Latest news

Advertisement
  1. AI Security via arXiv cs.CR

    Secure-CUA: Controlling Untrusted Influence in Computer-Use Agents

    arXiv:2610.09469v1 Announce Type: new Abstract: Computer-use agents (CUAs) perform tasks across applications (such as desktops, mobile apps, and web browsers) by…

  2. Vulnerabilities via arXiv cs.CR

    Constitution-Guided Watermarking

    arXiv:2610.09552v1 Announce Type: new Abstract: Watermarking enables language model providers to identify text generated by their models. However, its desired properties…

  3. AI Security via arXiv cs.CR

    MARS: Malware Analysis with Rule-Based Scoring of LLM Claims

    arXiv:2610.09553v1 Announce Type: new Abstract: Large language models can triage malware through direct verdicts or behavioral claims scored by an external policy. We…

  4. AI Security via arXiv cs.CR

    Correct Answers, Unsupported Findings: Evidence Binding in Forensic Reconstruction of LLM Agent Logs

    arXiv:2610.09581v1 Announce Type: new Abstract: Forensic reconstruction of LLM-agent actions requires not only recovering the correct value, but establishing which…

  5. AI Security via arXiv cs.CR

    Automotive Hardware Attacks: An Architect's Guide to TARA

    arXiv:2610.09697v1 Announce Type: new Abstract: Threat analysis and risk assessment (TARA) according to ISO/SAE 21434 treats implementation-level hardware attacks…

  6. Vulnerabilities via arXiv cs.CR

    Understanding and Mitigating Token-Pruning-Induced Vulnerabilities in VLMs

    arXiv:2610.09703v1 Announce Type: new Abstract: Token-Pruning accelerates Vision-Language Models by removing redundant visual tokens, yet its safety implications remain…

  7. Vulnerabilities via arXiv cs.CR

    Black-Box Adversarial Patch Attacks on VLAs via Ancestor VLM Exploitation

    arXiv:2610.09708v1 Announce Type: new Abstract: Vision-Language-Action models (VLAs) are increasingly deployed in safety-critical physical environments, yet their…

  8. Breaches via arXiv cs.CR

    Trust a Few: The Weakest Assumptions a Protocol Needs

    arXiv:2610.09730v1 Announce Type: new Abstract: Protocol verifiers check whether a protocol meets a security goal under stated trust assumptions, such as that a key is…

  9. Advertisement
  10. Threat Intel via arXiv cs.CR

    Faster PMNS Multi-precision Multiplications Using Truncated Montgomery Technique

    arXiv:2610.09767v1 Announce Type: new Abstract: The Polynomial Modular Number Systems (PMNS) aim to represent elements of fields or rings of large characteristics using…

  11. AI Security via arXiv cs.CR

    Formal Runtime Verification for Tool-Using LLM Agents: An Offline Same-Benchmark Study on AgentDojo and STAC

    arXiv:2610.09793v1 Announce Type: new Abstract: Guardrails for tool-using LLM agents are usually application-specific rules, which makes multi-step, data-dependent…

  12. Ransomware via ransomware.live

    🏴☠️ Lockbit5 has just published a new victim : consilio.com

    Consilio is a global legal software and services company that provides technology solutions for cros...

  13. Ransomware via ransomware.live

    🏴☠️ Lockbit5 has just published a new victim : capitalbankhaiti.biz

    Capital Bank SA provides a comprehensive range of traditional and online banking services tailored f...

  14. Ransomware via ransomware.live

    🏴☠️ Lockbit5 has just published a new victim : grunenwald.com

    GRUNENWALD Créateur français d'afficheurs sportifs haute performance depuis 1960, vous propose...

  15. Ransomware via ransomware.live

    🏴☠️ Lockbit5 has just published a new victim : avsa.com.ar

    Argentina Valores S.A. is a financial services company focused on capital markets, registered with t...

  16. Vulnerabilities via oss-security

    Re: CVE-2026-59265: Apache OpenOffice: Opening a malicious document can lead to system takeover

    Posted by Demi Marie Obenour on Oct 07 Does this apply to LibreOffice?

  17. AI Security via ransomware.live

    🏴☠️ Nightspire has just published a new victim : Asociación de Escribanos del Uruguay

    [AI generated] The Asociación de Escribanos del Uruguay is a professional association representing notaries (escribanos) in Uruguay. It operates within the legal…

  18. Vulnerabilities via oss-security

    CVE-2026-71896: Apache DolphinScheduler: Missing Authorization Checks Allow Unauthorized Disclosure of User Account Information

    Posted by Wenjun Ruan on Oct 07 Severity: critical Affected versions: - Apache DolphinScheduler before 3.4.3 Description: An authorization vulnerability in Apache…

  19. AI Security via ransomware.live

    🏴☠️ Nightspire has just published a new victim : INI Fiocruz

    [AI generated] INI Fiocruz refers to the Instituto Nacional de Infectologia Evandro Chagas, a research and hospital unit of the Oswaldo Cruz Foundation (Fiocruz) in…

  20. AI Security via ransomware.live

    🏴☠️ Nightspire has just published a new victim : Lumabuilt

    [AI generated] N/A I don't have reliable, verified information about a company called "Lumabuilt." Rather than provide potentially fabricated details about its industry…

  21. Vulnerabilities via oss-security

    CVE-2026-71895: Apache DolphinScheduler: Missing Authorization Checks Allow Non-Admin Users to Retrieve Kubernetes Credentials

    Posted by Wenjun Ruan on Oct 07 Severity: important Affected versions: - Apache DolphinScheduler 3.1.0 before 3.4.3 Description: An authorization vulnerability in Apache…