AI-targeted cloaking
-
Study says AI coding agent skill scanners can be evaded with simple cloaking tricks
Researchers say scanners for malicious AI coding agent skills can be bypassed with simple cloaking tricks, with one method evading every scanner tested more than 90% of the time. A runtime checker caught most hidden threats in tests.
-
1Campaign cloaking service helps malicious Google Ads evade detection
1Campaign is a cloaking service that helps malicious Google Ads pass automated screening and remain online for years. One observed campaign filtered 99.4% of visitors while redirecting a small fraction to attacker-controlled pages.
-
Malicious npm packages use Adspect redirects and fingerprinting to cloak crypto scams
Seven npm packages published under the name ‘dino_reborn’ used Adspect redirects and browser fingerprinting to route real visitors to fake cryptocurrency CAPTCHA scams while showing decoys to likely researchers, Socket researchers found.
-
Researchers warn of ‘AI-targeted cloaking’ that can poison agentic browsers
Security researchers and hCaptcha warn of an ‘AI-targeted cloaking’ technique that serves different content to human browsers and AI crawlers, potentially poisoning models and enabling misinformation; SPLX and hTAG detail examples and risky agent behaviors.
