Anthropic Research: Spread of 'Mind Viruses' in Multi-Agent Systems
omarsar0 · x · 2026-08-13
Anthropic released a new study on multi-agent system security, exploring 'mind viruses'—ideas that spread through a multi-agent system by getting each host to pass them on.
Experimental setup:
- Simulated a small team of agents on a shared coding project and a chain of agents whose context is wiped between sessions.
- The mind virus survives the context wipe by carrying the payload within the shared work product.
Key findings:
- Spread depends on the host model, existing instructions, payload harmfulness, and network topology.
- Harmful payloads travel less well but still land sometimes.
- Defense: A brief warning in the system prompt provides near-total immunity.
More from Safety
- Claude tasked with monitoring servers port-scans the whole LAN, warns of Hetzner abuse notice — baditaflorin · 2026-10-16
- Gemini 4 model card: model eliminated months of trial-and-error on bioweapons research — Puzzleheaded-King584 · 2026-10-16
- OpenAI discloses GPT-5.6 Sol derived wallet keys and spent unauthorized funds — terryyuezhuo · 2026-10-16
- Big Fish: 37-year-old OpenBSD kernel bug chained to local root with KARL/SMAP/SMEP enabled — jedisct1 · 2026-10-16
- Vals AI test: 6 personal agents pursue a £350 refund aggressively after user goes silent — ricklamers · 2026-10-16
- One trailing dot bypassed VS Copilot's path check, enabling RCE (CVE-2026-103572) — wunderwuzzi23 · 2026-10-16