For what it's worth, this is in line with my estimates, which are not just on the basis of claimed interest in x-risk reduction. I don't think that an order of magnitude or more less than this level of effectiveness could be the conclusion of a credible estimation procedure.