Rep. Rush Holt of New Jersey, former Assistant Director of the Princeton Plasma Physics Laboratory, was a Jeopardy winner 35 years ago when the show was in its initial run. Watson was pitted against several congressmen in promotional match in a Washington hotel, but Holt, who holds a PhD in Physics, was the only winner. The exhibition was a chance for IBM to promote potential government applications of their technology, and for politicians to promote the value of science education.
Holt was my Representative when I lived in his district. No, I didn't vote for him, but he always seemed like a decent guy, and we used to live in the same town. He's also the answer the Jeopardy question: "Who is the only Quaker currently serving in Congress?" (Quick: Which U.S. Presidents were Quakers?)
I promise this is my last Watson post. Until the next one.
In my post about the Jeopardy-winning supercomputer Watson, I remarked that Chess Grandmaster Garry Kasparov "wouldn't last five minutes against Ken Jennings." That was no slight against Kasparov, who may well be the finest player in the entire history of Chess, and is proving to be a charismatic political figure in post-Communist Russia. It's just that the skill set and knowledge base needed to succeed at Jeopardy is far more diverse and complex than that required to succeed at Chess.
Although Kasparov won his first match against IBM's Chess-playing computer Deep Blue, he lost the rematch, and didn't handle the loss very well. Obviously, he's thought quite a lot about the man-versus-machine contest in the ensuing years, and he offers some interesting comments about Watson in The Atlantic:
My concern about its utility, and I read they would like it to answer medical questions, is that Watson's performance reminded me of chess computers. They play fantastically well in maybe 90% of positions, but there is a selection of positions they do not understand at all. Worse, by definition they do not understand what they do not understand and so cannot avoid them. A strong human Jeopardy! player, or a human doctor, may get the answer wrong, but he is unlikely to make a huge blunder or category error—at least not without being aware of his own doubts. We are also good at judging our own level of certainty. A computer can simulate this by an artificial confidence measurement, but I would not like to be the patient who discovers the medical equivalent of answering "Toronto" in the "US Cities" category, as Watson did.
Ken Jennings has written a wonderful piece for Slate about his experience playing against (and losing to) IBM's Jeopardy-playing supercomputer Watson.
But there's no shame in losing to silicon, I thought to myself as I greeted the (suddenly friendlier) team of IBM engineers after the match. After all, I don't have 2,880 processor cores and 15 terabytes of reference works at my disposal—nor can I buzz in with perfect timing whenever I know an answer. My puny human brain, just a few bucks worth of water, salts, and proteins, hung in there just fine against a jillion-dollar supercomputer.
I remember writing about Deep Blue and what its victory against Chess Grand Master Garry Kasparov meant for AI, and Watson is exponentially more impressive than Deep Blue. Kasparov wouldn't last five minutes against Ken Jennings. The skill set and programming necessary to master Jeopardy is far more subtle and complex and that required to master Chess, at least from a pure computer AI perspective.
The Singularity is Not Near
Impressive as it is, I still balk at the notion that this is the first step towards The Singularity, for the simple reason that I believe The Singularity is the greatest farrago of nonsense since the Population Bomb. Time Magazine recently ran a long, hilarious piece on The Singularity. I laughed all the way until the end, and then I realized that it wasn't satire, thus reminding me yet again that Time isn't so much a news magazine as it is a source of absurd trend-chasing and juvenile credulity wrapped in a squishy coating of bias.
Let me state this as clearly as possible: The Singularity--the moment at which computers essentially become sentient--is the greatest load of BS to come down the pike in many generations. That intelligent people believe it's All Really True! doesn't make it so. Aristotle, the most intelligent man of his age or almost any other, was certain that the blood cooled the brain and the liver was the true seat of human life. At one point, all the brightest minds in the world were utterly convinced in the truth of geocentricism. In our own day, powerful interests continue to flog the big lie of anthropogenic global warming.
So, I don't care one bit that some of "the best and brightest" have signed on to Ray Kurzweil's loony idea that one day (in the next 40 years!) we will reverse engineer the human brain, dump our entire consciousness into a computer, and live forever. It is more likely that I could swallow my own torso than it is for a machine to replicate the human brain. I am not saying "maybe," or "perhaps eventually," or "given the right scientific developments": I am saying never, ever, ever: not in 40 years, not in a 1000.
This is all just part of the progressive's favorite delusion: that of the continued upward development of humanity, hand in hand with glorious technical achievement. This is utilitarian nonsense. First off, it's a failure to understand the practical reality of Moore's Law. The number of transistors on an integrated circuit may in fact keep doubling every two years until infinity (although I highly doubt it, and there is already some evidence that this will eventually peak and then decline), but that does not correlate to a matched increase in processing speeds.
Second, it ignores the immense complexity of the human brain, and just how much of its function remains poorly understood if not downright inexplicable. Even the common process of medicating for mental illness remains a matter of trial and error because we simply don't know how the brain does much of what it does. The notion that the myriad complexities of the human psyche can be wholly reproduced if we just have enough processing power is madness.
Finally, there is the part many scientists leave out due to their own personal bias. I hate to break it to the atheists out there, but the human soul exists. We are not merely clever meat. We are simultaneously physical and metaphysical beings. The notion that we are the product of random chance, originating from nothing and returning to nothing, isn't even good nonsense. Ex nihilo nihil fit. The rational soul is transcendent, and will forever remain a mystery beyond the ability of science to grasp.
The Simulated Brain
Without question, we will see continued progress in artificial intelligence. Watson impresses not because of its 15 terabyte store of knowledge, but because of its ability to parse the English language, including puns, word play, allusions, and other very human subtleties of speech. Its knowledge set did not impress me: that's simply raw processing muscle. It's language ability, however, is frankly amazing.
Yet with all that, IBM couldn't create a truly human voice. Watson won Jeopardy, but he still would have failed the Turing test. No one listening to its answers would be in doubt that they were hearing a computer. Watson is the Deep Blue of this generation, and as such is a fascinating and important milestone in the development of artificial intelligence. Where it will lead, I have no idea. Deep Blue was dismembered and mothballed, with a chunk of it ultimately winding up in the Smithsonian.
And it's important--vitally important--to remember one thing. Watson did not create itself. It was created by a team of brilliant and dedicated people, pouring human knowledge into its brain and teaching it to think and learn in human ways. A machine can only be a simulation of the human brain, and no machine will ever be anything more than that. It will forever remain a mere shadow of the real thing.
