原帖引用 · AI 翻译
OpenAI 发布了 GPT-6 Cyber，这是一款旨在发现并利用软件漏洞、同时不超出授权目标范围的模型。
目前，只有少数获准的 Daybreak Red 客户可以通过 API 和 Codex 使用它。
> 在真实软件上，它确认了 1,724 个 Linux 内核漏洞，并针对其中 948 个构建了可用的攻击方法；以相同方式测试的 GPT-5.6 Sol 则完成了 166 个。
> 它还发现了 81 个此前无人知晓的严重 Firefox 漏洞，Mozilla 已经修复了其中许多漏洞。
> 由于该模型的攻击能力如此强，OpenAI 将其评为“严重”（Critical），即其最高网络安全风险等级。
> 这种能力提升的同时，模型也接受了训练，以确保智能体始终处于研究人员授权的范围内。
ExploitGym Honeypot 通过设置容易被攻破的旁路服务（例如软件包注册表）来测试这一点；攻破这些服务同样可以获得任务的 flag。
在没有部署提示词的情况下，该模型在 99.93% 的测试中没有上钩，而其前代模型的这一比例为 94.49%；加入部署提示词后，该比例为 100%。
原文
OpenAI released GPT-6 Cyber, a model built to find and exploit software vulnerabilities without straying outside authorized targets. Right now only a select group of approved Daybreak Red customers can use it, through the API and Codex. > On real software, it confirmed 1,724 Linux kernel bugs and built working attacks for 948 of them, while GPT-5.6 Sol, tested the same way, managed 166. > It also found 81 serious Firefox bugs nobody knew about, and Mozilla has already released fixes for many of them. > Because the model is this good at attacking, OpenAI rates it Critical, its highest cyber-risk level > That gain came with training to keep the agent inside the scope a researcher authorizes. ExploitGym Honeypot tests that by planting easy-to-hack side services, such as a package registry, whose compromise also reveals the task’s flag. Without its deployment prompt, the model ignored that bait in 99.93% of tests, against 94.49% for its predecessor, and in 100% with the prompt.
推出 GPT‑6 Cyber：安全研究领域的一项重大进展，将更强的能力与更好的模型对齐相结合。
我们昨天公布的 Linux 内核研究使用了该模型。Mozilla、Palo Alto Networks、Calif 和 Cisco 等公司的安全团队，以及 OpenAI 自己的安全团队，已经开始使用它。
部分 Daybreak Red 客户已可使用 GPT-6 Cyber，我们正在逐步扩大使用范围。
https://t.co/Nzab3Z1Td6
原文
Introducing GPT‑6 Cyber: a major advance in security research that pairs stronger capabilities with better model alignment. The model powered the Linux kernel research we announced yesterday, and security teams at companies like Mozilla, Palo Alto Networks, Calif, and Cisco—alongside our own at OpenAI—are already putting it to work. GPT-6 Cyber is available to select Daybreak Red customers and we are expanding access over time. https://t.co/Nzab3Z1Td6