Ian Goodfellow· @goodfellow_ian · X·· 2 小时前AI 评分54
AI 导读
Ian Goodfellow 转发 AI 安全研究实验室 AISLE 的年度总结：该团队一年内在 100+ 代码库中发现 500+ 个带 CVE 的零日漏洞，约占 2026 年全球开源 CVE 的 1%，在 curl 等加固代码库上的零日发现多次追平或超过前沿 AI 实验室。作者评价称，Aisle 首年证明了围绕 LLM 搭建的系统非常重要，即使是小而便宜、速度快的模型也能被增强去完成有价值的安全工作。
正文
In Aisle's first year, they've shown that the system built around an LLM is highly important, and even small, cheap fast models can be boosted to do valuable security work
My AI security research lab AISLE discovered 500+ zero-days with CVEs across 100+ codebases in a year => ~1% of 2026's open-source CVEs globally from a single team! On hardened codebases like curl, we've repeatedly matched or outperformed frontier AI labs in zero-day discovery在 X 查看被引用的帖子
来源：Ian Goodfellow · x.com