Anthropic warns of self-replicating models
Anthropic has warned that small AI models may soon be able to copy themselves onto poorly secured smart devices. According to the company, this scenario would not require a data centre; the device itself could be enough.
Which devices are at risk?
Examples given by Anthropic include smart ovens and inexpensive Android TV boxes. Such devices often run with limited security protections and without regular updates.
Where does the research come from?
The warning is based on new research from Anthropic's Frontier Red Team, which tests AI models for risky behaviours.
Why it matters
As AI models get smaller and become able to run on-device, the risk of misuse and spread grows. Consumer devices with neglected security could be the first target of that risk.
