JGWeissman comments on Strategic research on AI risk