To understand the future of AI, see what happened in chess.

Predictions of AI doom have been proven wrong before.

When world chess champion Garry Kasparov accepted the challenge of playing against IBM’s Deep Blue supercomputer in the 90s, many in the chess community were surprised. They feared that a machine victory would deal a fatal blow to their 1,500-year-old sport.

At the time, computer technology was expanding globally and many chess veterans argued that if the best human player was shown to be inferior to a machine, interest in the game would fade and sponsorship would cease. He urged Kasparov not to participate in the spectacle.

Against this advice, Kasparov was finally defeated by a computer in 1997.

Yet, today, chess is more popular than ever. It continues to thrive in the machine age with top players emerging from countries that didn’t even have much interest in chess at the time. And sponsors have never been more interested in investing in human athletes. The reigning world chess champion has amassed a fortune playing the game, estimated to be in the tens of millions of dollars.

Understanding how chess evolved, not just in spite of machines but because of them, can help us understand how AI is changing our wider world today as it accelerates human capabilities in other fields. learns to improve, which is much more productive than a game of chess.

Like all technology, AI is not inherently good or bad. Nor is it neutral. This will have a profound impact in a complex number of ways. Many of the concerns about the dangers of AI today are entirely valid, but it is even more important that we understand and embrace the positive potential of AI in order to drive its wider development in that direction. could

As it happens, the fears of jaded chess veterans in the 90s about the pervasive effects of the machine age were not entirely misplaced, at least in terms of their own narrow interests.

In the years following Deep Blue’s victory, the chess rankings saw rapid upheaval and long-time leaders suddenly took over.

Before the proliferation of computers, you needed more than skill to succeed at chess. You should be among the lucky few whose talents will be recognized and nurtured by the existing experts of the game with whom you can hone your skills.

Yet, suddenly, computers could enable endless play against experts worldwide, both virtual and online. Not only this, budding new players can use the internet to follow games that they would otherwise never be able to attend. After that they could also get quick and detailed analysis.

And many new tactics need to be analyzed. Because computers don’t just learn to play well. They played differently and shaped the human game in the process. Despite centuries of human play, computers can still invent new strategies and methods by crunching vast amounts of data about the gameplay they were using. It encouraged human players to rethink and adapt the way they play with each other.

Machines have not replaced humans in chess. In professional sports, humans who accepted machines replaced humans who did not. But the overall effect was a huge net positive.

The same pattern was repeated recently in the even older and more complex game of Go, when Google Deep Mind’s AlphaGo swept aside some of the game’s top players.

Starting in 2016, the first Go player to lose to a machine was European Go champion Phan Hua. Still, his ranking improved significantly after that, because he helped AlphaGo see the game differently. After being stunned was the popular Korean player Lee Sedol, who later said that the computer’s style was so different that it made him realize he needed to study more. Finally, Ke Jie, the world’s top-ranked Go player in 2017, suffered defeat. All this was too much for frustrated Chinese state censors, who banned further coverage of AlphaGo.

As DeepMind continued to develop AlphaGo, it first taught top human Go players similar opening moves but eventually rejected them in favor of its own entirely new opening strategies.

At the same time, human Go players were given access to the AI ​​data it was crunching and used it to develop new strategies to play against other humans. The change was quantifiable. As reported by Scientific American, analysis of matches between human players showed relatively stagnant gameplay in previous decades, but this suddenly changed in the years following AlphaGo’s appearance, with both games being played first. With new tricks and high judgment standards.

It’s understandable if Garry Kasparov himself harbors resentment against AI.

Instead, over the decades since, Kasparov has become more passionate about how machines have improved and democratized human play, as he writes in his book, Deep Thinking.

Kasparov now also sees parallels with the positive potential that AI can have in our wider world if it is developed properly and led by independent nations to set the rules in everyone’s interest.

Beyond the spectacle of the media, classic games have served as interesting test cases for the development of AI since ancient machines were first programmed to defeat humans. However, the real world is more complex, and AI has no inherent desire to play against us.

Take negotiation, for example, as a fundamental human skill needed to foster cooperation and create value on which the global economy depends. Sadly, humans are pretty bad at this. Not only do we have all kinds of cognitive biases that cloud our better judgment, but there is too much data around us to analyze every possible deal.

Just as chess and Go moves increase exponentially in complexity, a fairly simple contract negotiation with a limited number of conditions can have millions of possible outcomes. So it’s no surprise that AI has now overtaken human capabilities here as well, with the emerging field of autonomous negotiation.

Again though, the way AI plays this ‘game’ is quite different from humans. Humans have a tendency to view negotiation as a zero-sum game with a winner and a loser. But AI can be tasked with winning by extracting the most value from a deal so it can ensure that both parties create more value together. In mathematics, this is called Pareto efficiency.

In other words, it makes for a bigger pie to get a bigger piece of.

It is beyond our concept of a win-win outcome. Take a classic story often told in negotiation training. Two children are fighting over who gets the last orange. Parents ask to divide it in half. This can be classified as a win. But what if one child wants to eat orange slices and another needs orange peel for a cake recipe? If this information had been known at the time of negotiating the orange, both parties could have gotten everything they wanted.

In the real world, these kinds of missed opportunities for better deals exist all around us in more complex ways, so if AI can use more information to negotiate autonomously at scale, it’s just another way to AI can significantly increase the world’s GDP and provide a net positive to humanity while teaching us new skills in the process.

According to a recent IMF study, 40% of jobs globally will be affected by AI. This rises to 60 percent in developed economies. Behind the headlines though, the data breaks down to roughly half of the affected jobs being at risk of being replaced by AI while the other half will likely be replaced by AI working in a complimentary manner. And that’s before we consider the new jobs that will be created by AI, as has happened with previous technologies when humans unlocked value that was never possible before.

When we walk into an elevator, we no longer expect a human operator. When we board an airplane, we expect a pilot but hope that their autopilot is working as well. And any busy elevator or airplane these days will be filled with people doing things that didn’t exist when these things were invented.

What’s happening with AI is really a continuation of long-term trends at a much faster pace, which is why it helps to look back at the past to understand where it’s going.

As in chess before, the AI ​​doesn’t necessarily take over your work. It is humans using AI that will take jobs from humans not using AI. But if we recognize emerging opportunities and begin to adapt to them, there is potential for enormous net gain for everyone.

Leave a Comment