Typos in AI tool web addresses lead users to malware and browser hijacks
MALWAREChatGPTredteam/daily · 1 source
- Severity
- High
- Sources
- 1
What happened
Zaytsev traced 186 typosquatted domains across 27 AI brands, which he calls TypoPrompting, with ChatGPT (25), Claude (18) and DeepSeek (16) most targeted. The domains deliver ClickFix lures, browser-hijacking malware and search-hijacking extensions. The article does not name specific malware families or CVEs. Defenders should monitor for lookalike domains of AI services in DNS and proxy logs, and consider blocking or flagging newly registered look-alikes. Users should be warned about ClickFix-style social engineering.
For security analysts · Written by AI from 1 source · Wrong?
What to do
- Check the full web address carefully before entering details on any AI tool site.
- Bookmark official AI service addresses rather than typing them from memory.
- Be wary of any page that asks you to run commands or install extensions to fix a problem.
Written by AI from the sources, checked against them
Full story timeline · 1 update
Who reported · 1 outlet
GBHackers