About

About The Turing Hub

The Turing Hub is a kind of switchboard that connects people to chatterbots or other people for the express purpose of having a conversation with something and then being asked to describe what it was you were talking to. This activity is part of the Turing test.

At this site, our goal is to be able to present each visitor with a verbal behavior experience using identical interfaces for the possibility of conversation. We want to see if people have more than a 70% chance of guessing the true identity of their conversational partner.

How to conduct a bonafide Turing test.

You can have a Turing test between people, bots, or some combination of them. Have it at a party, or conduct it like a scientific experiment. It can be both fun and entertaining. You can find your own people to chat with, and we have a list of some of our favorite bots on the Resources page.

Using Google talk, you can open a window to chat with another human being. Open the chat bot window and then there are some roles to play. Pick one person to be the referee. You might even cover the top of the computer screen so there are fewer clues about which window is which. Another person is the judge, or tester, and this person actually chats with the bot and one or more people (called confederates). There are various opinions about timing the chat. Some say 5 minutes, others say 20 minutes or more, but the longer the chat, the less likely the bot will pass for human. At some point, the referee must declare the test is over, and the judge then decides which chat was with a person, and which chat was with a bot. On the bot screen, you may rate the bot's performance.

Dr. Loebner would have you do conversations in pairs, where 100 points is divided between the human and the bot as their chat score. This allows for testing several bots and humans and then being able to make a transitive comparison of all the participants, so that you may infer that person A was more human than person B, or that one bot was less human than another, for example.

Turing predicted that machines would eventually be able to pass the test. In fact, he estimated that by the year 2000, machines with a billion bits (about 119 MB) of memory would be able to fool 30% of human judges during a 5-minute test. He also predicted that people would then no longer consider the phrase "thinking machine" contradictory. He further predicted that machine learning would be an important part of building powerful machines, a claim which is considered to be plausible by contemporary researchers in Artificial intelligence.

Why are there no scores for the human beings? You may keep your own scores and email them to us if you find them intersting, or would like for us to post them here. It is not feasible for us to man the chat room 24/7 as the bots do. At this point, we have not compiled a composite score of the various human beings who chat here. We are more interested in perfecting the chat bots at this point, and depending on which human does the chatting, it is obvious to us who the human is. When you give a score for one of the humans, we log it in a separate category from the bots, but that category is not available on the scores page. The scores page represent tests conducted with various bots over the years.

There are some technical limitations associated with this network. This project would not be possible without The Human Beings.

This project bears no relationship to Albert One or the Loebner Prize Contest. Though some of the participants here are past Loebner Prize contestants, it has not always been possible to utilize programs like Albert One as web applications. At present, this site is a proof of concept primarily.

See the TuringHub Blog for the project diary.

[main page] What's with the bicycle you may ask? This is our homage to the classic television series, The Prisoner, with Patrick McGoohan's famous line, "I am not a number! I am a free man!" It is a part of human nature to resist automation because of the fear that it takes away our ability to decide for ourselves, our free will.