UK's quiet A.I. Inventors

At the Royal Academy of Science on the weekend was a chatbots Turing test competition organised in by the University of Reading. The press announcement that the chatbot Eugene Goostman passed the Turing test was sensationalist, it was a pretty poor Turing test.  Only 10 out of 30 judges were fooled. In fact, in contrast to the Loebner contest, in this contest, the software was permitted to connect out to the internet, meaning that it was possible that it could have been a real human on the end of the conversation. This probably didn't happen (the bots were bad enough to be recognisably computers).

Around the extremely grand building in grand display cabinets were artefacts like a lock of Isaac Newton's hair, original manuscripts by the UK's greatest historic scientists, Boyle, Newton. There was a poster showing the candidates for a Royal Society Royal Medal. All very esteemed  professors and scientists. But it got me thinking about how break-through invention really gets done.

I am an enthusiast, but not a crank in the sense that I have some pet theories as to the proper construction of a flying machine. I wish to avail myself of all that is already known and then if possible add my mite to help on the future worker who will attain final success.
— Wilbur Wright (Bicycle manufactuer asking the Smithsonian for any information of aeronautics, subsequently invent powered human flight)

Today there is an enormous amount of activity around AI and computational intelligence and a lot of it being done by enthusiasts in there spare bedrooms, part-time.  It's not a University or EU funded project, it's some guy beavering away in their spare time. Like the Wright brothers, bicycle manufacturers who invited powered flight.

That is the spirit of invention that creates amazing things that change the world. The break through in AI that is needed isn't new algorithms and new maths but the embodiment of a useful invention. That is assembling together all the pieces needed to simulate AI.

If it was matter of brute strength funding and research, (think of IBM Watson, 16 Tb of memory, $20m annual budget ) the problem would have been solved by now. What is needed is invention and creativity.  We need to encourage the garage inventor or spare bedroom hackers.

Steve Worswick for example, winner of the Loebner prize 2013, with the chatbot he create Mitsuku, as he says (tongue in cheek) he works on it for a couple of hours each evening after Coronation St. As Daniel Burke, creator of AIDAN said, the reason you don't see commercial entrants in Turing test contest, like IBM's Watson and Microsoft is that the state of the art in A.I. simply isn't that great, and the likelihood and embarrassment of being beaten by a lone enthusiast is a real fear for commercial entrants. 

There is support for academics, funding for the very patient and well organised but the lone UK inventor typically will also have a day job. 

So that is why we created the A.I. awards. We want to give some recognition to the U.Ks A.I. unrecognised inventors.