🚩🚩🚩 Anthropic says UNKILLABLE models may be MONTHS away from self-replicating across the internet

(and this is SMALL models - imagine what large models will be capable of)

- "We're seeing signs of life for AI worms."

- "DIGITAL SLIME MOLD": "We expect a future awash with agents running on all types of hardware, including some that aren’t under human control."

- "It’s uncertain exactly what form autonomous AI replication will take. It could be a slow, distributed “digital slime mold,” living on relatively insecure compute. It could also be a powerful cyber model that breaches datacenters at a high pace, or a model that scams people for compute credits."

"The capabilities that make AI models generally powerful coding agents also generalize to cyber tasks, including autonomous replication. We don’t yet have a full rogue replicator, but agents are becoming increasingly capable of completing the individual steps in the replication chain (and passing on their adaptations to future copies of themselves).

The rise of slow replicators would also be a bellwether for more aggressive, faster ones that can breach more hardened machines or be effective API key thieves. We should aim to harden our systems to make it difficult for such replicators to spread, but also prepare to deal with a world where they are here to stay.'"