An artificial intelligence test was conducted at the University of Reading U.K. on Sunday, where Computers engaged in arguments, recited jokes and escaped tricky questions. 

Nearly a dozen volunteers indulged in two conversations, while typing away at split-screen terminals. The two conversations to be conducted at once, included one with a chat program and the second with a human. However, after five minutes when the volunteers were asked to discriminate between the two, may of them were not sure as to to whom or to what they were chatting to. 

Ian Andrews, one of the judges in Reading, reported, “There was one time when I was speaking to the two, and there was an element of humor in both conversations. That's the one that stumped me more than others.”

Some confident judges tried to trip programs up with questions related to day's weather, the global financial turmoil and the color of their eyes, which was exposed in the transcription of the conversation.

When tried to be tricked with such questions; Eugene Goostman, a “chatbot” created by Pennsylvania-based programmer Vladimir Vesselov quickly replied, “Blue Of course!” Eugene was among those five programs, which were designed to trick the volunteers. These programs chatted with the volunteers as a normal human being will do. However, Alice, which was supposed to be the sixth program dropped since it could not be set up in time. 

The Loebner Artificial Intelligence Prize's bronze medal was grabbed by Fred Roberts' Elbot, and it won it by fooling three out of 12 judges assigned to evaluate it. 

After receiving the prize, the Hamburg, Germany based consultant joked, “I wish I was as good at conversation as Elbot.”

The contest was organized and conducted with keeping the ideas of British mathematician Alan Turing in mind, who had come up with a subjective but simple rule to determine that whether the machines have the capability to think and reason. In his writings of 1950, Turing has specified that intelligence can be proved through conversation. This pointed to one justification that if in case the computers are capable of conversing like humans, then they thought like a human for all practical purposes. 

However, it is quite tricky to judge expressiveness of a computer, since there is a fair chance that humans might be prejudiced against a machine. So a test was designed by Turing in which a human judge had to indulge in two conversations simultaneously; out of which one would be with computer and the other one with computer.

The human judge will not be aware of who is who, and if incase the judge faces quite a difficult time while identifying the two, then it should be clear that computer has met the human standard of intelligence.