Elbot is a droid with a sarcastic edge — it points out that it's a machine even as it tries to convince you it’s human. And yesterday it proved somewhat successful, fooling three out of 12 judges during the University of Reading’s Turing Test and taking home the Loebner Artificial Intelligence bronze medal. But does that mean that Elbot is on the verge of sentience? His creator doesn’t think so.Loebner judges yesterday sat at split screen terminals, simultaneously chatting with one of the chatbots and a human volunteer. At the end of the conversation, the judge selects which is the human and which is the computer program. Programs used various tactics to appear human; bot Eugene Goostman comes with his own backstory, including a childhood in Odessa and an Aunt Sonya who lives in America. Elbot tries to throw judges for a loop by joking that it might be a machine:
"Hi. How's it going?" one judge began. "I feel terrible today," Elbot replied. "This morning I made a mistake and poured milk over my breakfast instead of oil, and it rusted before I could eat it."
A quarter of the time, it was successful, beating out the four other programs and meriting it a bronze medal. A program is supposed to have achieved sentience if it can fool human judges 30% of the time, but Elbot’s creator, Fred Roberts, isn’t buying it:
"I don't think it's anything like thought," he said of Elbot's conversational prowess. "If you know a magic trick, you know how it's done, it's not magic anymore. Sorry to be so pessimistic."