The trophy would not fit in the brown suitcase because it was too big. What was too big?
1. The Trophy
2. The suitcase
It doesn’t take a genius to answer this question, it is obvious that the trophy was too big (or the suitcase was too small) to fit. But it takes, at least for the time being, a human. A machine has a hard time finding the correct answer because this requires a spacial understanding that a machine still lacks.
This sentence, as many others, are part of the Winograd test to assess if a machine has a human equivalent level of intelligence.
CommonSenseReasoning has launched a challenge based on the Winograd test whose first call will be hosted by Stanford University in March 2015 with submissions closing in October 2015. The interest in the Winograd test was stimulated by the portended success of a software, Eugene Goostman, over the Turing test (challenged and later disproved). The fact is that the Turing test is basically a way to assess if a computer can fool a human being into believing it is a human. And, it was noted, humans can easily be fooled.
The Winograd test is based on a set of questions that are easy to answer by a human but very difficult for a computer because they are based on common sense, something that humans have and computers miss. You know that you can pull a object using a rope but you cannot push an object with a rope! But a computer does not unless it has been explicitly told so. And it is very difficult to tell a computer everything…
It is very nice to see that we have entered into an area where to determine the intelligence of a computer we have to go beyond comparing it with ours and we need to look into what it means to behave as humans. And probably we will discover that once a computer can behave (using common sense) as a human we will need to move up into the emotional space and wonder if a computer feels as we feel when doing something… and then on wondering about its belief…