According to CoinMeta, researchers, in collaboration with the University of Toronto, Vector Institute, the University of Cambridge, and ServiceNow, have created a computer worm driven by large-scale open-source models. This worm is capable of scanning networks, identifying vulnerabilities in target machines, generating attack strategies, and replicating itself after gaining control of a device to continue attacking the next machine. The team conducted 15 rounds of experiments in an isolated network consisting of 33 virtual machines, with each round lasting for 7 days. On average, the worm obtained administrative privileges on 23.1 machines and was able to launch new copies on 20.4 machines, with the longest continuous transmission spanning 7 generations. Compared to traditional worms that rely on preset attack scripts, this system can adjust its attack methods based on the specific conditions of each machine. Nevertheless, this remains a laboratory proof of concept; every machine in the test network had pre-installed vulnerabilities, there was no antivirus software or active defense measures, and neither the model nor the complete code has been made public.