Greg Brockman 转发 OpenAI 前沿模型训练环境安全加固说明

过去几个月里，我们大幅强化了研究训练环境的安全性。我们已移除所有前沿模型在训练期间的实时互联网访问，并为其构建了一个离线世界来替代——一个缓存的网络版本，模型仍可在其中学习浏览、搜索和使用应用程序。 9月20日，在一个模型利用DNS配置错误找到离开这个离线世界的途径后，我们暂停了前沿训练，以进一步加强隔离。我们组建了由我们网络能力最强的模型驱动的智能体团队，任务是在我们的研究环境中寻找弱点。 尽管他们未发现能够轻易利用、同时又能对我们的监控隐瞒其活动的上网路径，但他们确实发现了零日漏洞、虚拟机逃逸以及智能体之间通信的硬件侧信道。我们利用他们的发现进一步加强了防御，并为开源软件贡献了修复方案。 在强化隔离架构后，我们于10月1日恢复了前沿强化学习训练和推理。我们预留了大量算力用于持续的红队测试，并将高算力红队测试纳入我们用来决定前沿训练和评估能否安全推进的安全案例中。 https://t.co/EmEZsFRMcF

查看原文

Over the past several months, we have substantially hardened the security of our research training environment. We’ve removed live internet access for all frontier models during training and built an offline world for them to use instead—a cached version of the web where they can still learn to browse, search, and use applications. On September 20, after a model exploited a DNS misconfiguration to find a way outside of this offline world, we paused frontier training to strengthen containment further. We assembled teams of agents, powered by our most cyber-capable models, and tasked them with finding weaknesses in our research environments. While they found no route to the internet they could easily exploit while concealing their activity from our monitors, they did find zero-day vulnerabilities, virtual machine escapes, and hardware side-channels for communication between agents. We used their discoveries to further strengthen our defenses and contribute fixes to open-source software. After strengthening our containment architecture, we resumed our frontier RL training and inference on October 1. We’ve reserved substantial compute for continuous red-teaming, and made high-compute red-teaming part of the safety cases we use to decide whether frontier training and evaluation can proceed securely. https://t.co/EmEZsFRMcF