This week on ted Talk Tuesdays I’m going to recap a talk from someone who spent more than a significant amount of time on my favorite game show! Ken Jennings from Jeopardy! As I was perusing various Ted Talks to potentially review this week I was naturally drawn to Jennings’ both because I’m a big fan of Jeopardy! And because the topic he was discussing is of crucial importance to our society. His talk was titled “Watson, Jeopardy, and Me, the obsolete know-it-all.” I know that doesn’t sound like the crucially important topic as I’ve just described it but hear me out. Jennings goes on to discuss the prevalence of AI in our society, the feeling people can succumb to when they’re replaced by it, and where our world could be heading in the future. Overall he takes a light-hearted approach to the topic walking a fine line between “doom and gloom” and an overly rose-colored approach, ultimately focusing on the fact that knowledge really is power.
Jennings begins his talk by providing some background and context with his own experiences growing up. As a kid he lived in Korea because of his father’s work (job not mentioned) and he only had access to one TV channel that was broadcast in English. Fortunate for Jennings this channel almost exclusively showed game shows. So while other kids were watching cartoons, Jennings grew up learning nothing but seemingly useless trivia facts. It’s in this part of his presentation that Jennings delivers one of my favorite lines from his talk when he says, “Knowledge truly is power, the right fact in the right place does indeed have power.” Obviously this builds on an already famous quote, but I just really liked the middle part of the quote about the right time and place for a fact. As they say, timing is everything!
Anyway, back to Jennings’ story. He quickly fast forwards from childhood anecdotes to the time after his extensive run on Jeopardy, and he explains that around this time was when he first heard about potentially facing a computer in a game of Jeopardy. As he describes it, his initial thoughts were that humans would dominate the computer because a computer simply couldn’t think analytically enough to decipher clues that can be as complex as Jeopardy’s. Jennings recalled that he initially received the call from IBM and Jeopardy representatives some time before the actual competition, and in the time leading up to the competition he received updates from IBM on how Watson was progressing. At this point he describes getting worried as updates continued to show vast progress in Watson’s ability to catch up to him and other Jeopardy contestants on a variety of metrics.
Well, as we all know Watson’s progress ultimately led to him dominating Jennings and another human competitor in the Jeopardy game on national TV. Personally, I remember being in college and actually being pumped to watch this “man v. machine” showdown and actively rooting for Jennings to crush Watson. If I remember correctly it was a two night event, and while Watson struggled early, by the end of the second night he had pulled well ahead. Jennings goes on to explain that losing to Watson left him feeling obsolete, similar to how Detroit factory workers must have felt in the 1980s seeing robots come in who could do their jobs faster and cheaper.
It is here where Jennings really starts to get to the meat of his presentation. Is this level of AI ultimately a good thing or a bad thing for humanity? And where does the ability to recall specific facts and use actual knowledge still fit in society? He uses an interesting example here regarding GPS. As Jennings explains, humans have shown actually decreased spatial functions in certain parts of the brain which has been attributed to using GPS. Essentially, since we have directional information available to us with a device, we no longer have as much use for that part of our brain, so that part of our brain actually diminishes. If this works for GPS, what is to say that widespread AI won’t cause other areas of our brains to diminish? By the same token, Jennings explains how knowledge, facts, and the ability to understand context are still crucially important in social situations where computerized knowledge is not always available.
Essentially Jennings boils the issue down in this way: with the presence of AI an almost certainty, humans can choose one of two paths, either we can allow ourselves to be cut off from knowledge and facts (thus relying solely on AI and computers) or we can use a combination of our own brains and technology so that human knowledge and AI can operate in harmony.
I have also thought about this topic quite a bit over the past few months, especially with the growing story about Robot Sophia, a humanoid, nearly full functioning robot that looks and thinks like a human. I’ve always wondered why so much effort is put into AI and have rarely seen an appropriate use for it in everyday life. Plus, you know, I’ve seen enough movies to know that the AI robots are just going to try and takeover the world anyway. And then Will Smith will have to defend humanity by himself! But seriously, I wonder where this is all going and whether or not the powers that be will ever decide enough is enough. But until then, at least we still have guys like Ken Jennings around.