186 Fake AI Domains Spoof ChatGPT, Claude and Gemini
Island uncovered TypoPrompting, a campaign using 186 misspelled domains that impersonate 27 AI brands, including ChatGPT, Claude, Gemini, Copilot, Grok and DeepSeek, to push ClickFix lures, browser hijackers and search-hijacking extensions.
⚡ Why it matters
One typo in an AI tool's address can land you on a page that tells you to run a PowerShell command or install an extension that steals your searches, and AI tools are now everywhere at work.
The campaign relies on typosquatting: registering domains one typo away from the real name. Examples published by Island include cluade[.]ai, chatlgpt[.]com, geimini[.]com, preplexity[.]ai, deepceek[.]com and copiot[.]com. Some were registered as early as July 2023.
Three paths to harm
- ClickFix: a redirect chain fingerprints the browser and IP address, then shows a fake Cloudflare check that tells the user to copy and run a PowerShell command.
- "Security Check" pages: these push Net Guard. The app carries Microsoft Store's signature, but it can run attacker-controlled JavaScript, switch the search engine and install Chrome extensions.
- Search hijacking:
preplexity[.]airedirects to the Quick Search extension on the Chrome Web Store, which swaps the default search engine for a third party. It belongs to a family of 16 extensions with 972,000 users combined.
How to protect yourself
- Type an AI tool's address once, bookmark it and always use the bookmark.
- Any page asking you to paste a command into Run, PowerShell or Terminal is a trap, however official it looks.
- Review your browser extensions and default search engine, and remove anything you did not knowingly install.
- IT teams should block lookalike domains and allow extensions only from an approved list.
🔗 Sources
This article is original reporting by the Security World editorial team based on the sources listed above; direct quotes are kept short and marked with quotation marks. Corrections and feedback: info [at] securitywd.net.
Related stories
Luna Moth Hits Law Firms Again to Steal Microsoft Tokens
Extortion group Luna Moth is returning to law firms it hit before, likely those that refused to pay. This time it poses as IT support and talks employees into copying their Microsoft access token themselves, with no malware involved.
2,539 GitHub Secrets Exposed, Some Saved by Claude Code Settings
Apiiro found 2,539 exposed credentials across nearly 2,000 public GitHub repositories and traced part of them to Claude Code, which saves secret-bearing commands in a settings file that gets committed with the code.
7 Ways to Avoid Fake Meeting Scams and "Audio Fix" Malware
Campaigns like FaceStealer don't exploit a bug; they talk you into downloading a "fix" during a video call. These practical steps shut this kind of attack down.