The article we're discussing is about rating the accuracy of Kurzweil's predictions, it is not about the soundness of his reasoning. I agree that regardless of how we rate his predictions, the strategy he used to make those predictions has issues of its own.