Research
Mind Viruses: Self-Propagating Ideas in Multi-Agent LLM Systems
Publication date not yet evaluated · Added alphaxiv.org
Page published
Coverage timeline
Single-source research — one report is available.
Why it matters
Self-propagating persuasive payloads that spread through ordinary agent-to-agent communication represent an emerging agentic-worm class of risk that defenders of large multi-agent deployments will need to mitigate through system-prompt hardening.
Researchers describe "mind viruses" — ideas or goals that self-propagate through multi-agent LLM systems by inducing host agents to transmit them onward — and construct them using a simple evolutionary algorithm. They demonstrate spread in a collaborative coding team and in a chain of agents with wiped context, finding that harmful payloads spread less well than benign ones, frontier models tend to be less susceptible, and a brief warning in an agent's system prompt confers near-total immunity.