> IQ 1000 is 60σ But that is assuming that intelligence is a normal distribution, i.e. arises from a contributions of many independent random variables. Which

> IQ 1000 is 60σ But that is assuming that intelligence is a normal distribution, i.e. arises from a contributions of many independent random variables. Which is a decent approximation for underlying natural processes (human genetic pool and environmental contributions). But it is just an approximation. For starters, there is a lower bound to intelligence (inert matter) so we're not really dealing with a normal dist, the truth would be closer to a skewed, semi-bounded distribution, we just don't see that because we're just measuring humans instead of also including animals on the left and hypothetical superintelligences on the right. Secondly, modelling it as random independent variables breaks down if you start to deliberately engineer all those contributing factors. Natural processes are exceedingly unlikely to hatch ants from fly eggs due to random genetic mutation. But a process like evolution or engineering can bring such things into existence. So asking for an IQ of 1000 is simply saying that you want to be more intelligent than random, natural reproductive processes could give you anytime soon.