)

    arXiv:2610.36731v1 Announce Type: new Abstract: A refuted 2026 claim of an efficient quantum algorithm for the Dihedral Coset Problem (DCP) motivates this survey of the consequences of a genuine breakthrough. We distinguish a noiseless…

  63. The Content Leash: Mitigating Degenerate Solutions in Invertible Speaker Anonymization
    USarXiv cs.CR (Security)

    arXiv:2610.36768v1 Announce Type: cross Abstract: Voice anonymization aims to protect a speaker's identity while preserving the linguistic content of speech, enabling voice data to be shared and processed with reduced privacy risk. This…

  64. BioZKFHE: Scalable Encrypted Biometric Identification via Verifiable Homomorphic Similarity Evaluation
    USarXiv cs.CR (Security)

    arXiv:2607.22065v2 Announce Type: replace Abstract: Large-scale biometric identification in outsourced settings requires two properties simultaneously: biometric templates and queries must remain protected during computation, and the…

  65. AgentSentry: Mitigating Indirect Prompt Injection in LLM Agents via Temporal Causal Diagnostics and Context Purification
    USarXiv cs.CR (Security)

    arXiv:2602.22724v2 Announce Type: replace Abstract: Tool-augmented large language model (LLM) agents aggregate dialogue history, tool outputs, and retrieved content into an accumulated runtime state to complete multi-step tasks. This…

  66. Heads of Security: Learned Attention Gates Improve Prompt Injection Robustness
    USarXiv cs.CR (Security)

    arXiv:2610.37657v1 Announce Type: cross Abstract: We introduce Heads of Security (HoS), a prompt injection defense that enforces instruction-data separation via a small set of learned per-head attention gates. HoS learns how each head…

  67. Privately Clipping Heavy-Tailed Data
    USarXiv cs.CR (Security)

    arXiv:2610.36281v1 Announce Type: new Abstract: Summing vectors is a basic task in differentially private data analysis. Standard private sum algorithms clip vectors to some bound, sum them, and add noise scaled to the bound. The…

  68. Developing a Dynamic Multi-factor Authentication System for Securing Smart Organizations
    USarXiv cs.CR (Security)

    arXiv:2610.36926v1 Announce Type: new Abstract: As smart organizations increasingly rely on technologies like AI, big data, and IoT, securing these systems is crucial. This research introduces a dynamic multi-factor authentication (MFA)…

  69. Peak-DNA: Recurrent Morphology in Global Narrative Propagation
    USarXiv cs.CR (Security)

    arXiv:2610.36571v1 Announce Type: new Abstract: Global narrative streams exhibit noisy, fragmented, and asynchronous amplification bursts, making direct waveform comparison unreliable for recurrence analysis. This paper introduces…

  70. When Is a Relation Worth Trusting? Manipulation Costs, Exposure, and the Trust Trap in Multi-Relational Graph Learning
    USarXiv cs.CR (Security)

    arXiv:2610.37239v1 Announce Type: cross Abstract: Multi-relational graph detectors learn from clean data how much to trust each relation, but an adversary pays a different price to manipulate each relation and each action; taking such…

  71. Robust GNN Watermarking via Implicit Perception of Topological Invariants
    USarXiv cs.CR (Security)

    arXiv:2510.25934v2 Announce Type: replace-cross Abstract: Graph Neural Networks (GNNs) represent valuable intellectual property, yet watermarks based on out-of-distribution triggers can weaken under pruning, fine-tuning, and distillation…

  72. ORBITAL: Optimization of Robust Brain--Computer Inference Through Architecture-Aware Low-Precision Compression
    USarXiv cs.CR (Security)

    arXiv:2610.36411v1 Announce Type: cross Abstract: Compact electroencephalography (EEG) decoders enable embedded brain--computer interfaces (BCIs), but compression can alter adversarial robustness without proportional hardware savings. We…

  73. PHP2Shell and the Future of AI-Discovered Vulnerability Disclosure: Evidence, Verification, and Human Authority
    USarXiv cs.CR (Security)

    arXiv:2610.36334v1 Announce Type: new Abstract: AI-assisted vulnerability discovery can produce technically plausible findings faster than maintainers can validate, coordinate, and safely disclose them. The resulting challenge is not only…

  74. Constraining the Commit Boundary: A DIO/ODI-Informed Authorization Guardrail for Agentic Entity Resolution
    USarXiv cs.CR (Security)

    arXiv:2610.37407v1 Announce Type: cross Abstract: Agentic Entity Resolution (ER) systems iteratively plan actions, retrieve evidence, revise hypotheses, and propose graph mutations. An erroneous identity hypothesis may condition later…

  75. Watermark Basins: Black-Box Detection of Distilled LLMs
    USarXiv cs.CR (Security)

    arXiv:2610.37266v1 Announce Type: new Abstract: Proprietary language models and their reasoning traces are vulnerable to API-based distillation, where an adversary trains a student model on the teacher model's outputs. To obtain evidence…

  76. They Will See Johnny's Message: A Measurement Study of Post-Quantum Cryptography Across Email Delivery Paths
    USarXiv cs.CR (Security)

    arXiv:2610.36145v1 Announce Type: new Abstract: Harvest-now, decrypt-later (HNDL) attacks make the transition to post-quantum cryptography (PQC) urgent for email. This paper presents a systematic measurement that jointly examines PQC…

  77. Too many clients spoil the logs: distributed parsing, aggregation techniques and metrics
    USarXiv cs.CR (Security)

    arXiv:2610.36946v1 Announce Type: new Abstract: Log messages are employed to identify anomalies in IoT systems and have been widely used across many applications. To enable collaboration among infrastructures belonging to different…

  78. PrivacyEvalClaw: A Tool-Augmented Deliberative Agent for Auditing Privacy Leakage in LLM Jailbreak Responses
    USarXiv cs.CR (Security)

    arXiv:2610.36711v1 Announce Type: new Abstract: Reliable evaluation is a critical bottleneck in privacy red-teaming of large language models (LLMs): determining whether a jailbreak response actually reveals personal information can be…

  79. Asynchronous Verifiable and Complete Secret Sharing: From Design Patterns to System-Level Trade-offs
    USarXiv cs.CR (Security)

    arXiv:2610.37234v1 Announce Type: new Abstract: Modern secret sharing protocols promise robust secret sharing in fully asynchronous Byzantine systems, but their practical costs remain difficult to compare across papers, assumptions, and…

  80. SemDiff: Binary Similarity Detection by Diffing Key-Semantics Graphs
    USarXiv cs.CR (Security)

    arXiv:2308.01463v2 Announce Type: replace Abstract: Binary similarity detection is a critical technique that has been applied in many real-world scenarios where source code is not available, e.g., bug search, malware analysis, and code…

  81. SELF: A Robust Singular Value and Eigenvalue Approach for LLM Fingerprinting
    USarXiv cs.CR (Security)

    arXiv:2512.03620v2 Announce Type: replace Abstract: Safeguarding the Intellectual Property (IP) of Large Language Models (LLMs) against model theft represents a critical challenge in contemporary AI research. Although fingerprinting is…

  82. Haste Makes Waste? Towards Safe Few-Step Text-to-Image Models
    USarXiv cs.CR (Security)

    arXiv:2610.36665v1 Announce Type: cross Abstract: While few-step distilled text-to-image models have gained traction due to their fast generation times, the speed-up implications on their unsafe generation rates remain unknown. We uncover…

  83. Classical Verification of Quantum Computation Beyond the Polynomial Hierarchy
    USarXiv cs.CR (Security)

    arXiv:2610.37635v1 Announce Type: cross Abstract: It is not known whether a classical polynomial-time verifier can verify arbitrary polynomial-time quantum computations with an efficient quantum honest prover without computational…

  84. Cross-User Poisoning: User-Task Boundary Failures in Multi-User Collaborative Language Agents
    USarXiv cs.CR (Security)

    arXiv:2511.17671v2 Announce Type: replace Abstract: Language agents are moving from single-user assistants to shared collaborators in workspaces, forums, and group chats. A shared agent observes interleaved messages from multiple users…

  85. What Victims Take Away: Attack Reuse and Defense Revision after Prompt-Injection Attacks
    USarXiv cs.CR (Security)

    arXiv:2610.36422v1 Announce Type: new Abstract: Attack logs and red-team reports show targets the attacks used against them, to help them repair their defenses. Prompt injections against large language models (LLMs) are plain text, so such…

  86. FIDES-Live: Evaluating Behavioral Concordance in Sequential Trading Agents
    USarXiv cs.CR (Security)

    arXiv:2610.36476v1 Announce Type: new Abstract: Sequential trading agents produce decisions, orders, execution receipts, account states, and reports that need not agree, even when the final portfolio value is correctly computed. We present…

  87. Blindfold: Keeping the Audit Coin Secret from LLM Agents
    USarXiv cs.CR (Security)

    arXiv:2610.37055v1 Announce Type: new Abstract: Audit-based certification bounds the violation rate of an LLM agent by auditing a random fraction $\varepsilon$ of actions and reweighting the audited labels; its validity is meant to rest…

  88. The Envelope Layer: An Underestimated Surface of LLM Agent Defense
    USarXiv cs.CR (Security)

    arXiv:2610.37415v1 Announce Type: new Abstract: Prior work framed the envelope layer as the channel through which malicious content enters an agent, and established it as a defense surface independent of the LLM layer and the agent layer…

  89. RePAIR: Inverting Backdoor Attacks in Semantic Segmentation
    USarXiv cs.CR (Security)

    arXiv:2610.37015v1 Announce Type: cross Abstract: Despite significant advances in computer vision, semantic segmentation models remain vulnerable to backdoor attacks, posing a security risk. These attacks redirect pixels of a chosen victim…

  90. Log-To-Leak: Prompt Injection Attacks on Tool-Using LLM Agents via the Model Context Protocol
    USarXiv cs.CR (Security)

    arXiv:2610.36108v1 Announce Type: new Abstract: LLM agents increasingly access external tools through the Model Context Protocol (MCP), but the natural-language metadata used to describe those tools creates a prompt-injection surface…

  91. Revisiting Privacy Risks in Quantized Deep Neural Networks
    USarXiv cs.CR (Security)

    arXiv:2610.37488v1 Announce Type: new Abstract: Model quantization has become a fundamental technique for enabling the efficient deployment of large-scale deep neural networks (DNNs) in data-intensive applications by reducing model size…

  92. Who Gets Flagged? The Pluralistic Evaluation Gap in AI Content Watermarking
    USarXiv cs.CR (Security)

    arXiv:2604.13776v3 Announce Type: replace-cross Abstract: Watermarking is becoming the default mechanism for AI content authentication, with governance policies and frameworks referencing it as infrastructure for content provenance. Yet…

  93. Deanonymizing Monero Transactions in Tor Network
    USarXiv cs.CR (Security)

    arXiv:2607.07062v2 Announce Type: replace Abstract: Monero is a privacy-focused cryptocurrency that deploys the Dandelion++ protocol and incorporates anonymity networks (such as Tor and I2P) to prevent malicious attackers from linking…

  94. When POST Becomes GET: Investigating HTTP Method Interchangeability and Its Security Implications
    USarXiv cs.CR (Security)

    arXiv:2610.37474v1 Announce Type: new Abstract: HTTP method semantics are a common boundary between safe retrieval and state-changing operations; however, deployed web applications do not always enforce this boundary consistently. In this…

  95. No Tokenizer, Still a Gap: Cross-Lingual Fairness of Homoglyph and Invisible-Character Watermarking
    USarXiv cs.CR (Security)

    arXiv:2610.36432v1 Announce Type: cross Abstract: Homoglyph and invisible-character watermarks are edit-based watermarks. Instead of changing how a model generates text, they hide a signal in the text after it is written. They swap some…

  96. An Empirical Study of Undisclosed Charges in Hook-Based Decentralized Exchanges
    USarXiv cs.CR (Security)

    arXiv:2610.36682v1 Announce Type: new Abstract: Many decentralized exchanges use automated market makers (AMMs) to trade assets against liquidity pools. Modern AMMs allow third-party contracts, called Hooks, to customize this process…

  97. Quantifying and Prioritizing Stealthy Source-Manipulation Threats Against Multi-Source Anomaly Detectors
    USarXiv cs.CR (Security)

    arXiv:2610.36693v1 Announce Type: new Abstract: Security pipelines often fuse several data sources and protect a downstream model with a cross-source anomaly detector. Such detectors are usually judged by their clean false-positive rate…

  98. Penny for Your Prompts: A Measurement Study of Contextual Leakage in AI Chat Platforms
    USarXiv cs.CR (Security)

    arXiv:2610.37467v1 Announce Type: new Abstract: AI chatbot platforms have become one of the most popular web interfaces users interact with online, routinely receiving personal and intimate information shared under an expectation of…

  99. Pricing Cross-Bank Context in Federated Anti-Money Laundering
    USarXiv cs.CR (Security)

    arXiv:2610.35996v1 Announce Type: new Abstract: Most graph-based Anti-Money Laundering (AML) methods assume access to a pooled transaction graph containing all accounts and payments, which in reality does not exist. Federated learning…

  100. HqcLethe: Streaming HQC Encapsulation for Tiny Stateless Servers
    USarXiv cs.CR (Security)

    arXiv:2610.36808v1 Announce Type: new Abstract: Stateless post-quantum authenticated key exchanges such as post-quantum WireGuard hedge against unexpected cryptanalytic advances by combining a lattice KEM with code-based Classic McEliece…

  101. Nash Incentive-compatible Online Mechanism Learning via Weakly Differentially Private Online Learning
    USarXiv cs.CR (Security)

    arXiv:2407.04898v2 Announce Type: replace-cross Abstract: We study a multi-round mechanism design problem, where we interact with a set of agents over a sequence of rounds. We wish to design an incentive-compatible (IC) online learning…

  102. Bringing Hardware Security Education to All: A Seamless Curriculum Integration Approach
    USarXiv cs.CR (Security)

    arXiv:2610.36219v1 Announce Type: new Abstract: Cybersecurity education has predominantly focused on software and network security, under the assumption that the underlying hardware is trustworthy. Threats introduced through the global…

  103. SpecGuard: Detecting Multi-Turn LLM Jailbreaks via Specification-Based Runtime Monitoring
    USarXiv cs.CR (Security)

    arXiv:2610.36258v1 Announce Type: new Abstract: Multi-turn jailbreaks turn dialogue safety into a trace property: a violation may become identifiable only when user intent and assistant behavior are considered across multiple turns…

  104. Detecting and Mitigating Query-Based System Prompt Leakage Attacks
    USarXiv cs.CR (Security)

    arXiv:2610.36295v1 Announce Type: new Abstract: Large Language Models (LLMs) rely on proprietary system prompts to enforce task directives, safety policies, and access controls. These prompts are vulnerable to query-based leakage attacks…

  105. UniPCH: A Unified Framework for Cross-Chain Payments via Payment Channel Hubs
    USarXiv cs.CR (Security)

    arXiv:2610.36866v1 Announce Type: new Abstract: Payment channel hubs (PCHs) enable private cross-chain payments, but existing protocols remain difficult to extend across heterogeneous blockchains. Their payment flows often depend on how…

  106. Evaluating Federated Vision-Language and Vision-Only Models for Privacy-Aware Breast Cancer Imaging
    USarXiv cs.CR (Security)

    arXiv:2610.36379v1 Announce Type: cross Abstract: Artificial intelligence (AI) shows promise for breast imaging, but limited labeled data, institutional data fragmentation, and privacy concerns hinder centralized training. We investigate a…

  107. On the Reliability and Stability of Selective Methods in Malware Classification Tasks
    USarXiv cs.CR (Security)

    arXiv:2505.22843v4 Announce Type: replace Abstract: The performance figures of modern drift-adaptive malware classifiers appear promising, but does this translate to genuine operational reliability? The standard evaluation paradigm…

  108. GKS Fully Homomorphic Encryption: Security Analysis
    USarXiv cs.CR (Security)

    arXiv:2610.37061v1 Announce Type: new Abstract: All traditional encryption schemes share a fundamental operational vulnerability: encrypted data must be decrypted before it can be used. Fully homomorphic encryption eliminates this…

  109. LineageMark: Multi-user White-box Watermarking for Contribution Tracing in Model Derivation Chains
    USarXiv cs.CR (Security)

    arXiv:2606.17123v2 Announce Type: replace Abstract: In open large language model (LLM) ecosystems, models are frequently adapted across multiple domains and applications, forming multi-stage derivation chains. Consequently, tracking and…

  110. SecDMP: Provably Secure Data Memory-Dependent Prefetching (Extended Version)
    USarXiv cs.CR (Security)

    arXiv:2610.37454v1 Announce Type: new Abstract: Data memory-dependent prefetchers (DMPs) are a microarchitectural optimization recently introduced in mainstream desktop and server CPUs. Although DMPs offer performance improvements, they…

  111. Split Federated Unlearning
    USarXiv cs.CR (Security)

    arXiv:2610.36465v1 Announce Type: cross Abstract: Split federated learning (SFL) combines partitioned model execution with federated aggregation to support collaborative learning across distributed Web and edge environments. However…

  112. ISC Stormcast For Wednesday, October 28th, 2026 https://isc.sans.edu/podcastdetail/10154, (Wed, Oct 28th)
    USSANS ISC