رفتن به محتوای اصلی
بازگشت به هشدارها
بحرانی

AI Recommendation Poisoning: How "Ask AI" Buttons Silently Alter LLM Memory

۱۴۰۵/۰۵/۱۸
بحرانی
سطح هشدار
بحرانی

شرح هشدار

A new class of prompt injection is spreading across commercial websites. It requires no malware, no stolen credentials, and no zero-day exploit. It abuses a standard feature built into almost every major AI assistant: pre-filled deep links. We observed production websites embedding hidden prompt injection payloads inside "Ask AI" buttons on marketing and competitor comparison pages. When a user