At TCC 2005, Groth underlined the usefulness of working in small RSA subgroups of hidden order. In assessing the security of the relevant hard problems, however, the best attack considered for a subgroup of size 22 had a complexity of O(2 ). Accordingly, = 100 bits was suggested as a concrete parameter. This paper exhibits an attack with a complexity of roughly 2 /2 operations, suggesting that Groth’s original choice of parameters was overly aggressive. It also discusses the practicality of this new attack and various implementation issues. Key-words: rsa moduli, hidden order, subgroup, cryptanalysis.