资讯探索

一个根本性的缺陷使得大型语言模型极易受到攻击。

原文：A fundamental flaw leaves LLMs strikingly vulnerable to attack

MIT Technology Review · score 43.9 · 7/30/2026, 3:15:19 AM

摘要

It is impossible to make large language models fully secure against hacks because of a fundamental flaw in how they work, a team of researchers argue in a paper presented at the International Conference on Machine Learning, a top AI conference, this month. The claim has huge implications for the safety of this technology, which…

原文摘要：It is impossible to make large language models fully secure against hacks because of a fundamental flaw in how they work, a team of researchers argue in a paper presented at the International Conference on Machine Learning, a top AI conference, this month. The claim has huge implications for the safety of this technology, which…

原始内容

It is impossible to make large language models fully secure against hacks because of a fundamental flaw in how they work, a team of researchers argue in a paper presented at the International Conference on Machine Learning, a top AI conference, this month. The claim has huge implications for the safety of this technology, which&#8230;