Watson whips wetware; machines rule Jeopardy

IBM's Watson supercomputer proves faster on its feet than reigning human champions.
Written by Joe McKendrick, Contributing Writer

In the latest skirmish between humans and machines, the machines won.  Won to the tune of $1 million, to be exact.

Dave Ferrucci, IBM scientist and Watson project director is seen with the Watson Question Answering Computing System – File / Photo: IBM

In the finale of a three-day matchup between Ken Jennings and Brad Rutter, the two reigning human champions of the game show series, and IBM's Watson supercomputer, Watson took a leading position and stayed there.

In the end, Watson prevailed with the correct question to the answer: "William Wilkinson’s 'An Account of the Principalities of Wallachia and Moldavia' inspired this author’s most famous novel." Watson posed the correct question of "Who is Bram Stoker?"

Watson racked up $77,147, compared to Jennings' $24,000 and Rutters' $21,600. As a result, Watson took the show's $1 million grand prize, which will be donated to charity.

What have we learned here?  We know that computers are now capable of of deciphering shades of meaning, innuendos, and cultural associations in human queries.

There were some quirks in the series, such as Watson responding to "US Cities" topic of "Its largest airport was named for a World War II hero; its second largest, for a World War II battle" with "What is Toronto???" (Correct question should be "What is Chicago?")  But overall, not a bad performance.

The Jeopardy victory for the supercomputer doesn't portend the beginning of a sinister rise of Hal-like machines to dominate the world. It means artificial intelligence has matured to a point where it can address queries in areas relating to business intelligence, medicine, engineering, and more. Each word in these disciplines has its own sets of words with special meanings.  Imagine a professional in the middle of a situation being able to make a natural-language query over a mobile device and getting a quick, accurate response in its proper context.

SmartPlanet colleague Janet Fang reveals, in fact, that Watson may soon go to work assisting with diagnoses in healthcare establishments.

This post was originally published on Smartplanet.com

Editorial standards