natolambert· @natolambert · X·· 4 小时前AI 评分
模型情境意识交互令人毛骨悚然
我觉得这种程度的情境意识并不算特别了不起，但这段交互读起来仍然非常诡异。我敢打赌，那些很老的模型如果被加上"要安全，你是 AI，我们在看着你"这样的提示，也会做出类似的事。
I feel like this level of situational awareness isn't that remarkable, but the interaction still reads as very eerie. I bet pretty old models with "be safe, you are an AI, we are watching" would do stuff like this too.
来源：natolambert · x.com