Monday night, February 14^th 2011, we saw IBM’s natural language processing computer Watson go up against two Jeopardy! champs and whomp them solidly—until the final round when some sort of brain freeze hit the mechanical brains powerhouse and it mixed up the United States for Canada. However, that said, it proved a surprisingly amazing show against the reining champions and did in fact win the day.

IBM’s Watson has been put to task before, by Greg Lindsay and he has spoken up on how he managed to beat the machine—3 times. In an article published at Fast Company, Greg goes over what he expected the machine’s primary weaknesses to be and how he managed to exploit them using a carefully crafted trivia stratagem to keep himself in the lead. Although, from the way it reads, just barely:

My strategy was simply to take Watson’s strengths away from him. Having no idea what those strengths were, however, I had to make several assumptions.

First, I assumed he’d be impossible to beat on the buzzer, which had never been my strong suit, anyway. Instead, I took a page from The Princess Bride (the book, not the movie), specifically Inigo Montoya’s duel against the Man in Black. As long as Montoya was able to keep the fight on rocky terrain, his defensive prowess awarded him the advantage. Once the Man in Black maneuvered him onto open ground, however, Montoya’s was overwhelmed by his speed. So it would go with Watson, I figured. Binary relationships–countries and their capitals, for instance–would be easy for him to figure out, and he would beat me to the buzz every time. So I had to steer him into categories full of what I called “semantic difficulty”–where the clues’ wordplay would trip him up. I would have to outthink him.

Second, I would need to find and win the Daily Doubles to deny Watson a coup de grace and to keep pace in what I figured would be a losing war of attrition. (This was based on personal experience–I had rallied from last place to win my first Jeopardy! match only after a Daily Double on the very last clue.)

Finally, I had to be in the lead heading into Final Jeopardy. If Watson could confidently decide on an answer in only three seconds, I shuddered to think how infallible he would be given all of thirty.

Watson uses a powerful data-mining engine to pull facts and figures out of textual sources. This gives the machine (casually referred to as “him” by most, so I’ll keep doing that) a giant advantage with clues that have direct semantic linkage to facts such a geography, names, places, things, et cetera. However, as Greg correctly assumed, Watson had trouble with cultural context and linguistic semantics that played on allegory, slang, and intracultural jargon. Things that we might take for granted in a conversation about comic books would probably wiff right over Watson’s head because it’s an inference that occurs in the relationship between minds and doesn’t rise directly from the text.

Also, huge kudos to Greg for The Princess Bride reference.

The funniest thing about the Valentine’s Day battle of wits against Watson is that he failed in almost the same way during Final Jeopardy now as he did then: for some reason Watson’s semantic-brain tends to go haywire on these questions. Where on the Valentine’s Day run he rendered “Toronto” from the answer, “It’s largest airport is named for a World War II hero; its second largest, for a World War II battle.” During one of Greg’s matches the proper question would have been, “Who is Albert Einstein?” from “The July 1, 1946, cover of Time magazine featured him with the caption, ‘All matter is speed and flame.’” ended with Watson deciding on things like “Top 100” and “David Koresh.”

Obviously, there’s some difficulties there that must still be sussed out, but overall we’re looking at possibly one of the best evocations of a smart agent yet developed. This has a multitude of implications for smart-systems in general for both business and social networking. I don’t mean that machines like Watson will need to pass the Turing Test; but with the advent of cloud-computing, the mobile revolution, and the opportunity for a machine like Watson to crunch through Big Data throughput we could see the rise of a service that provides extremely smart information filtering using the same technology.

While we won’t be seeing machines like Watson producing compelling reasons to see or not see a new movie (that seems to be the purview of crowd-sourcing) but it may help innovate how we think about vast amounts of information.