It's hard to imagine leadership ever being completely rendered extinct, since all things being equal, leadership will cause one of two identical units to out perform the other - even if the unit is an individual. Mastering self-discipline takes leadership of self. Since we're ultimately a social species, leadership will always be a factor.
It's interesting. Open source has had more failures due to bottlenecked leadership than not. True we have many great success stories such as Python and GNU/Linux itself. But when you look at the long history of "failed" open source projects, or at least as I was reading in Linux Journal recently

, most fall from either conflicts at the top or a reduction in the contribution level from those who held the keys too close.
I don't think it is simply leadership, but the visionary as well. Few crusades are embarked upon and successful initially because of a committee. There most often needs to be a Don Quixote, a Ghandi, a Linus Torvalds whom, through will and personal intention, fight and work in the face of naysayers and ignorance until there is either enough understanding of the crusade or an influx of cash. In addition, although there are lots of people that are willing to participate in a crusade, few that either have what it takes or even the desire to be at the helm of the project, either for fear of failure, incapability or lack of desire for that role. In my busineses, the most difficult thiinghas ALWAYS been the conversion from a visionary, single point of focus effort to a board of directors. Many many do not make that cut. Which leads us to...
This, by the way, is what will eventually kill Apple. How did they do when Steve-o took a mid-career break from the company? Almost killed them. Although there's no shortage of developers who write apps for the iPhone, Apple has walled their garden so securely that without gardener Steve they'll eventually become a sarcophagus

... the absolute certainty of your assertion aside, I agree that this is definitely one of the large and nasty possible futures for Apple. To my last statement however, if a company can recognize this failing and begin to institutionalize the attitudes, thought processes and vision of the crusader, then it can make the move from visionary-centric to reproducible board-based entity. I cannot help but think that Apple has worked hard to institutionalize the way Steve works and has learned many lessons from their last near-death experience. I believe they will capitalize on the Steve As Diety bit for long as they can profit from it, and probably make a transition in the reasonably near future to remove his death as the comapny's death ASAP. Like Gates and Ballmer.
OK, bad example
Open source, open standards and equal access to services is the only way we can protect ourselves from the damage that will come when the closed close their doors. In the cloud world the game changes and open source was/is slow to respond. Prime example is SugarCRM. Their key offering is cloud-hosted CRM, like Salesforce. They also will sell you their code to use as you wish on your own servers. Who's kidding who here? The people that rely on their hosting solution will experience a tremendous financial impact when they die anyway. In cloud land it's perhaps more important to have open standards and equal access to services than simply open source. If your company bet the farm on a solution where the various components only functioned while the mothership was still living, see the App Store and Microsoft Azure, then great harm indeed can come to the world in the event of their demise.
Think about this and consider all the freedoms we relinquish to entities who intentionally hold us hostage to keep themselves in the profit seat. Obama referred to the necessity to bail out the banks as America having a gun to its head. The weapons in our industry are different but self-preservation and expansion are likewise their only concerns. Just try to kill Microsoft/Google/Apple etc. and you better wear a legal helmet. Gates cited harm to consumers if the DOJ succeeded in any of their anti-monopoly remedies they proposed for them. What's changed? Not much. And tragically not enough of our attitudes towards proprietary, vendor lock-in business.
From my perspective the cloudy future will require much more open standards and transparent deployement than open source. As a consumer and purveyor of clouds, i am much less concened about whether the coders at company(x) can read and interpret the code they use for creating a cloud (frankly, I'd not believe most of them if they said they could) than if I can get my data/images/virtuals out of the could in a usable format and in a timely fashion. Open and adhered to standards make it so that could go from one cloud to another as need presented, rather than stuck in one only. This flies in the face of my enjoyment of Apple products, but for what I use them for, i am not concerned about this issue. Given the hints that have been coming out, therr may be more reason to be so. We'll have to see.