It’s elementary, Watson
Next week, IBM’s Watson will compete on Jeopardy against two of the show’s top contestants. This article explains the contest, and how Watson works (also noting that Watson won a practice round), which leaves me free to share my thoughts…
I think Watson will win on February 14, but I don’t believe that in and of itself this will be innovative or revolutionary—in the end, it is only the latest application of bruteforce to a problem that human beings solve differently (like Big Blue’s win over chess champion Gary Kasparov in 1997). It is by now well known that in the game of man vs. computer, the computer can win. And this will be only the latest proof. Even if a quiz show like Jeopardy looks more difficult than a game of chess for a computer, the reality is that fundamentally, it must have a vast “memory” and the possibility to process information rapidly. Two characteristics in which computers notoriously excel.
Having said that, I don’t want to diminish the fantastic job of all the people who have contributed to the creation and development of Watson, because they have surely advanced this field, especially in the area of automatic language processing. And, it’s something that is positive for everyone involved (and it will especially be well-deserved publicity for IBM). But this victory will not bring us any closer to a computer that is able to learn and reason in a way that is truly comparable with a human being (no matter what everyone says after next week). In the end, this is good news—I don’t mind at all that humans will continue to be the only real form of intelligence we know.