> superintelligence is still 20 years away, and even then he only gives his prediction a 50% chance I don't know the details of Hinton's probability distribut

> superintelligence is still 20 years away, and even then he only gives his prediction a 50% chance I don't know the details of Hinton's probability distribution. If his prediction is normally distributed with a mean of 20 years and a SD of 15, which is reasonable for such a difficult and contentious prediction, that puts over 10% of the probability in the next 3 years. Is 10% a lot? For sports betting, not really. For Mankind's Last Invention, I would argue that it is.