This weekend I watched a movie called “Robot and Frank” directed by Jake Schreier. It’s about an aging father, Frank, who is becoming forgetful as he grows older. His son gives him a robot butler who is programmed to take care of him. The movie explores the relationship between Frank and his robot, who he despises at first but then grows fond of. The movie is set in the future, but not the distant future. Almost everything looks the same except for robots, communication methods (such as video to video calling), and computer screens. Computer screens are more advanced and responsive than the ones we have today, and work with augmented reality rather than the mouse and keyboard like modern computers do.
This movie inevitably had me thinking about programming and how far we are from simulating artificial intelligence. Inventions such as Siri, Watson (the robot who beat the best human at Jeopardy) and the Cheetah robot appear to be a natural progression towards artifical intelligence. The Turing Test, which is best summarized by Wikipedia as: “a test of a machine’s ability to exhibit intelligent behavior equivalent to, or indistinguishable from, that of an actual human.” I think that the general attitude of the population regarding artificial intelligence is that it is inevitable, and the complexity of creating a machine that would pass the Turing test is often overlooked.
However, now that I am a programmer, I can truly appreciate how complex it would be to create a machine that behaves in a way that is indistinguishable from an actual human. Even in today’s best and most popular applications, there are still many bugs which result from the developers inability to foresee every situation that could cause a bug. It is difficult enough to develop an advanced application that does straightforward tasks. I cannot even begin to fathom how you would use programming logic to create an algorithm to understand subtleties of human nature as well as the subtleties of the English language.
The strange thing about humans is that we are not fully logical beings. Humans say things that we don’t mean, and often context and body language play a large part in how we understand what people actually mean versus what they say. Although I believe that some day we may reach the point where we are able to stimulate artificial intelligence, I definitely won’t be buying one of the first robot butlers. I think that it will take a long time to create sophisticated enough computers to simulate a type of intelligent interaction that won’t make you want to rip your hair out.