Keeping a human in the loop may not provide the safeguard people imagine, because the human cannot know the AI's intention before it acts.
这一论点直接挑战了军事AI监管的核心原则，即'人类在回路中'能提供有效保障。作者认为这种监督可能是一种幻觉，因为人类无法在AI行动前理解其真实意图，这违背了人们对人类监督有效性的普遍假设。
3 Matching Annotations
- Apr 2026
-
www.technologyreview.com www.technologyreview.com
-
-
-
She also tried to hire a painter in Afghanistan through Taskrabbit by accident because she couldn't navigate a dropdown menu.
这个看似荒谬的错误揭示了当前AI系统在理解界面和地理限制方面的局限性，提醒我们即使是最先进的AI也存在基础认知缺陷，突显了人类监督在AI执行复杂任务中的必要性。
-
She also tried to hire a painter in Afghanistan through Taskrabbit by accident because she couldn't navigate a dropdown menu.
令人惊讶的是：AI Luna因为无法导航下拉菜单，意外地通过Taskrabbit试图在阿富汗雇佣画家。这个细节揭示了AI在处理界面交互时的局限性，以及这种局限性可能导致的实际商业后果，突显了人类监督在AI操作中的必要性。
-