emollick· @emollick · X·· 2 天前AI 评分
Anthropic 的 AI 网络安全观点应受重视
在不过分轻信的前提下，我们应该更认真对待来自各实验室的、看似“异想天开”的 AI 网络安全观点。例如：Anthropic 曾因表示 Mythos 级别的 AI 改变了网络安全而被嘲笑，@polynoamial 也曾因表示模型会找到新颖的协同方式而被嘲笑。
Without being overly credulous, we should be taking seemingly “out there” ideas about AI cybersecurity from the Labs more seriously. Examples: Anthropic was mocked for saying Mythos-level AI changed cyber & @polynoamial for saying that models would find novel ways to coordinate.
来源：emollick · x.com