Published: Thursday, 16th October, 2008 12:09pm

Elbot has the robot gift of the gab

by Annabel Williams

Comments (0) | Print | Email

MAN and machine became a little more like minded than usual at a competition hosted in Reading at the weekend.

The 18th annual Loebner Prize was held at Reading University on Sunday - an international artificial intelligence contest to find the computer programme which can fool the most humans into believing it is just as human as them.

Five computer programmers pitted their algorithms against a judging panel in another room, with assessors sitting at a split computer screen displaying two written conversations.

One of the conversations was with a human, the other with a machine, and after five minutes the judges decided whether they thought each side was human or computer programme.

The programme Elbot, created by Fred Roberts, won the $3,000 prize (around £2,250) which was presented by the sponsor, American scientist Dr Hugh Loebner.

But no machines managed to pass something called the Turing Test, a benchmark for artificial intelligence set up by British mathematician, Alan Turing, in 1950.

Turing's criteria is that "if, during text-based conversation, a machine is indistinguishable from a human, then it could be said to be 'thinking' and, therefore, could be attributed with intelligence".

Elbot convinced 25% of judges that it was human, the closest any machine has been to the 30% Turing threshold. Having never won the prize before, despite entering Elbot in 2002 and 2003, Mr Roberts said: "I'm elated, I'm really happy that the system has come through.

"It's surprising that something with a robot personality could come through something like this."

Mr Roberts, an American based in Hamburg, said his studies in psychology, as well as computing, helped him win because he knew how to steer topics of conversation back to subjects his programme could handle.

Turing Test organiser, Professor Kevin Warwick, from Reading University's School of Systems Engineering, said: "This has been an exciting day with two of the machines getting very close to passing the Turing Test for the first time.

"In hosting the competition here, we wanted to raise the bar in artificial intelligence and although the machines aren't yet good enough to fool all of the people all of the time, they are certainly at the stage of fooling some of the people some of the time."