I don't think Kurzweil is claiming to create anything close to human intelligence in the next few years. Likely, people will continue to say it's not "really" i

I don't think Kurzweil is claiming to create anything close to human intelligence in the next few years. Likely, people will continue to say it's not "really" intelligence for quite some time. I think computers are going to gradually start fuzzing the line between AI and humans. As they get better, we will slowly start to treat them more like we would a human (i.e. we would start literally writing a question for Google to answer, not worrying if it will know what we mean). Eventually, it might be the case that we speak with a phone operator for some company and don't notice that it's a machine - unless we ask it something off topic, and eventually even that won't be a reliable heuristic. At what point do we finally say the computer "understands"?