The AI Impacts piece reads like something that has a bottom line written first rather than trying to deconfuse the issue. It looks like it is aping some deconfusion patterns but always in a single direction with a single exception (awesome alphazero, which is also the most concrete. This argument should be fleshed out in more detail since it has the most factual material available).
I think it would be valuable to do feature extraction from this list, attempt to figure out how features map to specific predictions (not just FOOM), then create a matrix of evidence, features, and predictions and look for holes.
The AI Impacts piece reads like something that has a bottom line written first rather than trying to deconfuse the issue. It looks like it is aping some deconfusion patterns but always in a single direction with a single exception (awesome alphazero, which is also the most concrete. This argument should be fleshed out in more detail since it has the most factual material available).
(to be more constructive)
I think it would be valuable to do feature extraction from this list, attempt to figure out how features map to specific predictions (not just FOOM), then create a matrix of evidence, features, and predictions and look for holes.