BBC, CNN, Zoom and other major brands were caught in an Optimizely supply-chain attack. The websites were real. The dangerous code came from a service they trusted.
November 17, 2026
Imagine visiting a website you've used for years.
The URL is correct. The HTTPS connection is secure. You're not clicking a suspicious email or visiting a fake website.
Yet the browser quietly loads malicious code.
That's what happened on November 16, 2026.
Security researchers discovered that websites associated with 178 organizations, including BBC, CNN, Zoom, Salesforce, GitLab, HSBC and IKEA, were serving malicious JavaScript through Optimizely, a widely used website experimentation platform.
The code was designed to redirect cryptocurrency payments, steal payment card information and deliver attacks targeting iPhones and iPads.
And here's what makes this incident different from the phishing attacks people are constantly warned about:
You didn't have to visit a fake website to encounter the malicious code.
How Did 178 Organizations Become Involved?
Modern websites rarely operate entirely on code written by their owners.
They rely on outside services for analytics, advertising, customer support and website optimization.
Optimizely Web Experimentation is one of those services.
Companies use it to test different versions of their websites and improve how visitors interact with their pages.
To make this possible, websites load JavaScript from Optimizely's infrastructure.
On November 16, an attacker exploited a vulnerability in Optimizely and inserted malicious JavaScript into customer projects.
The altered code was then delivered through Optimizely's legitimate infrastructure.
Researchers at Report URI examined 1,611 JavaScript snippets and identified 188 infected scripts associated with 178 organizations.
This does not mean attackers penetrated the internal networks of all 178 organizations.
The problem was a shared service that could deliver malicious code to visitors across many websites.
What Could the Malicious Code Actually Do?
The investigation identified several dangerous capabilities.
1. Replace cryptocurrency wallet addresses
Imagine copying a cryptocurrency wallet address to send someone money.
The malicious script could replace that address with one controlled by the attacker.
You might believe you're sending money to the intended recipient while the transaction actually goes somewhere else.
Researchers observed cryptocurrency transfers to wallets associated with the attack.
However, the precise amount stolen from victims in this incident has not been fully established.
2. Collect payment card information
The code could capture card numbers, expiration dates and security codes entered into certain payment forms.
There was an important limitation: information entered inside separate, protected payment-provider frames could not be directly read through this mechanism.
The number of cardholders actually affected remains unknown.
3. Collect browser information
The malicious JavaScript could access certain readable cookies and capture page content.
That creates a risk beyond payment information, depending on what was available to the script on a particular page.
4. Target iPhones and iPads
Researchers also identified mechanisms for delivering two iOS exploit kits, known as Coruna and DarkSword.
These were intended to exploit vulnerabilities in Apple devices, particularly those running older software.
But identifying exploit delivery is not the same as confirming successful device compromise.
The number of iPhones or iPads successfully compromised in this incident has not been established.
What Happened After the Attack Was Discovered?
Report URI documented a series of changes to the malicious code throughout November 16.
The investigation identified an early injection in a BBC-related script at 04:21 UTC.
By 08:28 UTC, the main malicious payload was being loaded through the BBC website.
Payment card-skimming functionality appeared later in the day, followed by additional iOS attack-delivery mechanisms.
The attacker repeatedly modified the malicious code while the campaign was active.
Cleanup began that evening.
Optimizely subsequently acknowledged the security incident and said it had fixed the vulnerability, removed the injected code and secured affected accounts.
But removing malicious code doesn't automatically answer what happened to everyone who encountered it before the cleanup.
Visited an Affected Website? Here's What to Check.
If you visited one of the affected websites on November 16, don't assume your information was automatically stolen.
The risk depends on what code was delivered, what device you used and what actions you performed.
If you used an iPhone or iPad
Open your device settings and install the latest available security updates.
If you made a cryptocurrency payment
Check the completed transaction on the blockchain.
Compare the actual destination wallet address against the address you intended to use.
Don't rely only on the address shown in an old browser window or copied text.
If you entered payment card information
Review recent charges.
Enable transaction notifications if your card provider supports them.
Pay particular attention if you entered card details directly into a website's own form.
If your company uses Optimizely
Review Project JavaScript, experiment code, account access and related logs.
Investigate historical changes rather than checking only whether the script is clean today.
If customer payment or account pages loaded affected code, determine whether sensitive information might have been accessible.
The Bigger Question: Who Can Access Your Information?
Most online safety advice focuses on avoiding suspicious websites.
And that advice remains valuable.
But this attack exposes a different problem.
A real company can operate a legitimate website with a valid HTTPS connection and still deliver malicious code through a third-party service.
The visitor may have done everything they were told to do.
They checked the URL.
They recognized the organization.
They didn't click an unfamiliar link.
Yet the code running inside their browser could still have been altered.
That raises a question we rarely ask when entering personal information online.
When you give your information to one company, how many other companies' programs are allowed to interact with it?
A trusted website is not necessarily a website running only trusted code.
And most visitors have no practical way to know the difference.
The Optimizely incident was a warning about more than one vulnerability.
It revealed how much trust we place in software we never see.
Sources
Report URI — Scott Helme, November 16, 2026
World's Largest Brands Infected With Malware Via Optimizely
https://blog.report-uri.com/worlds-largest-brands-infected-with-malware-via-optimizely/
Optimizely — Official Incident Information
https://status.optimizely.com/
Reuters — November 17, 2026
Popular websites subverted in bid to steal crypto, target iPhone users
Top comments (0)