Cybersecurity Brief

Thursday 29 October 2026

In this brief Sony, Okta, Cisco, Japan Airlines, University of New Mexico, Telegram, +39 more
Vendors
Sony, Okta, Cisco, Japan Airlines, University of New Mexico, Telegram, GitLab, Adobe, Oracle, Apple, Microsoft, Cursor, Claude, VMware, NIST, Fragnesia, Hive, Firefox, OpenAI, GitHub, Google, Gitee, Staffordshire Police, CREST, Virtualine, DDoSecrets, Alpine Linux, Veeam, Juniper Networks, Copilot, Codex, Nefilim, Anthropic, Meta, Talos, Visual Studio Code
Threat actors
Volt Typhoon, GOLD OGRE, Hive, Yoba, GhostAction, Nefilim
CVEs
CVE-2026-9783, CVE-2026-101767, CVE-2026-41709

Major Incidents or Breaches

  • Sony Network Communications, a subsidiary ISP, confirmed a data breach exposing approximately 2.5 to 2.6 million customer records, including personal information, 54,401 email account passwords, and 16 bank account records [22][29].
  • Over 41,000 US Department of Homeland Security (DHS) employee records, including undercover ICE agents, were leaked online by a hacktivist group, exposing names, offices, contact details, and assignments. The data was reportedly stolen from an Okta single sign-on system earlier in 2026 and published by DDoSecrets [28][43][66][72].
  • The Belgian Federal Public Service Finance temporarily suspended tax services following a cyberattack, which occurred just before a filing deadline [45].
  • Japan Airlines disclosed a breach affecting about 2,000 customers, with stolen passport numbers, frequent flyer numbers, email addresses, and phone numbers [48].
  • The University of New Mexico restored hospital systems after a cyberattack that forced ambulance diversions and class cancellations at its medical school [46].

Newly Discovered Vulnerabilities

  • A critical vulnerability in UEFI firmware, widely used in servers and data centers, allows attackers to execute code early in the boot process, impacting most servers globally [42].
  • Cisco released patches for a high-severity flaw in ASA and FTD VPN software that has been actively exploited to crash firewalls via remote, unauthenticated access. This marks the second exploited Cisco firewall VPN zero-day since August [6][19][31].
  • A flaw in Telegram for Android allowed attackers to execute code on a victim’s device when answering a voice call. Only an ordinary phone number was needed for exploitation [3].
  • Attackers attempted to exploit a critical GitLab vulnerability (CVE-2026-9783) on Alpine Linux’s GitLab server shortly after a patch was released [8].
  • Adobe, Veeam, and Juniper Networks issued patches for critical vulnerabilities that could lead to arbitrary code execution, authorization bypass, and privilege escalation [26].
  • Oracle E-Business Suite flaw CVE-2026-101767 is under active exploitation attempts, with decoys recording exploratory exploits [63][52].
  • Apple released a security update for older iPhones and iPads, backporting a fix for a CoreGraphics flaw reportedly exploited in targeted attacks [90].
  • Researchers demonstrated that logic bugs alone, without memory corruption, can be chained to escape the Microsoft Edge sandbox and execute code on Windows systems [14].
  • Researchers found vulnerabilities in the browsers built into several AI coding assistants (VS Code, Cursor, Copilot, Codex, Claude Desktop), allowing web pages to break out of the assistant’s browser sandbox [61].

Notable Threat Actor Activity

  • The China-linked APT group Volt Typhoon used a covert, ephemeral access network known as “Silent Train,” built from rapidly rotating chains of compromised devices, to conduct zero-day exploitation and strategic intrusions targeting hundreds of critical infrastructure networks [12][23][62][74][30].
  • Volt Typhoon also exploited a Cisco Catalyst SD-WAN flaw to deploy custom malware (KeyDrift) designed to steal encryption keys, with infrastructure overlapping the Silent Train network [9][30][87].
  • Chinese state-sponsored actors exploited a VMware ESX zero-day logging flaw (CVE-2026-41709) to reduce audit trail visibility, allowing operations by attackers with administrative privileges to go unlogged [10][33].
  • A suspected China-nexus threat actor leveraged PHP2Shell vulnerabilities to root 25 servers and breach a US government-affiliated network, including chaining with a high-severity Fragnesia Linux kernel privilege escalation bug [11][15][17].
  • The GOLD OGRE threat group used AI in the development, orchestration, and execution of Aur0ra ransomware operations [84].
  • A Ukrainian affiliate of Hive and Nefilim ransomware gangs was sentenced to 70 months in US prison for attacks targeting businesses and hospitals [34][53].
  • Attackers used fake ChatGPT and Claude projects to distribute Yoba malware, hiding malicious code behind signed binaries [60].
  • Malicious Firefox extensions impersonated popular crypto wallets to steal recovery phrases and private keys [89].
  • A threat actor impersonated AI bots from OpenAI, Anthropic, and Meta to scan for exposed credentials [39].
  • Attackers abused legitimate Mobile Device Management (MDM) platforms to deploy rogue remote management tools, turning device management into a pathway for remote access and persistence [93].
  • GitHub issues were weaponized in a supply chain attack campaign targeting 61 public repositories in less than an hour, exploiting vulnerable workflows [50].
  • AI is increasingly used both offensively and defensively in cybersecurity. Google introduced Gemini 4 Carbon, a frontier AI model outperforming predecessors in vulnerability discovery, exploitation, and patching [1][64].
  • Cisco open sourced Project Quarry, a rules language to detect AI agent attacks that emerge across sequences of benign actions [25].
  • Cisco Talos observed a new AI-enabled command-and-control framework in the wild, dubbed NIK C2 [86].
  • AI agents are now capable of controlling authenticated browsers, interacting with corporate applications, and accessing sensitive data while mimicking legitimate users [16].
  • The supply chain attack campaign “GhostAction” shifted toward targeting developer endpoints and cloud access, indicating evolving tactics among cybercrime groups [37].
  • Online scam volumes have doubled and now outpace phishing, with over six million scam detections in Q2 2026 [55].
  • Phishing remains the leading initial access technique, while ransomware and pre-ransomware cases have increased, with finance and public administration being the top-targeted industries [85].
  • Attackers are increasingly hiding malware behind signed binaries of AI tools to evade detection [60].
  • Credential exposure continues to be a significant risk, with 26,400 valid cloud and AI credentials found in public Gitee repositories [5].
  • Attackers are turning GitHub issues into attack vectors for supply chain campaigns, automating exploitation of vulnerable workflows [50].
  • The use of AI-enabled tools such as Gemini 4 Carbon and NIK C2, as well as AI-assisted attack orchestration by groups like GOLD OGRE, demonstrates the growing sophistication of threat actors [1][64][84][86].

Regulatory or Policy Developments Affecting the Security Industry

  • The UK released a shortlist of age verification methods for its upcoming under-16 social media ban, noting that some adults will also require verification [49].
  • The UK Information Commissioner’s Office (ICO) reprimanded Staffordshire Police following a serious breach of personal information related to domestic abuse cases [56].
  • CREST published a framework advocating a whole-ecosystem approach to strengthening cyber resilience across the financial services sector [54].
  • The US National Institute of Standards and Technology (NIST) published guidance for federal agencies on prioritizing post-quantum cryptography adoption to meet the 2031 transition deadline [57].
  • Microsoft again urged Entra ID administrators to migrate users to phishing-resistant authentication methods ahead of the planned retirement of SMS and voice authentication in February 2027 [24].
  • A federal appellate court ruled that West Virginia’s law barring disclosure of public officials’ personal data is unconstitutional [47].
  • US law enforcement seized the primary domain of Virtualine, a major bulletproof hosting provider used to facilitate cybercrime and ransomware operations [20][71].

Sources

  1. Sony ISP subsidiary data breach exposes 2.5 million customer records - bleepingcomputer.com
  2. Hackers Steal 2.6 Million Records From Sony’s Japanese ISP So-net - securityweek.com
  3. Over 41,000 DHS Employee Records Leaked Online - securityweek.com
  4. Massive Leak Exposes 41,000 Homeland Security Workers—Including Undercover ICE Agents - wired.com
  5. Hacktivists leak 41,000 DHS staff records, revealing undercover ICE agents - helpnetsecurity.com
  6. Leak site publishes data on more than 41,000 DHS employees obtained by hacktivists - cyberscoop.com
  7. Cyberattack disrupts Belgian tax services before filing deadline - therecord.media
  8. Japan Airlines warns of data breach as country grapples with multiple cyber incidents - therecord.media
  9. University of New Mexico restores hospital systems, urges no contact with hackers - therecord.media
  10. Most Servers on the Planet Are Vulnerable to This Firmware Code-Execution Flaw - wired.com
  11. Cisco Patches Actively Exploited ASA and FTD VPN Flaw That Can Crash Firewalls - thehackernews.com
  12. New Cisco firewall zero-day exploited to crash SSL VPN devices - bleepingcomputer.com
  13. Attackers Exploit Second Cisco Firewall VPN Zero-Day Since August - securityweek.com
  14. Telegram Android Flaw Lets Attackers Run Code When Users Answer Voice Calls - thehackernews.com
  15. Attacker Tried to Exploit Alpine Linux’s GitLab via CVSS 10.0 Flaw Its Ruby Left Open - thehackernews.com
  16. Adobe, Veeam, Juniper Networks Patch Critical Vulnerabilities - securityweek.com
  17. Exploitation attempts against recently patched Oracle EBS flaw have begun (CVE-2026-101767) - helpnetsecurity.com
  18. Oracle celebrates end of Cybersecurity Awareness Month with quadruple trouble - theregister.com
  19. Have an older iPhone or iPad? Apple has a security update for you - malwarebytes.com
  20. Researcher Reveals How Logic Bugs Alone Cracked Microsoft Edge in $175,000 Pwn2Own Win - thehackernews.com
  21. Researchers broke the browsers built into VS Code, Cursor, Copilot, Codex and Claude Desktop - helpnetsecurity.com
  22. Volt Typhoon Used Covert Silent Train Network in Four Zero-Day Exploitation Campaigns - thehackernews.com
  23. Volt Typhoon hides zero-day attacks using 'Silent Train' framework - bleepingcomputer.com
  24. Volt Typhoon targeted enterprise control planes via ever-changing “Silent Train” relay network - helpnetsecurity.com
  25. Researchers uncover Volt Typhoon’s covert, ephemeral network - cyberscoop.com
  26. Cisco SD-WAN Zero-Day Exploitation Tied to Volt Typhoon’s ‘Silent Train’ Network - securityweek.com
  27. Hackers Exploited Cisco SD-WAN Flaw to Plant Malware Built to Steal Encryption Keys - thehackernews.com
  28. Inside KeyDrift: Analyzing historical UAT-11473 activity on Cisco SD-WAN - blog.talosintelligence.com
  29. China-Linked Hackers Exploited VMware ESX Zero-Day to Reduce Audit Trail Visibility - thehackernews.com
  30. Chinese Hackers Quietly Used Low-Severity ESXi Logging Flaw as Zero-Day - securityweek.com
  31. Attackers Exploit PHP2Shell Flaws in Internet-Exposed Palo Alto Networks Management Interfaces - thehackernews.com
  32. PHP2Shell Attack Roots 25 Servers, Breaches U.S. Government-Affiliated Network - thehackernews.com
  33. CISA warns of Fragnesia Linux flaw chained in PHP2Shell attacks - bleepingcomputer.com
  34. AI used in Aur0ra ransomware testing and execution - sophos.com
  35. Ukrainian Hive and Nefilim Ransomware Affiliate Gets 70 Months in US Prison - securityweek.com
  36. Serial ransomware crook gets 70 months behind bars for Nefilim, Hive roles - theregister.com
  37. Attackers Hide Malware Behind Signed ChatGPT and Claude Binaries - infosecurity-magazine.com
  38. Malicious Firefox extensions impersonate popular crypto wallets - malwarebytes.com
  39. Attacker Relentlessly Scans for Credentials by Impersonating AI Bots - darkreading.com
  40. Managed for Malice: How Attackers Turn MDM into Remote Access - huntress.com
  41. Attackers turn GitHub issues into supply chain campaign against 61 public repos in less than an hour - theregister.com
  42. Google's Gemini 4 Carbon Leads Benchmarks for Finding, Exploiting, and Patching Vulnerabilities - thehackernews.com
  43. Google says Gemini 4 Carbon beats its predecessors at finding and fixing security flaws - helpnetsecurity.com
  44. Cisco Open Sources Quarry to Catch AI Agent Attacks Hidden in Harmless Steps - securityweek.com
  45. NIK C2: The new AI-enabled command-and-control framework - blog.talosintelligence.com
  46. 3 Ways AI Agents Already Operate in Your Browser - bleepingcomputer.com
  47. GhostAction Hacking Wave Fizzles, But Attackers Lurk on in GitHub - darkreading.com
  48. Online Scam Volumes Double to Overtake Phishing - infosecurity-magazine.com
  49. IR trends Q3 2026: Actors impersonate IT help desks in vishing attacks as ransomware activity increases - blog.talosintelligence.com
  50. 26,400 Valid Cloud and AI Credentials Found Exposed in Public Gitee Repositories - thehackernews.com
  51. Shortlist of age checks for the UK's under-16 social media ban is out, 'some' adults will need verifying - theregister.com
  52. ICO Reprimands Staffordshire Police After Domestic Abuse Breach - infosecurity-magazine.com
  53. CREST Urges Whole-Ecosystem Approach to Financial Cyber Resilience - infosecurity-magazine.com
  54. NIST Publishes Guidance to Meet Federal Post-Quantum Deadlines - infosecurity-magazine.com
  55. Microsoft urges Entra ID admins to migrate users to passkeys - bleepingcomputer.com
  56. Court strikes down West Virginia law barring disclosure of public officials’ personal data - therecord.media
  57. Police seize Virtualine bulletproof service tied to ransowmare, cybercrime - bleepingcomputer.com
  58. Officials seize domain of top bulletproof hosting provider Virtualine - cyberscoop.com