ROHAN TECH OFFICIAL • Technology • AI • Apps • Tutorialsinfo@rohantechofficial.com
ROHAN TECH OFFICIALTechnology • AI • Apps • Tutorials • Digital World
Home → Tech News → Optimizely Hack Turns Major Websites Into Malware Dispensers

Optimizely Hack Turns Major Websites Into Malware Dispensers

By Rohan Jaspal • November 18, 2026 • Updated November 18, 2026
Optimizely Hack Turns Major Websites Into Malware Dispensers

A flaw in Optimizely’s sign-up process let attackers turn some of the web’s biggest sites into malware dispensers. The attack, dubbed “OPTIMJACK”, injected malicious JavaScript into customer projects and spread it through Optimizely’s own content delivery network.

Security researcher Scott Helme examined 1,611 Optimizely scripts and found 188 carrying malware, spread across at least 176 organizations. The list includes Zoom, GitLab, the BBC, HSBC and Salesforce — sites used by millions of people every day.

How the attack worked

The attackers exploited a flaw in how Optimizely’s Experimentation product handles sign-ups. From there, they could inject their own JavaScript into customer projects. Because the code was served from Optimizely’s own CDN, it arrived with the trust that browsers and visitors give to first-party scripts.

That trust is what made the attack so effective. Visitors to an affected site did nothing wrong — they just loaded a page, and the poisoned script loaded with it. A detailed write-up of the attack is available from StepSecurity.

What the malware did

The injected code did more than one job. Some payloads swapped crypto wallet addresses, quietly redirecting payments to wallets controlled by the attackers. Others skimmed payment card details from checkout pages. A third set tried to run exploit kits against older iOS devices.

Around $7,000 in crypto was traced to attacker wallets, though the true total may be higher — wallet swaps are hard to trace in full, and not every victim reports what happened.

How far it spread

Helme’s scan found infected scripts on at least 176 organizations, and the real count is likely higher. Optimizely’s experimentation tools sit on a huge number of high-traffic sites, so a single flaw in its sign-up flow became a supply-chain attack at web scale.

The incident is a reminder that a trusted third-party script is only as safe as its weakest link. When the script comes from the vendor itself, no amount of care on the customer’s side can stop it.

What Optimizely did next

Optimizely says it patched the flaw, removed the injected code, revoked tokens and notified affected customers. Its official site now carries the company’s response to the incident.

Site owners should still check their own Optimizely projects for unusual scripts and rotate any tokens that may have been exposed. For a related warning about hidden risks in AI systems, see Microsoft’s report on AI agent backdoors, and for how agent identities are being made verifiable, read our piece on the Linux Foundation Agent Name Service.