X Prize aims to show AI is friend not foe

February 17, 2016
X Prize founder Peter Diamandis speaks during a presentation at the 2012 International Consumer Electronics Show on January 10,
X Prize founder Peter Diamandis speaks during a presentation at the 2012 International Consumer Electronics Show on January 10, 2012 in Las Vegas

An X Prize unveiled on Wednesday promised millions of dollars to a team that could best show that artificial intelligence is humanity's friend, not its enemy.

X Prize founder Peter Diamandis and the newly-appointed head of Watson at US technology veteran IBM David Kenny challenged software savants to demonstrate "how humans can collaborate with powerful cognitive and AI technologies capable of solving some of the world's grand challenges."

Diamandis and Kenny announced a $5 million X Prize on the stage of a prestigious annual TED Conference here, telling the audience it would be here that the ultimate winner would be picked by the audience in the year 2020.

"Personally, I am sick and tired of the dystopian conversation around ," Diamandis said, referring to high-profile public debate whether self-aware and smart machines would annihilate humanity or help it thrive.

Diamandis weighed in on the side of AI being vital to surmounting huge problems in the world and hoped that would be made clear when three finalists for the new take to the TED stage in four years.

Competing teams will be free to define challenges their AI creations are designed to tackle.

A half-million dollars in prize money will be awarded during an elimination process at IBM World of Watson conferences intended to whittle the field down to three teams that will use AI to deliver "jaw-dropping, awe-inspiring" TED talks.

Conference attendees will pick the one they think is best. The winning team will get $4.5 million in Watson X Prize money.

"X PRIZE and IBM believe that the partnership between humans and technology has produced some of society's most groundbreaking modern advancements, from landing a man on the moon to addressing climate change to mapping the human genome," the AI prize partners said in a joint release.

"We believe there's an opportunity today for others to push the boundaries yet again."

Teaching machines to think the way people do has been a hot topic in technology, with companies such as Facebook, Google,Microsoft, and Apple working to build smarts into services or products.

Watson, which gained fame in 2011 for defeating human opponents on the "Jeopardy" quiz show, has been reaching into its computing power since then for an array of other services.

The quest to build real brains into machines has caused some to fear it could turn on humanity in terrifying ways depicted in science fiction films along the lines of "The Terminator."

Explore further: Team trains Watson AI to 'chat,' spark more creativity in humans

Related Stories

'Jeopardy!' to pit humans against IBM machine

December 14, 2010

(AP) -- The game show "Jeopardy!" will pit man versus machine this winter in a competition that will show how successful scientists are in creating a computer that can mimic human intelligence.

Formula E to stage driverless, AI 'Roboraces'

December 2, 2015

Real-time computing algorithms will replace humans in an innovative series of driverless races to be staged alongside the regular Formula E season, promotors said on Wednesday.

When machines can do any job, what will humans do?

February 13, 2016

Rice University computer scientist Moshe Vardi expects that within 30 years, machines will be capable of doing almost any job that a human can. In anticipation, he is asking his colleagues to consider the societal implications. ...

Recommended for you

Microbes help turn Greek yogurt waste into fuel

December 13, 2017

Consumers across the world enjoy Greek yogurt for its taste, texture, and protein-packed punch. Reaching that perfect formula, however, generates large volumes of food waste in the form of liquid whey. Now researchers in ...


Adjust slider to filter visible comments by rank

Display comments: newest first

5 / 5 (2) Feb 17, 2016
Woohoo! Machines that get hooked on meth and cheat on their taxes. I suspect when they say "Think like people" they aren't talking about the general population...
5 / 5 (2) Feb 17, 2016
How about an X-prize of millions of bananas to any group of non-human primates that can prove that humans are their friends and not foes?
not rated yet Feb 17, 2016
for real? " promised millions of dollars to a team that could best show that artificial intelligence is humanity's friend, not its enemy."
They wanted catered cherry picked results?? that they are openly willing to pay for? Sounds like the same thing the cigarette companies did and now coal. Why rush this unless you have some ulterior goal? Guys like this like to peddle off they're idea like its the end all of the grand picture... but its not and never is.
not rated yet Feb 18, 2016
That's like promising millions of dollars to show that Russel's Teapot is blue.

First you have to produce it for examination.

Competing teams will be free to define challenges their AI creations are designed to tackle.

That's the problem. AI designers define what Artifical Intelligence means in the first place, so we don't actually get artifical intelligence, just algorithmic dumbness that appears smart because nobody takes a closer look at what it's actually doing.

For "intelligence" to be meaningful it has to have qualitive differences to what is not intelligent, because intelligence is not a thing like a heap of sand or a head of hair. Intelligence is a way of doing things, and some things in nature do it that way, others don't.

Intelligence is a phenomenon and a property, and therefore the continuum idea of adding more of the same thing until it "becomes intelligent" doesn't apply. No matter how much non-intelligence you add, you don't get intelligence.
not rated yet Feb 18, 2016
Think of it this way. If we want to have "cycling" - which is a similiar phenomenon - then no matter how much "not-cycling" we have, we won't get "cycling" which is defined as people riding bicycles.

Modern AI research does to intelligence the same as if we took cycling, looked at the start and end conditions, and concluded that we can pick the person and bicycle up with a car, carry them to their destination, and call that cycling. We simply hide the car and make it seem like the person is pedaling, or just pretend that pedaling is happening through semantics, and call that "Artifical cycling".

The problem with AI is that nobody has sufficiently defined "intelligence" for there to meaningfully be "Artifical Intelligence". We haven't sufficiently studied or described what happens when "intelligence" happens.
not rated yet Feb 18, 2016
We need a prize winning EXTENDABLE OPEN SOURCE AI.
One not designed by accountants to drag you into their chosen cul-de-sac.
One that can be TAUGHT not take you to Bing search engine if you say "Ring Frank".
One that is MODULAR whether your target interest is medicine or 1970 rock music riffs.
One that definitely doesn't advertise, and is equally suited to either being confined to knowing your laptop media collection or extended to 11 dimensions looking for a credible new GUT on physics websites, depending on your mood and choice.
5 / 5 (1) Feb 20, 2016
Two recent movies on the subject which I really liked are Ex Machina

-and spielbergs A.I. Artificial Intelligence

-a real tear-jerker.
5 / 5 (1) Feb 20, 2016
some things in nature do it that way, others don't
Sounds like you're expecting AI to be just as survival- and reproduction-oriented as that found in nature.

What would intelligence be like that was not based on these attributes? We see it all around us already.

All machines have a modicum of intelligence. The proximity alarm responds to external stimuli just as any animal does, but it is not concerned with survival. We design it to respond to certain things in certain ways.

Animal creativity is similar, as responding to stimuli in ways which maximize it's survival and reproductive potential.

We consider human creativity as something novel, but it is only a more complex form which we have defined to accommodate our flaws and mistakes in very complex cultural contexts.

Artists and priests and politicians want to eat and spawn too, and not everyone can be growers and fighters.

We have been designing our tech without these constraints. Technology is AI.
not rated yet Feb 20, 2016
Incidentally if you want to know if you're a psychopath or not, watch the Spielberg movie without bawling your eyes out. No true human could.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.