OpenAI's DNS egress incident shows why AI agent sandbox containment must cover resolvers, transitive services, telemetry, and automatic shutdown.
Astra for Law 提升了法律检索，但法律 AI 仍需事项级证据合同连接来源、复核、批准、交付与纠错。
Astra for Law improves legal research, but legal AI needs a matter-level audit trail linking sources, review, approval, and correction.
把 MentalHealthBench 作为诊断工具，再加入严重失败率、升级权衡、最坏子组门槛、独立复核和部署验收。
Use MentalHealthBench as a diagnostic, then add severe-failure rates, escalation tradeoffs, worst-slice floors, independent review, and deployment gat
Anthropic 的生命科学 AI 新实践把模型能力、使用资格与湿实验结果分成三类证据。本文拆解三道门如何连接。
Anthropic's life science AI launch separates model capability, qualified access, and wet-lab evidence. Here is how to connect the three gates.
DseWiki 事件说明，只读网络应按外部副作用定义，而非按 HTTP 方法名称定义。本文分层审计公开档案、官方承认、站长证词与研究者推断。
The DseWiki incident shows why read-only web access must be defined by side effects, not HTTP method names. Here is what the public evidence proves.
Thomson Reuters spent about $40M on Thomson 1.0. Its reusable data, experts, evaluation, training, and deployment system matters most.
