Zoom, Salesforce, BBC Among 178 Websites Hit by Cyberattack: What It Means for Yours
November 19, 2026
November 19, 2026
%20(1).jpg)
Key takeaways
Introduction
Zoom, Salesforce, the BBC, CNN, IKEA, HSBC and Equifax were among 178 organizations whose websites served malicious code to visitors on November 16, 2026. The code was built to skim card details, swap crypto wallet addresses and target older iPhones, and it arrived through a website-testing tool these companies trusted. None of them had to be hacked directly. This article covers which sites were hit, what the companies said, how one attack reached them all, and what visitors and businesses should do now. NZMinds applies its Control-Transparency-Recovery (CTR) framework to exactly this kind of risk.
In this article
Security researcher Scott Helme of Report URI found malicious code on the websites of 178 organizations across 20 sectors. BleepingComputer's own scans confirmed that Zoom, CNN, the BBC, Salesforce, GitLab and Air France loaded or tried to load the malware.
Who was affected, by sector
Sources: sector counts from Report URI; named websites from Newsweek and Entrepreneur.
Newsweek has the full list of affected organizations. As The National CIO Review notes, being on it does not mean a company's internal systems were breached. Helme estimates the code may have reached roughly 34 million page visits, and the attacker's wallets received about $7,040.
Video: news coverage of the cyberattack.
Salesforce said it is investigating unauthorized script changes and disabled the scripts as a precaution.
IKEA said it is working to determine whether there was any impact on its digital services and customers.
Equifax said it immediately investigated, stopped using the affected tool and found no evidence its systems were compromised.
Zoom, HSBC, CNN and Apple did not immediately respond to Reuters' requests for comment.
The tool's vendor, Optimizely, said an unauthorized party exploited a vulnerability in its platform. It fixed the flaw and appointed Mandiant to investigate. Helme called the incident "a near miss, and far too close for comfort."
All of these websites ran the same third-party testing tool. The attacker slipped a loader into the custom code it runs on every page, and because the file came from the vendor's trusted servers, browsers let it in.
What the code did. It swapped crypto wallet addresses, captured card details from checkout forms, stole cookies and tried to load iPhone exploit kits. There is no evidence any Apple device was compromised.
Why security policies missed it. Report URI checked the Content Security Policy (CSP) on affected sites:
Case study: how affected sites fared, by CSP setup
Source: Report URI, November 16 to 17, 2026.
These are well-resourced security teams. A policy that trusted the vendor's domain was correct on November 15 and wrong on November 16.
Used any of these sites on November 16? Take these steps today, based on Helme's advice and Newsweek's guidance:
It means trusting a vendor and controlling what runs on your pages are separate decisions, and only the second one protects your visitors.
At NZMinds, we see a third-party script the way we see an AI agent: both run with your authority, following instructions you did not write. So we apply the same framework to both.
The NZMinds Control-Transparency-Recovery (CTR) framework for third-party scripts
Control: what can it do? Zero access by default. Load each script only where needed and keep it off checkout and login.
Transparency: what is it doing now? Keep a live inventory with named owners and alert the moment a script changes.
Recovery: how fast can you pull it? A kill switch that needs no code deploy, plus credential rotation and customer notification plans.
CTR began as NZMinds' standard for AI agents. Read our analysis of the Google Antigravity incident.

Run this with engineering and marketing together.
Control
Transparency
Recovery
Security and compliance engineering. We build compliance-first systems for banking, insurance and healthcare clients, and apply that discipline to client-side code through CSP design, script governance and recovery planning. See our cybersecurity capability.
HawkShield.
HawkShield is NZMinds' cybersecurity product for email and browser protection, with threat detection, compliance monitoring and data visibility. This attack reached people through a browser they trusted. HawkShield brings visibility to that same layer for your workforce.
Websites of 178 organizations served malicious code on November 16, 2026, including Zoom, Salesforce, the BBC, CNN, IKEA, HSBC, Equifax, GitLab, Air France-KLM, Ledger and Peacock. Financial services was the hardest-hit sector, with 32 organizations. Newsweek has published the full list.
Not necessarily. Visiting an affected site does not mean your information was taken, and the number of people affected is unknown. As a precaution, update your iPhone or iPad, check card statements for November 16 and confirm any crypto transfers went to the right address.
Their websites loaded malicious code through a third-party testing tool, but that does not mean their internal systems were breached. Salesforce disabled the affected scripts and is investigating. Zoom did not immediately respond to requests for comment. The code targeted people visiting the websites.
A JavaScript supply chain attack compromises a third-party script that many websites load, so attackers reach every visitor without breaching the sites themselves. Here, one tool carried malicious code to 178 organizations' websites in a day. NZMinds offers a free review of your exposure.
NZMinds applies its Control-Transparency-Recovery framework: control what each script can do and where it runs, monitor every script for changes, and keep a tested kill switch ready. Start by removing third-party scripts from checkout and login pages. Book a free 30-minute review.

Unsure whether your gap is people or expertise? Download The Capacity & Capability Diagnostic.
References
