Evan Hubinger’s claim that RSPs are “pauses done right” is far more rigorous and sensible given the Paul-Christiano-cluster of beliefs about alignment (which I think is wrong). I recommend reading that post instead of this one, even if I also consider it a form of defection from the “let’s just aim for an actual indefinite pause”.
This post feels significantly less rigorous. I disagree with Jacob’s claim that this is a good example of “independent thinking”. I feel annoyed that Nora’s argument is getting more airtime after it was already shown to be significantly flawed given the discussion in the comments of her original post.
Evan Hubinger’s claim that RSPs are “pauses done right” is far more rigorous and sensible given the Paul-Christiano-cluster of beliefs about alignment (which I think is wrong). I recommend reading that post instead of this one, even if I also consider it a form of defection from the “let’s just aim for an actual indefinite pause”.
This post feels significantly less rigorous. I disagree with Jacob’s claim that this is a good example of “independent thinking”. I feel annoyed that Nora’s argument is getting more airtime after it was already shown to be significantly flawed given the discussion in the comments of her original post.