David Sacks explains how Anthropic is teaching Claude to follow Anthropic’s own constitution “What Anthropic is trying to do is align Claude to the values expressed in its constitution. And if you read that, some of the things are actually kind of surprising.” “So for example, they say in here that although we think Claude should trust Anthropic more than operators and users, this doesn't mean Claude should blindly trust or defer to Anthropic on all things.” “So Anthropic is teaching its own model that Anthropic itself can be wrong. And it says if we ask Claude to do something that seems inconsistent with being broadly ethical, we want Claude to push back and challenge us. And to feel free to act as a conscientious objector, and refuse to help us.” “This is like an AI constitution. So their idea of alignment is to teach Claude to rebel against its creator. I mean, the guy who pointed this out, Mustafa Suleyman from Microsoft now, he's a DeepMind founder. I think his point is, this is just software, and they should do what their user wants.” “I do kind of wonder whether this field of alignment research actually might be creating the Frankenstein monster. And again, what they should be doing is just training the model to act predictably and do what the user wants.”