2026-11-13 · Cyber Security News · LOW
TypoPrompting Turns AI Site Typos Into Malware Traps
60-WORD SECURITY BRIEF
Researchers discovered TypoPrompting attacks that redirect users typing mistyped AI tool URLs to fake verification pages hosting malware installers. The study mapped 186 lookalike domains across 27 AI brands, including Claude, ChatGPT, and Midjourney, exposing a widespread phishing surface. Defenders should enforce strict URL filtering, educate users on correct domain spelling, and deploy anti‑phishing solutions to mitigate the risk.