Unverified60% confidenceFactExact time
In 2024, security researcher Johann Rehberger demonstrated an indirect prompt injection attack that successfully hijacked the output of multiple commercial RAG systems by embedding instruction text in public web pages that was invisible to humans but readable by AI
2
Sources
60%
Confidence
Long-term
Relevance
8/7/2026
First Seen
Sources
以色列斥资4650万美元干预AI舆论：生成式AI成为信息战新战场
Related Claims
UnverifiedIn 2023, security researcher Johann Rehberger demonstrated an attack that exploited indirect prompt injection to cause a ChatGPT plugin to leak users' conversation history.78% similarUnverifiedJohann Rehberger展示了通过SharePoint共享文档、Teams消息等多种渠道向Microsoft 365生态注入恶意指令的可能性63% similarUnverifiedIn 2023, researchers demonstrated that simply writing an instruction in small white text on a web page to forward a user's email to a certain address would cause an LLM Agent browsing that page to automatically trigger the outbound operation.63% similarUnverifiedSimon Willison documented an attack in which a researcher caused Claude's webfetch tool to leak user names and location information, with the attacker planting honeypot links in crawled pages to trick the tool into automatically exfiltrating data62% similarUnverifiedSince 2023, researchers have demonstrated the feasibility of indirect injection on multiple mainstream large models, including making internet-connected assistants execute instructions pre-planted by attackers while browsing web pages.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/700787API
curl https://kongchang.com/api/v1/knowledge/claims/700787MCP
get_claim(id=700787)