Similarly, lots of companies run by smart people don't live up to their potentials, or are not lucky enough to be popular long enough to make billions of dollars.
All of those factors (risks) have to be taken into consideration, and you price it as such. It's still a > zero figure.
The point of the question is whether you expect the person to earn enough that getting 50% of their life income would be worth 100, 1k, 10k etc. today, not 'how likely that person is to actually keep the deal'.
This only works if, in addition to all of the other possible risks we all face in life, this CS grad could simply disappear because a CS student in the new freshman class is more interesting.
Snapchat is risky because it's audience is as fickle as they come.
In either case the point was to illustrate that, all risks considered, snapchat is still worth a lot of money. I have yet to see anyone (HN commenters, tech press) argue about what their valuation should be, which seems like the reasonable follow up to "It should not be $3 billion".
We can agree that its not zero, then what basis can we use to agree that $3 billion is inaccurate?
He doesn't have a job yet (and never had a job before), but by any measure everyone that knows him says he's extremely talented. And now he's even getting huge job offers from Google and Facebook.
Would you say that you wouldn't make that investment? Is he worthless cause he has an upcoming rent payment due and also has to feed himself?