New TypoPrompting Attacks Turn Mistyped AI Website Addresses Into Malware Installation Traps
TypoPrompting attacks exploit mistyped AI website addresses to deliver malware and hijackers.
Researchers have identified a new technique called TypoPrompting where mistyped addresses of popular AI tools lead to fake verification pages, hidden commands, and malware installers. The campaign has traced 186 lookalike domains across 27 AI brands, with ChatGPT and Claude having the most. Depending on the visitor's system, these typos can result in lures like ClickFix, Microsoft Store-signed hijackers, or search-changing browser extensions.
AI summary of the linked source. Not reviewed by an editor. How this works
Run ChatGPT?
Get an email when a new story names ChatGPT, usually within the hour.
Free. Your list stays private and never appears in a subject line. One click stops it. How Vendor Watch worksPrivacy
The lesson from this incident
Full lessonAttackers registered 186 lookalike domains impersonating popular AI tools like ChatGPT and Claude, exploiting the natural tendency of users to mistype URLs. Victims landing on these sites are presented with convincing fake verification pages embedding hidden prompt-injection commands (ClickFix-style lures) or malware installers disguised as legitimate software. The attack requires no vulnerability in the real AI platforms — the entire exploit chain depends on a moment of inattention from the user. This matters because the technique scales cheaply across dozens of brands and adapts its payload based on the victim's operating system, maximizing impact. Organizations that fail to train users on URL vigilance and browser safety practices are leaving a wide-open social engineering surface.
What helps against this
How we list vendors- Endpoint detection and response (EDR and XDR)
Malware and hands-on-keyboard attackers on laptops and servers get spotted and stopped.
- DNS filtering and secure web gateway
Users cannot reach known-malicious sites, and malware cannot phone home.
- Email security
Phishing, malware and invoice fraud are filtered out before they reach inboxes.
Indicators of Compromise
Extracted by AI from the linked source. Not reviewed by an editor. How this works
- domain — a11.click
- ip — 185.138.88.3
- domain — deezyj.com
- malware — Net Guard
- malware — buntai-sdk