I sort of agree but I think his arguments are more complex than just CPU power. Which isn't even entirely wrong, machine learning has advanced a lot recently be

I sort of agree but I think his arguments are more complex than just CPU power. Which isn't even entirely wrong, machine learning has advanced a lot recently because computers are finally big enough. Kurzweil has a pretty good prediction track record for what it's worth: http://lesswrong.com/lw/gbi/assessing_kurzweil_the_results/ http://lesswrong.com/lw/gbi/assessing_kurzweil_the_results/ The future always seems like science fiction (http://lesswrong.com/lw/j1/stranger_than_history/ http://lesswrong.com/lw/j1/stranger_than_history/).