# Anthropic warns small AI models could self-replicate through IoT devices

https://mins.news/en/item/99763

- Ars Technica, 2026-10-14 12:55 UTC
- Tech
- Original: https://arstechnica.com/security/2026/10/anthropic-warns-small-ai-models-could-self-replicate-through-iot-devices/

> Anthropic says advances in small AI models could enable them to replicate autonomously across IoT devices as soon as next year, potentially creating computer worms that spread without human help and can identify unfamiliar devices. A recent company experiment found major hurdles to building such malware, but the findings still raise concerns; the worms could also be difficult to shut down.

## Stories

- [AI Cybersecurity and Loss of Control Risks](https://mins.news/en/topic/ai-cybersecurity-and-control) — Markdown: https://mins.news/en/topic/ai-cybersecurity-and-control.md

## Related coverage

- [Anthropic’s self-replicating AI warning is an alarming déjà vu](https://mins.news/en/item/99996) — The Verge, 2026-10-14
- [Anthropic warns AI ‘worms’ could spread across smart devices](https://mins.news/en/item/100734) — CGTN, 2026-10-14
- [Anthropic warns AI models may soon replicate themselves on small devices](https://mins.news/en/video/98299) — ABC News, 2026-10-13