Third-party dispatches, not graded receipts
Anthropic red team says some AI models may be months away from self-replication capability
A new Anthropic research report explores scenarios where small models form autonomous, self-propagating botnets that are hard to shut down. The assessment is attributed to Anthropic's Frontier Red Team.
“The latest fear-drop from an AI lab is [this report from an Anthropic team](https://www.anthropic.com/research/self-replicating-ai), working through scenarios where small AI models could form autonomous, self-propagating worm botnets that are “challenging” to shut down, and what to do to try and stop it from happening:”