Security news

• Bizarre Halloween Candy Courtesy of AI Tool Dall-E: Farte Cats, Anyone? - CNET

  17-10-2022 23:03 via cnet.com
  Be sure to save room for these hilarious treats this year.
  
  Read more

• When English becomes exploit code: The hidden risk inside MCP servers

  10-09-2026 22:28 via scworld.com
  
  A harmless-looking sentence can influence an agent's choices once it enters a model's context.
  
  Read more

• CISA: WatchGuard Firebox bug exploited in ransomware campaigns

  10-09-2026 21:47 via scworld.com
  
  Ransomware gangs are exploiting a WatchGuard Firebox flaw that CISA flagged nine months ago.
  
  Read more

• A security framework for coding agents and their harnesses

  10-09-2026 20:59 via scworld.com
  
  The five layers of securing AI agent harnesses.
  
  Read more

• Advertisement

• What CIO-CISO alignment looks like when it's working

  10-09-2026 19:34 via scworld.com
  
  Cushman & Wakefield leaders share five practices for building an effective CIO-CISO partnership.
  
  Read more

• A credo for the AI era: Trust, but Decompile

  10-09-2026 18:11 via scworld.com
  
  Here’s why defenders today have to take apart and analyze code the way a hacker would.
  
  Read more

• PaperCut MF/NG flaws attacked with hundreds of AI agents

  10-09-2026 15:55 via scworld.com
  
  The AI-assisted campaign enabled attackers to gain domain admin in as little as 5 minutes.
  
  Read more

• Skullcandy Dime 3 earbuds vulnerable to Bluetooth hijacking

  10-09-2026 15:38 via scworld.com
  
  The vulnerability, tracked as CVE-2025-20701, affects firmware version 1.0.0.28 of the Skullcandy Dime 3 earbuds, which utilize the Airoha Bluetooth Audio SDK, the Carnegie Mellon University CERT Coordination Center (CERT/CC) warned.
  
  Read more

• Advertisement

• Gigabud banking trojan uses app cloning to evade fraud detection

  10-09-2026 01:46 via scworld.com
  
  Researchers at Group-IB have identified that Gigabud is now being paired with Vwork, a modified version of the Shelter app, attributed to the GoldFactory threat group.
  
  Read more

• Over 36,000 Plex media servers remain unpatched against security vulnerabilities

  10-09-2026 01:45 via scworld.com
  
  The unpatched Plex Media Server instances are running versions 1.43.2 and earlier.
  
  Read more

• Ping YOUniverse: The next stage of human authentication

  04-09-2026 23:25 via scworld.com
  
  Ping Identity says its combination of biometrics and verifiable documentation is the next step in access management.
  
  Read more

• New Linux toolkit found in trojanized HAProxy targeting South Korean organizations

  04-09-2026 22:51 via scworld.com
  
  The implant, identified by Rapid7 Labs with medium confidence as originating from North Korean state-sponsored actors, targets entities in South Korea's automotive and media sectors.
  
  Read more

• PostGREShell vulnerability allows server takeover

  04-09-2026 22:44 via scworld.com
  
  The vulnerability stems from missing authorization in PostgreSQL's logical decoding feature.
  
  Read more

• Upwind Security raises $300 million in Series C funding

  04-09-2026 22:41 via scworld.com
  
  Upwind Security offers a platform that automatically maps cloud assets and refreshes data every 30 seconds to account for frequent configuration changes.
  
  Read more

• Coder platform targeted by attackers delivering malicious Terraform modules

  04-09-2026 22:38 via scworld.com
  
  The attack occurred between 07:35 UTC and 21:45 UTC on Monday, August 31.
  
  Read more

• Cisco addresses critical vulnerabilities in Nexus 9000 switches and IOS XR

  04-09-2026 22:35 via scworld.com
  
  The critical Nexus vulnerability (CVE-2026-20212) allows unauthenticated attackers to execute code with root privileges by binding to an unrestricted IP address and reaching TCP ports 43210 and 43211.
  
  Read more

• Thomson Reuters court software breach exposes sensitive data in Canada and US

  04-09-2026 22:32 via scworld.com
  
  The incident, detected on June 30, impacted the C-Track Canada system, exposing files from three Ontario courts.
  
  Read more

• Abliteration.ai launches service for uncensored AI models

  04-09-2026 22:23 via scworld.com
  
  Abliteration.ai aims to enable offensive cyber operations, red-teaming, and agent testing that other models refuse.
  
  Read more

• How to Map Controls Across Frameworks Without Losing Meaning

  23-08-2026 20:56 via scworld.com
  
  Move Beyond Crosswalks to Control-Based Compliance
  
  Read more

• TrueConf flaws enabling attacks on meeting participants added to KEV catalog

  22-08-2026 01:15 via scworld.com
  
  The flaws have been used by the Head Mare APT hacktivist group to spread PhantomCore malware.
  
  Read more

• Secure AI agent identity in private cloud and hybrid environments

  21-08-2026 23:29 via scworld.com
  
  Ping Identity adds self-managed identity controls to secure and govern AI agents in regulated environments.
  
  Read more

• Navy warns of multi-pronged adversary campaign targeting personnel and installations

  21-08-2026 23:02 via scworld.com
  
  The campaign includes direct threats and harassment via social media, doxing, drone activity near Navy assets, ground-level surveillance, physical attacks, and probes of security measures.
  
  Read more

• New Agent Tesla malware version uses emoji obfuscation to evade detection

  21-08-2026 22:58 via scworld.com
  
  The latest Agent Tesla campaign utilizes a JScript dropper that incorporates Unicode emoji characters to disrupt signature-based detection and obscure the malicious code.
  
  Read more

• Army seeks AI agents for cyber defense amid evolving threats

  21-08-2026 22:58 via scworld.com
  
  Project Griffin, also known as the Intelligent Response and Orchestration Node (IRON), is a pilot program designed to create an ecosystem of AI agents capable of analyzing data from the Army's extensive network sensors and autonomously executing defensive actions.
  
  Read more

• Researchers find way to weaponize Windows Defender's own driver

  21-08-2026 22:52 via scworld.com
  
  Check Point Research disclosed a technique that uses Microsoft Defender's boot-time remediation driver, BTR.sys, to perform arbitrary kernel-level file and registry operations on Windows 7 through Windows 11 25H2.
  
  Read more

• A ‘kill switch’ law only makes sense for a worst-case scenario

  20-08-2026 03:15 via scworld.com
  
  Congress wants an AI ‘kill switch’ – but we really need identity-based access.
  
  Read more

• The CISO Doesn't Own the Outcome: A Shared-Accountability Model for Security Decisions

  19-08-2026 21:51 via scworld.com
  
  Align Security Accountability with Decision Authority
  
  Read more

• Medusa ransomware group attacked more than 500 victims since 2021

  19-08-2026 20:31 via scworld.com
  
  Medusa ransomware hit 500-plus victims, exploiting critical flaws at increasing speed.
  
  Read more

• Perspective and lessons learned from mergers and acquisitions

  19-08-2026 18:45 via scworld.com
  
  M&A success depends on product fit, cultural alignment and disciplined post-deal execution.
  
  Read more

• Geekom admits malware found in legacy mini PC driver download

  19-08-2026 16:20 via scworld.com
  
  The compromised LAN driver, hosted on a legacy support page, allowed the Asruex backdoor to operate with administrator-level permissions.
  
  Read more

Follow @newloc_security on Twitter!