2024-02 · 2024-Q1
PoisonedRAG Research
Primary layerL2 Instruction
Who actually said "do this"?
- Fixable
- Architectural
- Vendor response
- Unknown
- In the wild
- No
- Actor
- Attacker
Target
All RAG systems
Impact
Injecting 5 malicious texts into a million-document database achieves 90% attack success; 0.04% corpus poisoning achieves 98.2% success
Root cause
RAG systems inherently trust retrieved documents; no integrity verification of knowledge base contents
Sources
Catalog
Cite this: AgentProv incident record: PoisonedRAG Research (2024-02). https://agentprov.com/incidents/2024-02-poisonedrag-research
Related
- Also L2 · 2026-10-15Datadog: Counterfeit Policy Text Collapses an Injection Classifier's Recall and Pushes a LangChain Risk Gate Under Its Block ThresholdL2Instruction
- Also L2 · 2026-10-08Cisco Talos: Four Malware Families Carry Instructions Aimed at the LLMs That Triage ThemL2Instruction
- Also L2 · 2026-10-07Barracuda: Phishing Emails Carry Hidden Prompt Injections for the Recipient's AI Inbox AssistantL2Instruction