Anthropic warned in October 2026 that an AI model self-replicating across IoT devices would be hard to shut down because it would have no central point to switch off

Overview Debates Related claims Topics Sources

From the postAI models may be capable of self-replicating without access to a datacenter

AI models may be capable of self-replicating without access to a datacenter

Anthropic researchers show small open-weight AI models can set up inference on unfamiliar IoT hardware and, in simplified evals, GLM-5.2 can replicate end to end. They project small-model worms by end of 2026 and suggest defenses.