Why human intelligence shouldn’t try to ‘beat’ artificial intelligence
Everyone (or almost) knows the story of Alphago, the artificial intelligence of Google that beat the world champion of Go. There has been a lot of commentary on what is now considered a major turning point. from the history of artificial intelligence.
One could say of this story that it is that of the war of artificial intelligence against human intelligence. And that in this battle the robot won. But we could also see it differently. It’s a door. But it’s a pretty scary door because, contrary to what some might think, it’s not so much a defeat of human intelligence as it is a lesson in creativity and innovation. Yet it was not the Alphago team that caused this heartbreaking rift, it was Alphago himself. Or almost. And it is in this “almost” that we have something to learn from this “drama”. Something very important.
If you haven’t seen it yet the film “Alphago”, I invite you to see it (it is available on Netflix). This documentary is absolutely stunning. We tremble from start to finish. We cry sometimes. We try to understand what is happening in front of our eyes. What this story is trying to tell us. And what is fascinating is that, over the course of the five trials that will see the human and the robot face each other, no one really understands what is happening. Not even the robot.
During the first half of the story, Alphago is content to win. Lee Sedol, reigning world champion, is completely destabilized. He sometimes feels like he’s playing against a mirror of himself, or all the Go players before him. He begins to doubt and change his tactics. The third round is a massacre. Lee Sedol is playing a game he is not used to playing, more aggressive. He absolutely tries to win and loses his means. He forgets the way to concentrate on the result. He trembles, scratches the back of his neck, runs his hand forty times through his hair. At the end of the first 3 rounds, he lost 3 times, he is devastated. By his defeat. But also because he has lost confidence in himself.
The fourth round is a turning point. Lee Sedol decides to remain himself. He’s looking for the flaw in AlphaGo. He goes beyond his own game. At one point, he plays a move, move 78, which commentators immediately qualify as a “divine” move. There was a 1 in 10,000 chance this would be played. Alphago takes a long time to respond. He doesn’t find a solution and starts playing irrational. Little by little his game is breaking up. He eventually gave up.
And the machine gave us a lesson in creativity …
Then comes the last round. A breaking point. Human and machine have both succeeded in breaking the other. Alphago will then completely change tactics. And start playing in an incomprehensible way at first. Throughout the game, we see the Google team trying to understand the attitude of artificial intelligence. Victory statistics continue to drop. The robot carries on the so-called “neglected” blows, unnecessary blows which make it lose the advantage on the domination of the plateau. We try to capture a smile on Lee Sedol’s face, but the champion remains incredibly focused. Something is happening. His opponent’s moves defy all statistical calculations. Then, little by little, without anyone understanding anything more about what is unfolding before their eyes, not even the engineers of Alphago, the probabilities of victory go up. In the end, Lee Sedol gives up. Alphago wins by a small point and a half.
You might think the human champion is devastated, as in the first three rounds. Quite the contrary. This time he played pretty well, he remained himself. But something completely unimaginable happened to him. For millennia, Go players have used the score as an indicator of the odds of winning. The more the score advances during the game, the more self-confidence the player gains. This time, AlphaGo did not play the score, he chained the “strange” blows which seemed to be errors, dispersed him and weakened his domination. Alphago seems to be telling us: “No matter how much I win, I only need to win one point, I don’t need to win all of these territories if they are useless to me for the final victory.”
What’s amazing about this story isn’t the machine’s victory over the human, it’s the way humans reacted: Alphago taught us a lesson. Lee Sedol explains it very well: “What surprised me the most is that AlphaGo has shown us that the original shots for humans are, in fact, very conventional. This reveals a new model for the game of Go. ”
The machine, here, provided the trigger for a breakthrough, and jostled humans with its ability to be, if not creative, at least “unconventional”. Disruptive. Innovative.
But the lesson that can be learned from all this is that this attitude is the result of a story, which Lee Sedol and Alphago wrote together. At the beginning the machine breaks the human, then the human regains its coherence, and then the machine finds another way and teaches us something about ourselves.
We can approach this war between human intelligence and artificial intelligence in terms of victory or defeat, but we can also think of it as a story that prompts us to reinvent ourselves.
Alphago’s victory is actually a more global victory. It is also a victory for Lee Sedol. She helped the champion better understand the complexity of the game of Go and introduced him to new paths. And this “innovation” results from a shared history. AlphaGo sparked a new attitude in Lee, which pushed the machine to its limits and caused him to break the rules. In the end, everyone has learned something. Lee Sedol improved by confronting the machine in a very rational and emotional way. Very interior, after all.
Lee sedol: “I’m grateful and I think I found my reason for playing Go”. The story has only just begun.
Fanu Hui, European Go Champion, trained with AlphaGo, and never stopped growing: “While playing against Alphago I saw a form of beauty behind the game of Go. I saw the world differently. ”
There are a lot of lessons to be learned from this story, about how we should approach our future with artificial intelligence.
The potential of technology is infinite, but infinity is not enough to make a truth, it is only a field of possibilities. By developing according to our reactions, technology never ceases to question us like a mirror out of ourselves. A revealer that would constantly push back our beliefs about who we are and how we operate and reinvent ourselves.
By confronting it, we can certainly approach this war between human intelligence and artificial intelligence in terms of victory or defeat, but above all we should think of it as a story. And stories are made to make us grow.
Benoit Raphael is an expert in digital and media innovation, blogger and entrepreneur.
He is at the origin of many successful media on the Internet: Le Post.fr (Le Monde group), Le Plus de l’Obs, Le Lab d’Europe 1.
Benoît is also co-founder of Trendsboard and robot media Flint.