Should be trivial to show a working proof on a smaller-than-usual RSA number if "this really destroys RSA".