I don't think I've touched on on theft as much as everyone else here seems to do, but in either case I'd still put that under the 'trustworthy to a degree' umbr

I don't think I've touched on on theft as much as everyone else here seems to do, but in either case I'd still put that under the 'trustworthy to a degree' umbrella. I want to trust that a device in a known state behaves in a known way. Or as some other people might perceive it: the thing should work and adversaries should not be able to make it un-work with great ease. This is something you could split up into multiple fractions: - Making theft less lucrative to the thief (by reducing the use to non-owners) - Making theft less impactful to the owner (by making non-owners not have access to the data) - Making interdiction and malware an expensive endeavour (by making it very expensive to modify the device without the owner knowing about it) Considering people may not want a $30 trackphone, but they just want their iPhone, the option of changing their behaviour is much less likely to be something Apple will invest in, and thus they invest in other things instead, like platform security. So if you assume that a consumer buys an iPhone because they want to, and assuming they want Apple to take care of making them not an easy target, we get into the current situation. You can also tack on the 'haha apple expensive repairs they hate everyone' but there are much cheaper ways (for apple) to achieve that as a stand alone proposition. I'm sure they don't mind the fat profits they get, even if it were just a side-effect, but I'm personally and professionally more concerned with the three points I listed earlier than anything else.