> Electronic calculators are superhuman at arithmetic. Calculators didn’t take over the world; therefore, there is no reason to worry about superhuman AI. Wel

> Electronic calculators are superhuman at arithmetic. Calculators didn’t take over the world; therefore, there is no reason to worry about superhuman AI. Well but they did overtake arithmetic. > Historically, there are zero examples of machines killing millions of humans, so, by induction, it cannot happen in the future. Well there are also zero examples of a superintelligent AI that doesn't kill humans, does that mean that it cannot happen either ? > No physical quantity in the universe can be infinite, and that includes intelligence, so concerns about superintelligence are overblown. How is intelligence physical ? > we can always just switch it off Yeah, except at a time when we get so comfortable with an AI that turns things on and off for us. Add an "innovation" of not being able to turn the AI off by design (e.g. internet connection on SmartTVs nowadays) and there you go. It's suprising to me that these people call themselves scientists, perhaps the quotes were taken out of context or even paraphrased.