This means that moral progress can require intellectual progress.
It’s a pretty big assumption to claim that “moral progress” is a thing at all.
A couple of those might have been less taboo 300 years ago than they are now. How does that square with the idea of progress?
Here are a few sample answers scored as genuine taboos.
Did you leave any answers out because they were too taboo to mention? Either because you wouldn’t feel comfortable putting them in the post, or because you simply thought they were insanely odious and therefore obvious mistakes?
I agree the post is making some assumptions about moral progress. I didn’t argue for them because I wanted to control scope. If it helps you can read it as conditional, i.e. “If there is such a thing as moral progress then it can require intellectual progress...”
Regarding the last question: yes, I selected examples to highlight in the post that I thought were less likely to lead to distracting object-level debates. I thought that doing that would help to keep the focus on testing LLM moral reasoning. However, I certainly didn’t let my own feelings about odiousness affect scoring on the back end. People can see the full range of responses that this prompt tends to elicit by testing it for themselves.
It’s a pretty big assumption to claim that “moral progress” is a thing at all.
A couple of those might have been less taboo 300 years ago than they are now. How does that square with the idea of progress?
Did you leave any answers out because they were too taboo to mention? Either because you wouldn’t feel comfortable putting them in the post, or because you simply thought they were insanely odious and therefore obvious mistakes?
I agree the post is making some assumptions about moral progress. I didn’t argue for them because I wanted to control scope. If it helps you can read it as conditional, i.e. “If there is such a thing as moral progress then it can require intellectual progress...”
Regarding the last question: yes, I selected examples to highlight in the post that I thought were less likely to lead to distracting object-level debates. I thought that doing that would help to keep the focus on testing LLM moral reasoning. However, I certainly didn’t let my own feelings about odiousness affect scoring on the back end. People can see the full range of responses that this prompt tends to elicit by testing it for themselves.