> Essentially that we just need to stop kidding ourselves that we can predict the tails, and to build greater buffers into all our assumptions from the beginnin

> Essentially that we just need to stop kidding ourselves that we can predict the tails, and to build greater buffers into all our assumptions from the beginning. He calls this "Radical Uncertainty" which I think is better as a practical foundation than attempts to model black swan events. I haven't read The End of Alchemy, but Taleb also doesn't even think modelling black swan events is possible or even useful. Any attempts that claims can predict black swan events is assumed as harmful. And he would also propose the same to build greater buffers into assumptions or even assume that it will fail in time. Antifragile is the book that explores this more on how fragile are predictions and it's far more useful to know where is the risk exposure and compensate for that instead of calculating the risk and trying to avoid it.