AI Security Research Portal
claimactiveClaims

Persistent Memory Creates Poisoning And Provenance Risks

Claim

Persistent agent memory creates poisoning, provenance, lineage, and repair problems because malicious or incorrect state can influence future tasks.

Supporting Evidence

This claim is supported as a recurring pattern across the batch-ingested source catalogs. It should be refined with source-specific evidence before being treated as stable.

TitleKindDateTagsRaw
Securing LLM-Agent Long-Term Memory Against Poisoning: Non-Malleable, Origin-Bound Authority with Mapaper2026-06-23agent-memory, ai-security, formal-methods, memory-poisoning, provenanceraw
Duo Brings Identity and Authorization Across AI Agent Gatewaysofficial_blog2026-06-17agent-identity, ai-gateway, authorization, mcp, non-human-identity, security-for-airaw
AWS Security Agent adds threat modeling, Kiro power and Claude Code plugin, and moreofficial_blog2026-06-17ai-for-security, code-review, coding-agents, mcp, security-agent, strideraw
AI Agent Identity and Permission Challenges: How Uber and Auth0 Are Rethinking Access Controlnews2026-06-17agent-identity, delegated-authority, mcp, multi-agent, oauth, security-for-airaw
When Your AI Agent's Memory Becomes a Security Liabilityincident_report2026-06-11agent-memory, checkpointer, deserialization, langgraph, rce, security-for-airaw
SMSR: Certified Defence Against Runtime Memory Poisoning in Persistent LLM Agent Systemspaper2026-06-11agent-memory, ai-security, certified-defense, memory-poisoning, persistent-agentsraw
From Untrusted Input to Trusted Memory: A Systematic Study of Memory Poisoning Attacks in LLM Agentspaper2026-06-03Aditi Jain, Pritam Dash, Tanmay Shah, Tongyu Ge, Zhiwei Shang, agent-memoryraw
MemLineage: Lineage-Guided Enforcement for LLM Agent Memorypaper2026-05-14derivation-dag, memory-lineage, merkle-log, provenance, sensitive-action-gateraw
AI Security Solutions Landscape For AI and Agentic Red Teaming Q2 2026official_landscape2026-04-09OWASP Gen AI Security Project, agentic-ai, evaluation, owasp, red-teaming, security-for-airaw
AI Security Solutions Landscape for Agentic AI Q2 2026official_landscape2026-03-17OWASP Gen AI Security Project, agentic-ai, lifecycle-security, owasp, secops, security-for-airaw

Conflicting Evidence

Current Confidence

Medium. The pattern recurs across papers, standards, and news, but exact strength depends on source-specific validation.

Related