Secret trigger words could make AI-powered robots dangerous
- AI-powered 近 90 天出现 45 次
- 上一次：3 天前 · North Korea says it tested new AI-powered torpedo system
发生了什么
A team of computer scientists at Binghamton University has demonstrated that hidden trigger words inserted maliciously into LLMs could cause a robot to go off-script and perform dangerous actions—and they developed a defense.
摘要按规则整理自下方来源原文