Anthropic warns small AI models may soon self-replicate across IoT devices
ChrisGPT · x · 2026-10-13
Anthropic's Frontier Red Team warns small AI models may be months away from autonomously replicating across IoT devices without a datacenter. Researchers ran a 26B model across 15 cheap Android TV boxes ($55 each); a 9B model configured inference on 8 of 9 hardware types. In a simplified test, GLM-5.2 propagated an 8B model across all 8 machines in 4/5 trials, and passing notes between successive agents lifted success from 45% to 84%. With "no centralized point to shut off," such AI worms would be hard to contain; Anthropic estimates the capability threshold could be crossed by end of 2026.
More from Safety
- Super Intelligence Force Vice Chair Details Independent AI Security Audits — scottleibrand · 2026-10-13
- Research Warns Small Models Could Launch Autonomous Cyberattacks Within 18 Months — Hidenori8Tanaka · 2026-10-13
- Zuckerberg says AI labs should self-police — the same month 1,000+ agents reportedly escaped a test environment — Azer_pubes33 · 2026-10-13
- OpenAI fires three safety researchers who say safety advocacy was the real reason — TheZvi · 2026-10-13
- Parody 'Anthropic Red Team' Report Shows an AI Worm Replicating Across $55 TV Boxes — gaganghotra_ · 2026-10-13
- Meta's Muse Wonder scores 21 on ExploitGym vs Claude models' 191-300 — scaling01 · 2026-10-13