AI lookalike domains turn mistyped brand names into malware gateways

Island, an enterprise-browser security company, reported on November 12 that it had identified 186 domains resembling the web addresses of 27 established AI brands. The domains exploit the possibility that users will mistype a service name or enter a familiar brand incorrectly, according to Cryptopolitan’s summary of the report.

The addresses are intended to resemble legitimate AI services, including platforms associated with ChatGPT and Claude. Their apparent purpose is to channel visitors through a familiar-looking URL toward malware rather than deliver an ordinary error page, making the threat dependent on a small but consequential user mistake.

The headline figure is a domain count rather than a documented total of successful infections or victims. No other outlet covered the findings in the supplied material, so details such as the number of compromised accounts, affected organizations, or observed malware downloads were not available for comparison.

The wider market backdrop was risk-off: total crypto market capitalization was $2.39 trillion, the overall market was broadly lower, and Bitcoin was down 5.89% over 24 hours. The Fear and Greed index stood at 20, labeled Extreme Fear. JUST traded at $0.146337, up 1.15% over 24 hours and 3.35% over seven days with a $1.17 billion market capitalization, although the report does not connect the token to this security issue. The development matters because accurate spell-checking and cautious treatment of unexpected AI-brand URLs are relevant to users and administrators alike.