"It's not our fault," the cry goes up as hedge funds start reporting their losses. Chief among them David Viniar, CFO of Goldman Sachs, who says “We were seeing things that were 25-standard deviation moves, several days in a row".

No, you weren't. You were seeing things that your model told you were 25 standard deviations away from the norm. And then, the next day, it told you that you were seeing them again.
Now, from my dimly-remembered statistics lectures, something 7 standard deviations away from the mean of a normal distribution will occur on the order of one in a hundred billion times. Something 25 standard deviations away... well, it's meaninglessly unlikely. If you wrote an asset price on every atom in the solar system, one of them might be that far off the mean.

Or, alternatively, the model might be wrong, and be vastly underestimating the probability of improbable events. The Fat Tail strikes again. But it's depressing Viniar didn't say "well, obviously, our models were wrong" rather than bringing in Infinite Improbability...