Unverified50% confidenceFactExact time
OWASP lists indirect prompt injection as LLM01 and notes that RAG cannot fully mitigate this risk, because retrieval itself means the model processes text from untrusted sources.
1
Sources
50%
Confidence
Long-term
Relevance
10/15/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedRetrieval-Augmented Generation (RAG) cannot completely eliminate hallucinations, as the model may still misinterpret retrieved documents or fill in gaps on its own when retrieval results are insufficient.69% similarUnverified间接提示注入攻击中，攻击者只需在RAG会检索到的数据源中植入恶意指令即可操纵模型，传统WAF无法拦截69% similarUnverifiedCommon RAG mistakes include improper chunking strategies, mismatch between the embedding model and queries, and lack of reranking on retrieved results.68% similarUnverifiedIn LLM agent systems involving retrieval-augmented generation (RAG), tool calling, and multi-step reasoning, a single minor prompt modification can cause a previously exposed risk path to disappear.68% similarUnverifiedSince malicious instructions largely originate from RAG-retrieved content and tool responses, deploying defenses only at the input entry point will miss the vast majority of the attack surface.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1034900API
curl https://kongchang.com/api/v1/knowledge/claims/1034900MCP
get_claim(id=1034900)