GPT-6.2 Astra 系统卡披露模型权重安全协议与前沿训练隔离机制
OpenAI 把权重保管、虚拟机隔离、离线 RL 训练这套流程写得非常细,像是给同行看的防泄露手册。做智能体安全或部署前沿模型的团队值得对照参考。
OpenAI 在 GPT-6.2 Astra 系统卡中公布了针对未发布模型权重的安全协议。权重在存储和传输时均加密,访问需通过多因素认证、多方审批和详细日志记录,并配有专门的检测机制识别未经授权的权重转移尝试,配合 Detection and Response、Threat Intelligence 和 Insider Risk 三个安全项目运作。
自 GPT-6 Astra 发布以来,OpenAI 进一步强化了前沿训练和评估的隔离措施。Linux 代码执行和 computer-use 任务运行在轻量级虚拟机内的受限容器中,与宿主机隔离;网络和云控制限制出站流量以及系统间未经授权的移动。这些控制设计目标是在智能体逃逸沙箱并控制宿主机的情况下仍能维持隔离。
所有前沿强化学习运行使用缓存的网页内容构成的离线环境,替代真实互联网访问,让模型可以练习浏览和搜索但不作用于外部系统。这些限制适用于研究环境,已发布模型则采用产品层面的安全防护措施。
Interesting section from the GPT-6.2 Astra system card on OpenAI's security protocols for unreleased model weights and frontier training runs. I believe this is the most detailed look we've gotten yet at weight security.
'We encrypt unreleased model weights when stored and transferred and limit access to authorized personnel, with controls that include multi-factor authentication, multi-party approval, and detailed logging. We also use dedicated detections to identify possible attempts to remove model weights without authorization. Our Detection and Response, Threat Intelligence, and Insider Risk programs help identify, investigate, and respond to threats.
Since the release of GPT-6 Astra, we have further hardened containment for frontier training and evaluation. For Linux code execution and computer-use tasks, restricted containers run inside lightweight virtual machines that isolate them from the host. Network and cloud controls restrict outbound traffic and unauthorized movement between systems, including through supporting services. These controls are designed to maintain containment even if an agent escapes its sandbox and controls the host.
All frontier reinforcement learning runs use an "offline world" of cached web content in place of live internet access. This lets models practice browsing and searching without acting on external systems. These restrictions apply to our research environments; released models operate with the product safeguards described elsewhere in this card.'
本条情报归属 AITOP「模型」核心栏目,重点聚焦北美前沿实验室与硅谷科技巨头。在跨区域产业观察坐标中,该动态直接关联「各国 AI 监管法案、合规治理与主权算力政策」「中美大模型技术代差与商业化对照」研究主线,反映出当前技术演进与商业落地的最新风向。涉及核心实体与技术议题包括:GPT-6.2 Astra、OpenAI、AI安全、权重保护,读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。
信源核验说明:本文原始线索采编自「Andrew Curran」(@AndrewCurran_),经 AITOP 多维评分管线结构化提炼。截至目前,全网已有 11 家独立信源对该事件进行交叉印证与跟进报道(含 OpenAI、Vaibhav Srivastav、rohanpaul_ai 等),属于近期高关注度行业信号。