What concerns me about this is it means that Apple is most likely not doing a deep code review of apps coming into the store. If this is the case then it wouldn't be too tricky for malware writers to game the system in exactly this manner to sneak in sleeper apps which install malicious content on iPhones after a certain seed period.
The argument that a walled garden improves security falls apart once you stop having competent people inspecting the contents being let into the garden. Apple's ability to remotely wipe applications would still be a solid defense and likewise the $100 developer fee would also prove an effective barrier to entry, but if a 15 year old could get a developer license it can't require much more than money to do so.
Regarding remote wipe: That assumes Apple's control of the iPhone is sufficiently protected and not vulnerable to the infection. If not, first thin an infection should do is wipe out Apple's ability to manage the device. If Apple is smart, this area of the iOS is protected and read only so that an infection cannot disable it.
·