I'm not sure I understand your point. Who is stupid, besides those who apparently don't understand the importance of true randomness in key generation? There are 27,000 RSA keys out there that offer no security whatsoever. In any case, I would say that the title of the paper is quite stupid. There is no reason to make a distinction between "really bad" (two identical keys) and "really, really bad" (two keys with a prime gcd). Both cases should be considered as completely insecure. The title suggests that the solution is to switch from RSA to a single-secret system, whereas the actual solution is to make sure your random generator is in working order. Hopefully the CRYPTO referees will point this out. Also, the 99.8% security claim doesn't make much sense, at least not to me. I don't think it is reasonable to assume that, say, Google's RSA keys are as likely to be bad as those used at the average homemade online store. I would rather guess that there is one group of RSA keys (Google, Microsoft, Amazon, government bodies, and so on) that are almost always properly generated and another group of RSA keys (the homemade ones) that are almost always flawed in one way or another. Taking the average over two such disparate groups seems a bit strange. Jakob Warren Smith:
front page of NY Times just featured this paper
http://eprint.iacr.org/2012/064.pdf
which is hilarious! Just when you thought you'd seen the ultimate in human stupidity...