Advances in interpretability would boost capabilities more than alignment
Interpretability is considered by many a major bottleneck for alignment. As a result, a lot of people who are concerned about AI safety, and especially about extinction risk and other catastrophic risks, work on interpretability. In this post, I want to argue that better interpretability methods would likely speed up capabilities dramatically. Worse, they would speed up capabilities much more than they would speed up alignment. This is true for the current world, and for any world without extrem...
Read full article →