Researchers: when given 15 CVE descriptions, GPT-4 autonomously exploited 87% of the vulnerabilities, compared to 0% for every other model tested (Thomas Claburn/The Register)

Thomas Claburn / The Register:
Researchers: when given 15 CVE descriptions, GPT-4 autonomously exploited 87% of the vulnerabilities, compared to 0% for every other model tested  —  While some other LLMs appear to flat-out suck  —  AI agents, which combine large language models with automation software …

source http://www.techmeme.com/240420/p11#a240420p11