Optimizely Malware Scandal Affects Hundreds of Websites
Researchers at ReportURI discovered that hundreds of websites using Optimizely's personalization platform unknowingly served malware to visitors.

A recent discovery has revealed that hundreds of websites using the Optimizely website personalization platform have unknowingly served malware to their visitors. This malicious code targeted cryptocurrency wallets and stole sensitive information such as payment card data, cookies, and screenshots. The affected websites also delivered exploit kits designed to infect iPhones and iPads.
According to researchers at ReportURI, a total of 178 websites were compromised, including prominent names like CNN, Zoom, the BBC, and Salesforce. These websites had JavaScript code signed by Optimizely that was altered to include malicious components. When these components successfully installed on visitor devices, they replaced cryptocurrency wallet addresses with those controlled by the attackers.
The modified JavaScript code was designed to run on every page served by affected customers, making it a stealthy way for attackers to distribute their malware. Optimizely serves JavaScript snippets that combine company-provided code with customer-specific code. The compromised code was somehow embedded into these snippets and delivered from cdn.optimizely.com.
Researchers found that the malicious code had been signed with a valid certificate for the mentioned domain, making it difficult for security software to detect as malware. This sophisticated attack highlights the risks associated with using third-party services like Optimizely, which can inadvertently allow attackers to inject malicious code into websites.
The attackers didn't use a CDN cache poisoning attack to deliver the malware. They instead leveraged Optimizely's legitimate infrastructure, including its own domains, to serve the tainted code. This was done through the company's customer JavaScript files for www.optimizely.com and login.optimizely.com.
ReportURI researcher Scott Helme pointed out that the delivery mechanism didn't rely on a CDN cache poisoning attack. Instead, it used Optimizely's infrastructure to inject malicious code into websites. The exact method by which the attacker gained access is still unknown.
Helme noted that the attackers could have obtained write access through various means, including a compromised system or credential with cross-customer access. However, the specifics of how this occurred are for Optimizely to clarify. The company's representatives failed to respond to email queries regarding the incident.
The malicious scripts, named track.js, contained four components that became active simultaneously on affected devices. The responses from these devices varied depending on their platform and what they were doing when exposed to the final page delivering the code.
The malicious script's activities extend beyond just credit card information collection. It also engages in cookie theft, capturing sensitive data stored on users' devices. When it swaps an address next to a wallet or payment term, the script sends not only the document.cookie but also the original replaced address and a screenshot of the page to https://1688360.cc/hit.php.
This script's capabilities are not limited to individual exploits; it is also capable of delivering entire exploit kits. Specifically, it delivers the DarkSword and Coruna kits for iOS devices through a file named boot.js served from 5vobauua3ojpklx.icu and eil786npto9u5bq.icu. These files create hidden iframes that contain tailored exploit chains based on the device's iOS version.
The DarkSword kit, for example, can infect devices running versions 18.x of iOS, while Coruna can target versions 13 to 17. Interestingly, a separate chain is also present for devices with an iOS 26 fingerprint, but its purpose remains unclear. Researchers have noted attempts by hackers linked to the Kremlin to deliver DarkSword through compromised web analytics services earlier this month.
The list of websites identified as delivering these malicious scripts has reached 178 sites, including many well-known names. However, it is uncertain whether this list is exhaustive. ReportURI and security firm Cynet provide indicators of compromise that website administrators can use to identify potential vulnerabilities on their platforms.
Helme noted in an interview that Ars Technica itself was not affected by the malicious scripts, and there were no other indications of exposure. Website administrators are advised to review both the ReportURI post and the Cynet indicators to ensure their websites remain secure.
Researchers are still gathering information about the success of the exploit chains in delivering malware payloads to unsuspecting visitors. However, it's not uncommon for detailed information to be lacking in initial reports of mass malware attacks, as experts need time to thoroughly analyze the evidence.
The indicators provided by Cynet may help users identify if they have been affected, but applying this knowledge requires specialized skills that many individuals do not possess. As more details become available, readers are advised to check back for updates on this developing story with a conclusion that no one should visit websites without first verifying their security.
Facts based on reporting originally published by Ars Technica.
You may republish this story, in full or in part, if you credit Noti Group and link to it (licence CC BY 4.0). Photos are not included.












