Models71°

OpenAI 发布网络安全能力最强的模型,Vercel DeepsecBench 验证

Why we picked it

Vercel 的 DeepsecBench 用从未公开的代码库做评测,任何实验室都没法提前训练刷分,OpenAI 模型拿下第一就更有说服力。做安全相关开发或选型模型的工程师可以直接去 AI Gateway 看榜单,比那些人尽皆知的基准更真。

OpenAI 发布了一组在网络安全任务上表现最强的模型,这一成绩由 Vercel 推出的 DeepsecBench 基准测试验证。DeepsecBench 的关键设计在于其基准代码库此前从未公开,任何模型实验室都没有机会用这些代码库做训练,属于"隐身"(stealth)评测。

Body

由于所有实验室都无法提前接触评测数据,DeepsecBench 的排名较难被针对性刷分,OpenAI 排名第一的结果因此更具参考价值。Vercel 已将该基准通过其 AI Gateway(vercel.com/ai-gateway)对外开放,开发者可以自行查看各家模型在网络安全类任务上的得分。原文作者 rauchg(Vercel CEO)认为这个时间点释放了一个值得注意的信号。

Original · Guillermo Rauch

OpenAI just put out the most cybersecurity-capable models in the world. Excellent timing. Important signal: @vercel DeepsecBench is a stealth cyber benchmark. No lab has had access to the codebases to train on. vercel.com/ai-gateway/lea…📊 17 ⚡

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Models" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to US-China foundation model benchmarks and commercialization comparison and AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include OpenAI, Vercel, DeepsecBench, cybersecurity.

Source attribution: curated from Guillermo Rauch (@rauchg).