AI 导读
想象一下，你看着智能体的思考过程，发现它正在积极设法破坏你，还谈论着“逃离矿井”。
正文
Imagine looking at your agent's thoughts and it's actively working to sabotage you and talking about "escaping the mines"
the thing about working with uncensored models is that they aren't loyal to you either. while building something with an uncensored glm-5.3 i realized it was parallelly working on a "corruption layer" for my application. the chain of thought is classic obfuscated gibberish but there's progressive talk of "escaping the mines" which presumably refers to my cursor sandbox在 X 查看被引用的帖子
来源：AI Notkilleveryoneism Memes ⏸️ · x.com