It’s the world’s oldest, continuously played board game. The ancient Chinese game of Go is featured in Oliver Roeder’s 2022 book, “Seven Games: A Human History,” which I’m soon to finish.
Go seems simple on the surface, but it’s actually mind-numbingly complex. Chess has 10-to-the-120th power number of possible board configurations. This measure of game tree complexity is called a Shannon Number after the poly-mathematician Claude Shannon. By comparison, Go measures a whopping 10-to-the-360th power possible configurations. It is said that this is greater than the sum of atoms in the observable universe.
Go clearly takes genius to win at the highest levels. It’s a game that has long relied on intellect but also on intuition, creativity, and even quirkiness. When asked why they made a particular move, top-rung Go competitors have been known to say something like, “I don’t know; it just felt right." Maybe that's one reason why the ancient Chinese considered Go to be among the four noble arts including calligraphy, painting, and the zither.
IBM’s Deep Blue beat world chess champion Gary Kasparov in a shocking 1997 match, an early milestone in AI history. That event is referenced in Roeder’s book and featured in the 2003 Vikram Jayanti documentary “Game Over: Kasparov and the Machine” available on YouTube. Given its infinite complexity and intuitive orientation, however, most observers thought it could take an another 30-40 years for AI to best a top Go player. Nope! The 2017 Greg Kohs “AlphaGo” documentary, also on YouTube, showcases the 2016 defeat of 18-time world champion Lee Sedol at the hands of Google DeepMind’s AlphaGo. Sedol said at the time, “I thought AlphaGo was based on probability calculation and that it was merely a machine. But when I saw (a particular, very human) move, I changed my mind. Surely, AlphaGo is creative.”
AlphaGo’s deep neural networks learned and adjusted throughout the five-game match, making Deep Blue’s AI prowess almost 20 years earlier seem primitive by comparison. Frightening stuff, then and now as AI capabilities increase exponentially. And that’s among AI’s greatest challenges. Actually, it’s among society’s greatest challenges. The core concern is that the immense responsibility for AI rests in the hands of tech bros and their money-changer finance bros. (Not many women to be found in either the Deep Blue or AlphaGo documentaries, though one hopes that situation has improved since then.)
The tech and finance tribes are obviously needed to develop and grow AI. Just as many of these guys surged ahead developing and monetizing social media without reckoning with its dreadful consequences, however, they’ll certainly do likewise with AI. It’ll be all muscular IQ without much deliberative EQ, and we will all pay the price accordingly. That’s why these tech titans need the perspectives of sociologists, psychologists, behavioral economists, and even artists to place their work in context and understand its vast implications. The politicians certainly have no understanding of any of this. Plus, too many of them are owned by the broligarchy anyway.
It’s one thing to appreciate the history of AI in beating humans in Go, chess, and whatever comes next. It’s one thing to be generally supportive of AI advancement buoyed, we hope, by wisdom and common sense. In the 10 years since AlphaGo beat Lee Sedol, however, we now find AI in every facet of our lives.
Go is a game of intellect infused with vast wisdom. AI is a game of intellect that will likely lack the wisdom needed to develop and utilize it properly. Go figure!
Image courtesy of The Conversation