
University of Toronto researchers built a proof-of-concept AI worm that uses a locally hosted open-weight LLM to generate attack strategies and replicate across a network without human input. In 15 runs on a 33-host network, it identified 31.3 vulnerabilities on average and replicated to 62% of hosts.
Tap to vote and see what everyone thinks.
AI finds 10,000 flaws as China copies models
Summary by ByteBrief