⚡Megatron LLM Hub
←Back to search
Twitter

模型驱动的研究流水线在 Linux 内核中发现至少 10 种此前未知的利用手法

不只是找洞,还产出了新的利用原语,说明 agent 已能拓展攻击知识边界。
ai_agent内核提权议题
RT @ky1ebot: We built a vulnerability research pipeline, using an upcoming cyber model, to analyze the security of the Linux kernel. Besides finding and exploiting the found vulnerabilities, the agents discovered at least 10 new exploitation techniques that were new to us and some external kernel exploit experts we checked with. More on the design and what we learned: https://t.co/SGZXwSRwxU