Security & privacy

Optimizely supply chain attack injected malware into 178 major websites

Attackers exploited Optimizely to serve malware on sites like BBC and Zoom, highlighting the critical need for strict Content Security Policies.

A broken digital shield failing to stop malicious script arrows from hitting a website.
Illustration generated for this article

On November 16, 2026, a security vulnerability in the Optimizely platform allowed attackers to inject malicious JavaScript into the websites of at least 178 major organizations. High-profile victims included the BBC, CNN, Equifax, Ledger, and Zoom, exposing their visitors to data theft and device exploitation without any breach of the companies' own infrastructure.

What happened

The incident began when an unauthorized party exploited a vulnerability in Optimizely Experimentation, a tool used for A/B testing and user experience customization. According to security firm Report URI, the attackers served malicious code to visitors of 178 organizations throughout the day. The payload was dynamic, rebuilt 26 times to evade detection, and delivered a multi-purpose malware kit capable of stealing payment card details, cookies, and screenshots. It also swapped cryptocurrency wallet addresses with those controlled by the attackers.

Beyond simple data theft, the injected code targeted specific hardware vulnerabilities. It deployed zero-click exploit kits known as DarkSword and Coruna, which chained together multiple security holes to seize control of outdated Apple iPhones and iPads. These exploits required no user interaction, executing silently as soon as the page loaded. Optimizely confirmed the incident, stating they identified the vulnerability, blocked access, deployed a fix, and removed the injected code from affected projects. They are currently working with third-party experts on a comprehensive investigation.

How it works

This attack illustrates the risks inherent in modern web development, where sites rely heavily on third-party JavaScript. The median homepage now loads at least 22 JavaScript files, many from external vendors like Optimizely. These scripts run with high privileges on the user’s device, allowing them to read browser data and modify page content. Because Optimizely snippets are designed to load early for accurate A/B testing, they often bypass consent dialogs, meaning the malware executed immediately upon page load for most visitors.

The spread of such attacks is largely enabled by weak or missing Content Security Policies (CSPs). A CSP is a browser security feature that lets site operators specify which domains are allowed to execute scripts. Without a strict CSP, if a trusted vendor like Optimizely is compromised, the attacker can inject code from new, malicious domains that the browser will trust by default. In this case, the lack of restrictive policies allowed the malicious scripts to run unhindered on the majority of affected sites.

Key details

  • Scope: At least 178 top web destinations were affected, including BBC, CNN, Equifax, Ledger, and Zoom.
  • Timeline: The attack occurred on November 16, 2026, with the payload rebuilt 26 times during the day.
  • Malware capabilities: The kit stole payment data, cookies, and screenshots, and swapped cryptocurrency addresses.
  • Device targeting: Zero-click exploits named DarkSword and Coruna targeted vulnerabilities in older iOS devices.
  • CSP failure: Only 15 percent of affected sites had CSPs that fully blocked the malicious payload; 68 sites had no CSP at all.
  • Financial impact: Attackers stole approximately $7,500 in virtual currency, plus an unknown number of payment card records.

Why it matters

For software engineers and technical leads, this incident underscores that trusting a third-party vendor is not a security strategy. Even if your own infrastructure is secure, a compromise in your dependency chain can give attackers full control over your user’s experience. The fact that Optimizely code runs early and outside consent flows makes it a high-value target. Developers must recognize that every external script is a potential entry point for supply chain attacks, especially as AI tools make it easier for attackers to generate evasive malware variants quickly.

The widespread absence of effective Content Security Policies remains a critical gap in web security. Report URI found that only 24 of the affected sites had CSPs that successfully blocked the attack, while overall CSP deployment among the top million websites sits at just 10 percent. Without these policies, organizations are operating in a "Wild West" environment where they run code without knowing its true origin or intent. Implementing strict CSPs is no longer optional for serious web applications; it is a fundamental defense against the inevitable next supply chain compromise.

What you can do

  • Audit all third-party JavaScript dependencies and remove any that are not strictly necessary for core functionality.
  • Implement a strict Content Security Policy that whitelists only known, trusted domains for script execution.
  • Monitor CSP violation reports regularly to detect unauthorized script loading attempts in real time.
  • Require vendors to provide transparency on their security practices and incident response capabilities.
  • Use subresource integrity (SRI) hashes for critical external scripts to ensure they have not been tampered with.
  • Educate users on the importance of using ad blockers or script-blocking extensions as a personal firewall against malicious JavaScript.
  • #supply chain security
  • #content security policy
  • #javascript
  • #web security
  • #optimizely

Tools from the Bytechap store

Keep reading

All stories