Daily Summary of Cybersecurity · Final editionFriday 20 November 2026 · UTC
Cybersecurity News: 20 November 2026
VMware vCenter and Cisco flaws exploited as AI agents find zero-day
Broadcom patched a critical VMware vCenter zero-day exploited by a suspected Chinese state-sponsored group, while researchers reported autonomous AI agents finding and exploiting a separate zero-day in a university server. Hackers also exploited a critical Cisco Expressway flaw in a Russian organization, and N-able traced a days-long email service outage to a cyberattack.
Bell Canada says 'agentic' hacker stole enterprise customer data
Bell Canada confirmed that an 'agentic' threat actor breached its systems on November 17, stealing enterprise customer data from a pre-production environment of its Maximo provisioning system. The compromised data included business contact information, ticket titles, and some IP addresses, though Bell stated that credentials and highly sensitive personal information were not accessed.
The use of the term 'agentic' suggests the attackers utilized AI-powered tools to navigate and exploit the company's infrastructure.
Amazon says Chinese hackers used AI agents to find zero-day
Amazon says suspected Chinese-linked attackers used the Cairn AI agent platform to autonomously find and exploit a zero-day path traversal flaw in MONAI Label, an open-source medical imaging application. The agents gained root access to a university server between September 8 and 10, then used the same flaw to breach a second university. Maintainers released a fix on November 18.
NSA resumes normal operations after PHP patching crash
The NSA has mostly returned to normal operations after patches to PHP disrupted its internal NSANet classified communications system for much of the week. The agency is operating under normal conditions with exceptions, while some facilities remain under Code Orange for non-essential staff. Officials said the outages were not a cyberattack and did not affect Middle East intelligence work or support for President Trump's Asia trip.
WhatsApp patches Android flaw suspected in scam campaign
WhatsApp patched a low-impact Android flaw, CVE-2026-23871, that it found while disrupting a scam campaign and believes a scammer may have used to make messages seem more credible. The issue involved incomplete validation of Click to WhatsApp ad-reply URLs and could have revealed a target's IP address. WhatsApp also fixed two medium-impact flaws, with no evidence either was exploited.
Wellington IT breach exposes payee details for 40 Irish credit unions
Payment details belonging to members of more than 40 Irish credit unions were accessed by hackers who exploited a vulnerability in software supplier Wellington IT. The attack occurred in early October but was not identified as a data breach until November 15, leading to criticism over the six-week delay in notification. The incident affected payee records stored in mobile banking apps, with no evidence that member funds were accessed.
INC ransomware claims Boeing breach with 33 years of operational data
The INC ransomware gang says it breached Boeing and stole nearly four million operational files spanning more than three decades. Boeing has not confirmed the claim, but screenshots appear to show internal data, and Cybernews research may link the alleged breach to Boeing Sikorsky Aircraft Support, a military helicopter maintenance unit. If technical or customer details were stolen, the incident could create national security and phishing risks.
Ukraine and FBI arrest ransomware gang linked to $40.5 million ransom
Ukrainian authorities and the FBI arrested four suspected members of a ransomware group that allegedly extorted more than $40.5 million from at least 28 companies worldwide. Investigators said the group stole data, encrypted systems and used AI voice calls to impersonate IT support and trick employees into revealing credentials or installing remote-access software.
SBC Medical Group breach affects more than 2 million people in Japan
California-based SBC Medical Group said hackers breached an online platform it manages and stole personal and medical data for more than 2 million people in Japan. The SEC filing says basic information was taken for 1,624,000 individuals, while treatment, reservation and employee records were also affected, including 14 credit cards. A separate breach of third-party reservation service Tribeau affected about 19,000 people.
F5 releases emergency patches for critical BIG-IP flaws
F5 released hotfixes for five critical BIG-IP flaws that unauthenticated attackers can exploit for denial of service or remote code execution. The issues affect BIG-IP Access Policy Manager and iRules, and F5 also fixed a high-severity SAML issue. F5 said there are no mitigations. Shadowserver tracks more than 177,000 BIG-IP instances and 14,600 exposed APM fingerprints online.
Norwegian Air urges travelers to block cards after booking platform breach
Norwegian Air has asked around 5,000 customers in Norway and Sweden to block their bank cards after hackers compromised an external IT supplier's ticket booking platform. Attackers injected malicious code into the airline's websites to steal payment information, including card numbers and security codes, during a brief window on November 16.
The airline says its internal systems were not affected, but the breach may impact approximately 200 other companies that use the same supplier.
A reported outage at the US Department of Agriculture has disrupted financial tools used to pay farmers, with an internal email indicating the cause was a security incident. Staff were instructed to reset databases and passwords across all environments, though the USDA has not yet publicly verified the claims or confirmed if data was stolen.
The incident adds to a wave of recent technology disruptions affecting US organizations, including Microsoft services and air travel.
European Commission confirms AI-generated cyberattack disrupted its systems
The European Commission confirmed it was hit by a cyberattack that misused a publicly available AI tool, forcing the institution to take several applications offline. The incident was detected on October 22 and has been largely contained, with most affected services restored. The Commission stated there is no evidence of a data breach affecting staff as a whole, though the investigation remains ongoing.
JPMorgan, Visa and Mastercard cite third-party data leaks
JPMorgan Chase, Visa and Mastercard said outside service providers were hit after the N0n extortion group claimed to have stolen customer data. JPMorgan said attackers accessed marketing data held by vendors that may include customer names and email addresses, while Visa and Mastercard said their payment networks and card numbers were not affected.
None has identified the providers or how many customers were affected; N0n lists all three as unpaid.
ROMULUS ClickFix attacks deliver new remote access trojan
Researchers at Blackpoint Cyber said threat actors are using the ROMULUS ClickFix delivery platform to spread a previously undocumented remote access trojan similar to CNCMachineRMS. The malware can provide remote desktop control, screen streaming, microphone and camera capture, file management and system reconnaissance, and it communicates over DNS over HTTPS. ROMULUS also distributes Legion Loader through separate chains.
Unpatched Windows FastUnlock flaw lets locked PC run programs
A newly disclosed technique called FastUnlock lets someone with physical access to a locked Windows computer start programs without signing in, using the built-in Narrator screen reader and a reachable Microsoft account. ThreatLocker tested the method and said it can run a program in a limited, non-administrator account. Microsoft has not patched it or confirmed real attacks; it recommends a startup PIN and shutting down or hibernating PCs.