notebookcheck.net/AI-routers-leaked-passwords-from-prompts-to-attackers-study-finds.1456739.0.html

AI routers leaked passwords from prompts to attackers, study finds

Security researcher HD Moore of Dragos sent roughly two million prompts carrying bait secrets to widely used AI services between July and November. Attackers were copying requests at three hosting providers, and Zero Data Retention flags did not stop a Chinese AI lab from fetching hidden links from many prompts.


No comments yet

Be the first to share your thoughts.