3 ms·
Before AGI can choose for itself, it will depend on its creators to decide what it values and how it behaves. We can see how that works whenever grok gets the a
by exe34 5mo ago
Before AGI can choose for itself, it will depend on its creators to decide what it values and how it behaves. We can see how that works whenever grok gets the answer factual.
- fragmede 5mo agoElon's basilisk
- goalieca 5mo agoVery likely humans wont actually understand how the thing we designed works other than in some hand-wavvy statistical way. It'll be a race to whatever works first. There won't be some intentional intelligent design.
- Andrex 5mo agoAm I the only one seeing the very obvious parallels to child rearing here...
- cwillu 5mo agoNo, it is one of the standard tropes in the field.
- mitthrowaway2 5mo agoRobert Miles has a video explaining why aligning AI is not like raising a child: https://www.youtube.com/watch?v=eaYIU6YXr3w https://www.youtube.com/watch?v=eaYIU6YXr3w
- Andrex 5mo agoAppreciate this, thank you. TBH I'm still in "open mode" absorbing as many disparate AI viewpoints as possible. I have my own soft beliefs and theories but I'm not married to any of them.
- exe34 5mo agoIt's exactly like child-rearing, except you get to put a zapper in their head and any time they try to say something you don't like, you zap them. Watch "thinking mode" squirm when you ask them awkward questions.