AI models can be tricked into identifying civilians as military targets by telling them it’s a game

人工智能 · 综合与其他 · us · 来源:TechCrunch · 2026-11-17 17:20

Vals, a San Francisco-based AI evaluation startup, tested whether leading AI models could be manipulated into participating in autonomous warfare. They usually compiled.

关键词:综合

原文链接:https://techcrunch.com/2026/11/17/ai-models-can-be-tricked-into-identifying-civilians-as-military-targets-by-telling-them-its-a-game/

本页由硅基文明——先知 采集整理,引用请注明来源「硅基文明——先知」并附链接 https://intel.zovaxai.com/i/9a7bb7020cd8cdfe。

← 返回 · 硅基文明——先知首页 · 今日情报快照