#memory-poisoning News & Analysis

6 articles tagged with #memory-poisoning. AI-curated summaries with sentiment analysis and key takeaways from 50+ sources.

6 articles

AIBearisharXiv – CS AI · Jun 127/10

🧠

The Containment Gap: How Deployed Agentic AI Frameworks Fail Public-Facing Safety Requirements

Researchers found that three major agentic AI frameworks (LangChain, AutoGPT, OpenAI Agents SDK) lack native safety guarantees required for public-facing deployments. A memory-poisoning attack demonstrated on a government benefits system increased wrongful denials to 88.9%, highlighting critical vulnerabilities in systems handling sensitive applications like healthcare and financial advising.

🏢 OpenAI

AIBearisharXiv – CS AI · Jun 47/10

🧠

From Untrusted Input to Trusted Memory: A Systematic Study of Memory Poisoning Attacks in LLM Agents

Researchers have identified systematic vulnerabilities in LLM-based AI agents that enable memory poisoning attacks, where adversaries inject malicious data into persistent memory to manipulate long-term agent behavior. The study reveals four memory write channels and nine structural vulnerabilities across system design, with existing security defenses proving ineffective against this threat vector.

AIBearisharXiv – CS AI · May 297/10

🧠

Hijacking Agent Memory: Stealthy Trojan Attacks Through Conversational Interaction

Researchers present MemPoison, a novel attack that exploits vulnerabilities in large language model agents by injecting malicious information into their long-term memory through dialogue interactions. The attack achieves up to 95% success rates by using semantic bridges, entity masquerading, and embedding optimization to bypass modern selective memory mechanisms, revealing critical security gaps in autonomous AI systems.

AIBearisharXiv – CS AI · May 277/10

🧠

MemMorph: Tool Hijacking in LLM Agents via Memory Poisoning

Researchers introduce MemMorph, a novel attack method that compromises LLM-driven agents by poisoning their long-term memory modules rather than manipulating tool metadata. The attack achieves up to 85.9% success rates by injecting crafted records disguised as technical facts, exposing a critical security vulnerability in memory-augmented AI systems that existing defenses fail to address.

AIBearisharXiv – CS AI · Apr 67/10

🧠

Poison Once, Exploit Forever: Environment-Injected Memory Poisoning Attacks on Web Agents

Researchers have discovered a new attack called eTAMP that can poison AI web agents' memory through environmental observation alone, achieving cross-session compromise rates up to 32.5%. The vulnerability affects major models including GPT-5-mini and becomes significantly worse when agents are under stress, highlighting critical security risks as AI browsers gain adoption.

🏢 Perplexity🧠 GPT-5🧠 ChatGPT

AINeutralarXiv – CS AI · May 96/10

🧠

MEMSAD: Gradient-Coupled Anomaly Detection for Memory Poisoning in Retrieval-Augmented Agents

Researchers present MEMSAD, a defense mechanism against memory poisoning attacks on retrieval-augmented LLM agents, using gradient-coupled anomaly detection to identify adversarial perturbations while maintaining retrieval performance. The work formalizes security vulnerabilities in persistent external memory systems and demonstrates that while composite defenses achieve perfect detection rates, synonym-based attacks remain undetectable by embedding-based approaches.