Unverified50% confidenceFactExact time
If malicious instructions are mixed into the knowledge base of a RAG (retrieval-augmented generation) system, the retrieval stage becomes an entry point for prompt injection.
1
Sources
50%
Confidence
Long-term
Relevance
10/22/2026
First Seen
Sources
Can AI Agents Be Socially Engineered? Understanding Prompt Injection and Tool Permission Risks
youtubeRuleArena 攻防實驗室10/22/2026
Related Entities
Related Claims
VerifiedThe bottleneck of a RAG system often lies not in retrieval itself, but in how to tell the model via the prompt how to utilize the retrieved content.78% similarVerifiedRAG (Retrieval-Augmented Generation) upgrades memory-based reasoning to retrieval-based reasoning by dynamically accessing external knowledge bases or real-time search engines during the inference stage.78% similarVerifiedRAG技术通过将问题转为向量检索最相近知识片段并拼入提示词,可大幅减少大模型幻觉78% similarUnverifiedThe core of RAG is to use vector retrieval or keyword retrieval before inference to fetch the top-k chunks from an external knowledge base and splice only these fragments into the prompt; its advantages are precision, citability and controllable cost, and its drawbacks are possible missed retrievals or semantic breaks across chunks.77% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1085887API
curl https://kongchang.com/api/v1/knowledge/claims/1085887MCP
get_claim(id=1085887)