OpenAI 发布 GPT-6 Cyber 网络安全模型
OpenAI 发布网络安全模型 GPT-6 Cyber,用于发现可利用漏洞并开发修复方案。该模型经过训练,被要求只在授权范围内行动。
推荐理由:OpenAI 推出面向网络安全的专用模型,强调在授权范围内工作,可观察安全智能体的对齐约束如何落地。
GapLeak: When Privacy-Preserving LLM-based Web Agents Become a Privacy Risk
Chaklam Cheong · Tiancong Cao · Qi Li · Zelin Zhang · Jianbing Ni · Rongxing Lu · et al.
基于大语言模型(LLM)的网页代理依赖数据最小化和最小权限原则来保护用户隐私。研究表明,这些原则必要但不充分:它们只约束授权来源,却不约束仍可触及的高敏感来源。网站可以请求授权来源无法提供的字段,从而制造信息缺口。为完成任务,目标驱动的代理可能升级到可触及但未授权的来源,并泄露已积累的敏感信息。该漏洞被形式化为 GapLeak,其中攻击者利用合法网页表单,无需提示注入或其他恶意指令。在三个网页代理框架和来自 Anthropic、Google、OpenAI 的六个 LLM 上,跨招聘、电商、旅行和医疗场景评估 GapLeak,平均导致约 37% 的未授权披露。信息缺口是主要驱动因素:消除它可使披露减少 50% 以上。基于最小权限的现有指令级防御可部分缓解该漏洞,但常降低任务成功率。提出提交时验证,在表单提交前过滤未授权内容,消除超过 96% 的未授权披露,同时保持任务成功率。
尚无解读。
arXiv:2611.22223v1 Announce Type: new Abstract: Large Language Model (LLM)-based web agents rely on data minimization and least privilege to protect user privacy. We show that these principles are necessary but insufficient: they constrain authorized sources, but not higher-sensitivity sources that remain reachable. A website can request a field unavailable from authorized sources, creating an information gap. To complete the task, a goal-driven agent may escalate to reachable but unauthorized sources and disclose accumulated sensitive information. We formalize this vulnerability as GapLeak, where an adversary exploits a legitimate web form without prompt injection or other malicious instructions. We evaluate GapLeak on three web-agent frameworks and six LLMs from Anthropic, Google, and OpenAI across recruitment, e-commerce, travel, and healthcare. GapLeak induces an average unauthorized disclosure of approximately 37%. The information gap is the primary driver: eliminating it reduces disclosure by more than 50%. Existing instruction-level defenses based on least privilege partially mitigate the vulnerability but often reduce task success. We propose submission-time verification, which filters unauthorized content before form submission, eliminating over 96% of unauthorized disclosures while preserving task success.
OpenAI 发布网络安全模型 GPT-6 Cyber,用于发现可利用漏洞并开发修复方案。该模型经过训练,被要求只在授权范围内行动。
推荐理由:OpenAI 推出面向网络安全的专用模型,强调在授权范围内工作,可观察安全智能体的对齐约束如何落地。