Researchers Develop Self-Replicating AI Worm Using Local Models
Researchers from the University of Toronto have developed a proof-of-concept AI-driven computer worm that operates using a locally hosted open-weight large language model. This worm is capable of autonomously navigating networks, generating tailored attack strategies, and replicating itself without human intervention. The research, led by associate professor Nicolas Papernot, involved testing the worm on a deliberately vulnerable network, where it successfully identified and exploited numerous vulnerabilities. Unlike traditional worms, this AI worm does not rely on pre-encoded exploit chains or commercial AI services, making it a novel approach in cybersecurity threats. 
