Watson: You can be way smarter. As long as you do not look like me

Remember Watson? The IBM computer that successfully beat the living beejezus out of an army of Jeopardy champions? Turns out most people think it’s kind of cute. Same thing happened when Deep Blue (still called Deep Thought in 1997) won over six chess games against world champion Garry Kasparov.

It’s cute that a machine beats a human. Just… don’t call it thought: that would be immensely disturbing. Deep Blue is fine Deep Thought is creepy. Watson is fine, as long as he looks like a set of silicon on steroids. It needs to stay a thing. Our psychological and ethical sensors seem to take all AI (artificial intelligence) nudging machines as ultimately creepy when they take a human form.

Admit… you would not mind losing to a chess computer. Chess computers are built to outperform and outsmart us. But, imagine a chess computer looking like this?

What people find most disturbing in Watson is not that it (he?) has tremendous linguo-analytic capabilities, parallel thinking power and advanced AI logarithms… but that it speaks. People would prefer it to just dot words on a screen.

Humanity my age and older does not want to see computers and interfaces become human, in any way. Younger generations however are completely fine with human/machine interaction that is stooled on humanoid processes. Computers with a human voice, looking like us – be it on screen or in real life- do not disturb the gen-Y’ers in any way. Can you imagine a Google interface that looks and talks like Hale Berry? We are not that far off! Watch how Kate interacts with virtual Milo in this Microsoft Kinect demo:

YouTube Preview Image

David Levy, PhD, goes further by showing that psychology has identified roughly a dozen basic reasons why people fall in love, “and almost all of them could apply to human-robot relationships. For instance, one thing that prompts people to fall in love are similarities in personality and knowledge, and all of this is programmable. Another reason people are more likely to fall in love is if they know the other person likes them, and that’s programmable too.” Levy predicts that robots will become so human-like in appearance, function and personality that many people will fall in love with them

Maybe that is why grumpy old men like me do not want our gear to look too human. A robot looking like George Clooney, with my sparkling personality, and IBM Watson smart would simply be too hard to compete with! 😉

And what if you really could ASK Google?

As I predicted, IBM’s Watson super computer configuration won a game of Jeopardy from some very smart humans on US national television. A bit humiliating for us, the human race, but on the other hand, we did design the thinking machine… so our pride should not take too big a hit.

YouTube Preview Image

Fascinating on Watson winning this brainy contest about hands down, is the machines striking ability to deal with complicated linguistic clues: Watson can process natural language with surprising ease, and is even able to filter and crawl through complicated things as irony and sarcasm, and cynic one-liners (that’s more than some humans I know :-)).

Apart from winning primetime TV quizzes, Watson and a comparable research logarithm Wolfram|Alpha open new horizons that are way more down to earth than thinking machines taking over the planet. “Search” for instance. The best search engines available to the public (Google, Bing, etc) are based on keyword driven search. You type in a series of keywords, and the search logarithm gives you a series of links where those keywords are found. Mind you: neither Google or Bing answer your question, give you an answer to what you want to know. It gives you a suggestion of pages where your keywords can be found. Often, just finding a page where the keywords are featured enables us, humans, to browse ourselves through the content of the page, and find the answer of our question right there. We all got good at this keyword based content suggestion. We even think it gives us answers. But it is not ;-).

Using revolutionary natural language driven concepts like Watson and  Wolfram|Alpha, the game will soon change. With their ability to cope with the eccentrics of our language, we will be able to ask these search engines a question and they will provide us with an aswer. As human-to-intelligence interface this is a giant leap forward, because we’re not taking shortcuts anymore but will be able to elevate “search” to its fullest extent: the ability to ask the web a question. Watch the video: Can you feel I’m excited? 😉

Connected Brains, artificial intelligence… and you

We’re getting connected, anywhere, anytime. For some this is a scary thought, for some everyday reality. As miniaturization enables computing and transmitting devices to get smaller and smaller, we can make more little things smarter. Wireless connectivity, Bluetooth Low Energy technology, NFC, and high performance location are being built in a plethora of devices. Even consumer goods packaging gets its smart chips incorporated.

All of these devices are being hooked up to the internet, creating de facto an internet of things. Machine to machine data enables you to use your credit cards, helps road side assistance, combats shop lifting and enhances home automation and security (to name a few). Connected sensor devices are playing an increasingly dominant role in fast moving areas such as home care, patient follow-up and healthcare in general.  Small armies of sensors, analyzers and detectors connect through the local area networks, or through the all-mighty smart phones to the internet.

This feeds the web and its connected servers a vast amount of crunchable data. And our artificial brothers are getting better and better in smartly analyzing… and dealing with whatever is thrown their way. Artificial Intelligence, it is a dirty word…

But let’s face the facts: intelligent, interconnected machines help us evolve faster as a species. Much like the control of fire and the inventions of the wheel, steel, writing, electricity and social security the connected computer power propels us to a next level.

Stand-alone computers have beaten the world-chess-champions decades ago…  Google connected cars have driven countless hours through our towns…without human assistance, solely steered by web connected computers. Now IBM’s “Watson” is beating the heck out of human Jeopardy! champions. Watson runs on a small independent network of 90 servers, and is able to deal with complicated language challenges and a need to flawlessly understand complexities of humor, dark puns, vague metaphors and other subtleties that were usually guarded for the human brain’s pattern recognition and analyzing capabilities. Imagine what a Watson-like setup can do when linked to the full power of the millions of servers on the web.

Ray Kurzweil sees in his “fifth Epoch” a merger of technology and Human Intelligence, a moment where the methods of biology will be integrated into the human technology base. I do not know about that, but I do know that the power of the internet of things, enhanced with a sentiment loaded social and semantic web is emerging fast, in countless small devices very near  you…

And that, is not necessarily a bad thing ;-)…. or is it?