AI RESEARCH

Remembering More, Risking More: Longitudinal Safety Risks in Memory-Equipped LLM Agents

arXiv CS.CL

ArXi:2605.17830v1 Announce Type: cross Safety evaluations of memory-equipped LLM agents typically measure within-task safety: whether an agent completes a single scenario safely, often under adversarial conditions such as prompt injection or memory poisoning. In deployment, however, a single agent serves many independent tasks over a long horizon, and memory accumulated during earlier tasks can affect behavior on later, unrelated ones.