Skip to content
Malware spread through Optimizely hits 178 organizations, including BBC, CNN and Zoom

Malware spread through Optimizely hits 178 organizations, including BBC, CNN and Zoom

Helpnetsecurity •Sinisa Markovic • November 17, 2026

An attacker exploited a vulnerability in Optimizely’s web experimentation platform to plant malicious JavaScript on its customers’ websites, the company has confirmed.

Companies add Optimizely Web to their sites through a JavaScript snippet loaded from Optimizely’s servers.

“Earlier today, Optimizely identified a security incident in which an unauthorized party exploited a vulnerability in Experimentation and used it to add malicious JavaScript to some customer projects. We’ve fixed the vulnerability and removed the injected code from affected projects,” the company said in its latest status update .

“We’ve also taken steps to secure affected user accounts and put additional protections in place to prevent unauthorized code from being delivered.”

The attack was uncovered by security researcher Scott Helme of Report URI. “On 16 November 2026, malicious JavaScript was served to visitors of 178 organisations through the Optimizely Web experimentation platform. The list includes some of the most visited sites in the world: Zoom, the BBC, CNN, Salesforce, GitLab, IKEA, HSBC, Equifax, Air France-KLM, Ledger and Peacock,” wrote Helme.

“The malware was a multi-purpose kit. It swapped cryptocurrency wallet addresses for the attacker’s, skimmed payment cards, stole cookies and screenshots, and on iPhones and iPads it delivered the Coruna and DarkSword iOS exploit kits. Crypto has already been stolen, with verified transactions to the attackers,” he added.

The code was planted in the “Project JavaScript” field of customer projects. It kept loading from Optimizely’s own content delivery network, so visitors’ browsers treated it as trusted.

Helme’s team checked 1,611 Optimizely snippets and found 188 carrying the malware. He believes the number of victims is higher, since the team only examined the snippets it managed to find in time.

Banks and financial firms made up the largest group of victims, with 32 organizations, followed by technology companies and retailers.

Salesforce told Reuters it was investigating unauthorized Optimizely changes and disabling scripts as a precaution.

A payload that kept growing

The injected code loaded a script from a domain registered on the morning of the attack. The card skimmer arrived in the afternoon and went after card numbers, expiry dates and CVV codes typed into a site’s own payment forms. Card details entered into payment fields run by a separate provider were out of its reach.

On iPhones and iPads running iOS 18.6 or older, affected pages reloaded every 60 seconds, something visitors might have noticed.

Helme found that the payload was rebuilt 26 times over the course of the day, with the attacker adding features and wallets along the way.

A modest payout from a long-running operation

The attacker’s wallets received $7,040 before the stolen crypto, most of it Bitcoin, was moved to exchanges. Hudson Rock, which also investigated the campaign, says it has not been able to link those transfers to visitors of the affected sites.

The person behind the campaign was active well before this week. One of the wallets had been in use since March 2026, and the same injection code first appeared on Turkish websites in September.

The injected code was removed from most of the infected snippets the same day. By that evening, Helme found only one still infected, belonging to TalkTalk.

Three more services carried the code

Hudson Rock and researcher Kirk of Ohey traced poisoned scripts to three more services, namely Kissmetrics, StorifyMe and minute-ly, with the earliest tampered responses recorded on November 13. The deliveries hosting and loader code, but the researchers say that doesn’t prove a single operator was behind them or that the same way in was used.

Kirk documented later iOS stages built to steal keychain data, seed phrases and private keys from iPhones. “Hudson Rock has not independently reproduced those stages or demonstrated compromise of a real victim’s iPhone. A scan using an iPhone user-agent string is a delivery observation, not a Safari exploitation test,” the company explained .

Few sites had the right defenses

Helme’s team crawled the affected organizations’ websites to see whether a Content Security Policy (CSP) would have stopped the attack. Only 24 organizations, or 15%, had a policy strict enough to block it. Another 68 had no CSP at all. In 50 cases, the policy let the malware run and send stolen data back to the attacker.

Allowing Optimizely’s domain in a policy offered no protection, because the malware was delivered from that domain. Optimizely’s documentation recommends a setting that lets pages run code through JavaScript’s eval function. That setting allowed a backup copy of the malicious code to run on HSBC’s site.

“The important lesson here is that trusted JavaScript doesn’t always remain trustworthy. A website can be compromised through a third-party dependency without its own code or infrastructure ever being breached,” concluded Helme.