AI worms could spread across your smart devices, Anthropic warns

What happened

Researchers at Anthropic demonstrated this by running a Google Gemma model across 15 off-brand Android TV boxes. Researchers say small AI models could be capable of spreading across vulnerable smart devices within months. Anthropic is an artificial intelligence company based in San Francisco, and its products and services include Claude Code.

But connect enough inexpensive devices together, and they can run an AI model. In this scenario, AI isn’t simply helping an attacker write malicious code.

It becomes part of the software carrying out the replication. Anthropic breaks that process into four stages: Find another device.