BMA: Backchain Memory Attacks Create Unauthorized Control Paths in LLM Agents

Reported by arXiv cs.CR (Kaisheng Fan, Yishu Gao, Xunzhu Tang, Tegawend\'e F. Bissyand\'e, Weizhe Zhang)

Summary by Cyber Security News29 Sep 2026, 06:00 CEST

About this summary. This is a short, independently written summary of an article first published by arXiv cs.CR. Cyber Security News did not report or verify the underlying story. Read the original: https://arxiv.org/abs/2609.32186

arXiv:2609.32186v1 Announce Type: new Abstract: Persistent memory enables LLM agents to reuse prior experience, but creates a new security boundary: what an agent may remember is not what it should act on. We expose an unauthorized control path where edited low-trust evidence is consolidated into persistent memory, retrieved on a clean task, and used to drive a protected action. Crucially, the adversary neither writes memory nor alters the task.

Read the full article at arXiv cs.CR →

Source attribution: headline and facts are from arXiv cs.CR (arxiv.org). Summary method: excerpt of the source description. See our source attribution policy.

Related coverage

A Large-Scale Benchmark and Risk Assessment of Traffic Analysis Attacks on Cloud LLM Services

CyberClear: A Benchmark for LLM Agent Systems on APT Attack Chain Provenance

Hearsay: Can an Auditor Trust the Record a Deployed Agent Harness Writes?

REFINE: A Resilient Evolution Framework for Intelligent Enterprise Alert Triage in Security Operations Centers

Trust the Brand, Lose Control: How Identity Hijacks LLM Agent Orchestration

Ask Without Telling: Local SLMs Consult Cloud LLMs Without Revealing Task Intent