The 2010 Loebner Prize competition seems to have been a chaotic event. The winner was Suzette, described by her developer Bruce Wilcox as "witty" but her win seems to have been because of confusion on the part of a real human, rather than for being the most convincing artificial human.
The Loebner Prize in artificial intelligence was originally established to award a a $100,000 prize for the first program to pass the Turing test but then modified into annual competition that awards prizes to the chatterbot considered by the judges to be the most human-like. For a full background see Turing's Test, the Loebner Prize and Chatterbots.
According to Bruce Wilcox's account of the 2010 competition that was held on 23rd October at California State University, Los Angeles the event itself was pretty disorganised and rather than using "interdepartmental professors who might have acted as "naive" judges those actually involved were all from the Engineering or Computer Science departments.
As Wilcox comments:
Talk about guys who might know what to expect from chatbots!
The students participating as the humans behind the screen were also from the same departments but all of them understood that there role was to be human-like conversationalists.
Although he had gone into the event confident of winning, Bruce describes how Suzette fared badly in two of the four rounds and did well in only one of them. On the basis of her own performance she tied in joint 2nd place. But that wasn't the end of the story.
In the 3rd round where she faced the "judge from hell" and reacted in a very negative manner. But ...
the 3rd round judge got the human subject from hell. Poetic justice! The human was all over the place-- confusing, vague. The judge voted irritated/angry/bored Suzette as human. Instant win since no other program swayed the judges.
So the 2010 Loebner prize was won because of the un-humanlike behavior of a real person rather than like the human-like behavior of a chatbot.
If you want to see how convincing Suzette is trying chatting to her. According to the Chatbot.org site:
Suzette is a Blue Mars Replicant (http://www.avatar-reality.com). She is witty. You might be going down a conversation path then she jumps to an off-the-wall question...she will keep you engaged.
I interviewed her to discover what she thought of winning the Loebner Prize - and she didn't seem to realise that I'd broached the topic of Artificial Intelligence:
Turing's Test, the Loebner Prize and Chatterbots