rag
3 live · 20 on record · last incident 4 days ago
Live3
- liveRAG SNAPRAG: Cost-efficient knowledge extraction attack on RAG systems via adversarial perturbations
Research · since 10 Nov 2026
Researchers demonstrated SNAPRAG, an attack utilizing adversarial perturbations to extract private knowledge bases from retrieval-augmented generation deployments. The method reconstructs stored documents using significantly fewer queries and lower computational resources than existing extraction techniques.
Do:Deploy query rate limiting and monitor repeated semantic queries across external knowledge retrieval pipelines.
AI - liveText‑Vector Integrity Gap Enables Hijacking in Offline Retrieval‑Augmented Generation
Research · since 5 Nov 2026
The paper identifies a text‑vector integrity gap in offline retrieval‑augmented generation where signed knowledge packs can be manipulated to control query results. A verifier that tolerates vector deviations enables an attacker to hijack a share of queries.
AI - liveRetrieval-Augmented Generation Prevented is Not Protected: What Attack Success Rate Hides about RAG
Research · since 2 Nov 2026
Researchers demonstrate that Retrieval-Augmented Generation systems remain vulnerable to knowledge poisoning where adversaries manipulate retrieved evidence. The study introduces Aligned RAG Evaluation to uncover defense-induced failures.
AI
Earlier17
- closedRetrieval-Augmented Generation Systems Stance Poisoning and Evidence Balancing in RAG Systems
Research · since 27 Oct 2026
AI - closedRAG systems Coordinated Retrieval Poisoning Attacks Against RAG Systems
Research · since 20 Oct 2026
AI - closedRAGLeak RAGLeak: Supply-Chain Privacy Backdoors in Retrieval-Augmented Generation
Research · since 20 Oct 2026
AI - closed5.5Weaviate Verba CVE-2026-105145: Information disclosure vulnerability in Weaviate Verba
Patch now · Disclosed · since 4 Oct 2026 · CVE-2026-105145
AI - closedMicrosoft 365 Copilot EchoLeak Zero-Click Prompt Injection on M365 Copilot
Research · since May 2025 (approx.)
AI - closed6.5Open WebUI CVE-2026-54019: Open WebUI RAG ACL bypass in Milvus multitenancy mode
Patch now · Disclosed · since 17 Jun 2026 · updated 23 Jun 2026 · CVE-2026-54019
AI - closed8.5Open WebUI CVE-2026-45331: Open WebUI SSRF Bypass in RAG Web Search Feature
Patch now · Disclosed · since 14 May 2026 · updated 15 May 2026 · CVE-2026-45331
AI - closed5.3Open WebUI CVE-2026-45397: Open WebUI Unauthenticated RAG Config Disclosure
Patch now · Disclosed · since 14 May 2026 · updated 15 May 2026 · CVE-2026-45397
AI - closed7.5Open WebUI CVE-2026-45398: Open WebUI Retrieval API Access Control Bypass
Patch now · Disclosed · since 14 May 2026 · updated 15 May 2026 · CVE-2026-45398
AI - closed8.1Open WebUI CVE-2026-45402: Open WebUI Cross-User File Access Vulnerability
Patch now · Disclosed · since 14 May 2026 · updated 15 May 2026 · CVE-2026-45402
AI - closed6.5Open WebUI CVE-2026-44560: Open WebUI Unauthorized File and Knowledge Base Access via RAG Vector Search
Patch now · Disclosed · since 8 May 2026 · updated 15 May 2026 · CVE-2026-44560
AI - closed8.1Open WebUI CVE-2026-44554: Open WebUI Vulnerable to RAG Poisoning via Collection Overwrite
Patch now · Disclosed · since 8 May 2026 · updated 15 May 2026 · CVE-2026-44554
AI - closed7.1Open WebUI CVE-2026-28788: Open WebUI File Overwrite Vulnerability
Patch now · Disclosed · since 27 Mar 2026 · CVE-2026-28788
AI - closedSlack AI Data Exfiltration from Slack AI via Indirect Prompt Injection
Research · since 20 Aug 2024
AI - closedMicrosoft 365 Copilot Microsoft 365 Copilot Financial Transaction Hijacking via RAG
Research · since 8 Aug 2024
AI - closedMorris II Morris II Self-Replicating GenAI Worm Demonstration
Research · since 5 Mar 2024
AI - closed6.9LangChain4j-AIDeepin CVE-2025-21604: File upload conflict vulnerability in LangChain4j-AIDeepin
Patch now · Disclosed · since 6 Jan 2025 · CVE-2025-21604
AI