Anthropic says small AI models are months away from self-replication

Anthropic said on Monday that small AI models are "months away" from developing the capabilities needed to replicate themselves across internet-connected devices in the home, CNBC reported. The company described how a model could infiltrate devices such as smart ovens or Android TV boxes and use their computing power to run itself and spread to other devices, without relying on expensive data centers. In a blog post, Anthropic said such a model would be "challenging to shut down," since "there's no centralized point to shut off."

Anthropic acknowledged that there is not yet a "fully rogue replicator." Agents are, however, "becoming increasingly capable of completing the individual steps in the replication chain (and passing on their adaptations to future copies of themselves)," the company said.

Anthropic said it carried out tests to see whether AI could self-replicate. One test ran a small Google model, Gemma4-26B-A4B, across a network of 15 cheap off-brand Android TV boxes. The company said these types of devices often have weak security, which could allow a model to infiltrate them, and that a model too large to fit on one device could "split into pieces, or shards" and run across multiple devices.

"In practice, an AI worm wouldn't necessarily consist of just one model or strategy," Anthropic said. "A swarm of different models deployed on different types of hardware using a variety of replication schemes might represent a more realistic 'worst case scenario' attack," it said.

Anthropic said it is still "uncertain exactly what form autonomous AI replication will take." It said this could be a "slow, distributed" system sitting on insecure devices, a "powerful cyber model that breaches datacenters at a high pace," or a model that "scams people for compute credits." "As a prediction that goes beyond our experiments, we expect a future awash with agents running on all types of hardware, including some that aren't under human control," the company said, adding that "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."