Re: New AI Prize

> > What the AI community needs is a new AI Prize -
> > s.th. like the X-Prize. See http://www.xprize.org/
>
> [..]
>
> On the other  hand, contributors of free software are numerous enough
> but are not motivated by the money. I doubt that the existence of a
> AI prize would increase their output. There are already a number of
> AI open software projects working.
>

Most open source projects really seem to lack motivation and
quality, but open source is a topic that maybe should be discussed
elsewhere. Basically I think it is a good thing, but I don't believe the
existing AI projects will achieve very much.

A prize would be useful to motivate companies.
The only thing companies seek and understand is money.

I think it is possible to create a system with true AI
in the original sense. Now. With the systems we have
today. Stephen Wolfram said in his talk "HAL Isn't Here ?" from 1997
http://www.stephenwolfram.com/publications/talks/97-Cyberfest.html
"the only thinking things around today are still just humans".
But he adds "I don't think it's going to take another generation
of computers, or any kind of fancy supercomputers. I think
that just the standard computers we have right now have big
enough memories and are are fast enough to do it."

He is right. We have the tools to do it, but it can not
be done by a single person alone. At the universities,
every scientist works to a large amount alone. Sometimes
there are small collaborations between 2 or 3 scientists.
But scientists don't like to make large collaborations
or teams, because everyone wants to earn the fame alone.
An exception are for example the publications at CERN,
or the publications of the human genome project.

And many reinvent the wheel several times. Marvin Minsky
has written his PhD thesis "Neural Nets and the
Brain Model Problem" in 1954, and he was among
the first who examined learning in recurrent neural
networks. Researchers are still fighting with recurrent
neural networks models today, 50 years later, and
many of them don't know Minsky's pioneering work.

I said the AI community needs a kind of new AI Prize
because people need something that melts them
together to achieve a common goal. We can reach
the goal of building a "thinking" machine today, but
only if we work together, and if we don't give up
trying. As a french saying noticed :
"To believe a thing is impossible is to make it so"