I think you missed the point they were making. The risk is the downside.

At a startup that's when your paper millions turns out to be worthless, you're realise you're never going to see the last six months of salary that the company couldn't afford to pay you "at the moment", and you're in debt to your eyeballs.

At Google, the downside is you don't get your millions, but you keep your cool job at Google and continue to collect your pretty decent Google paycheque.

Having done the startup thing too many times, at this point, I'd prefer Google.

I remember I was using an old Amiga disk-doctor type utility, and I got this wonderful error message:
'Cannot mark bad blocks because the block used for marking bad blocks is bad.'
Say THAT 10 times fast.
I've been telling this wonderful story to my computer friends for ages, and finally, I have an online outlet for it! Yay!