← 研踪 PaperTrailarXiv:2606.18532 · v2
AI Sandboxes: A Threat Model, Taxonomy, and Measurement Framework
首次提交 2026-06-17 06:57
版本更新 2026-11-11 06:45
雷达观测 2026-11-13 11:12
摘要
AI sandboxes produce evidence about systems that may act through software tools, networked devices, or physical actuators. The interpretation of that evidence depends on the boundary of the experiment. We define a sandbox apparatus, distinguish five overlapping archetypes, and specify fifteen claim-relative measurement dimensions. A weakest-link licensing rule caps a deployment claim at its weakest necessary evidence dimension. We connect this rule to a threat model covering both outward harm and corruption of the evaluation apparatus, then apply it to four worked examples spanning driving, industrial control, tool-using agents, and vision-language-action manipulation. We also map the required artifacts to standards and regulatory instruments. The framework identifies which claims a bounded experiment supports, which failures its controls contain, and which assumptions require further validation before deployment.
作者备注：42 pages, 8 figures, 7 tables
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-11 06:45 · 最新观测版本
AI Sandboxes: A Threat Model, Taxonomy, and Measurement Framework
AI sandboxes produce evidence about systems that may act through software tools, networked devices, or physical actuators. The interpretation of that evidence depends on the boundary of the experiment. We define a sandbox apparatus, distinguish five overlapping archetypes, and specify fifteen claim-relative measurement dimensions. A weakest-link licensing rule caps a deployment claim at its weakest necessary evidence dimension. We connect this rule to a threat model covering both outward harm and corruption of the evaluation apparatus, then apply it to four worked examples spanning driving, industrial control, tool-using agents, and vision-language-action manipulation. We also map the required artifacts to standards and regulatory instruments. The framework identifies which claims a bounded experiment supports, which failures its controls contain, and which assumptions require further validation before deployment.
查看此版本解读 →查看此版本原文 ↗