Researchers are building a self-replicating AI worm with BYO LLM

A team of researchers at the University of Toronto in Canada has put together a self-replicating malware – a worm – that is capable of worming its way through networks and developing new attack strategies for each machine it encounters, rather than relying on fixed, specific exploits.

Additionally, the CleverHans Lab team, led by Associate Professor Nicolas Papernot, used a small, free Large Language Model (LLM) on the worm, showing that it does not require extensive commercial infrastructure to power it.

The worm carries a copy of a single open-weight graphics processing unit (GPU) LLM that the…