Teaching AI to think like humans may be the wrong path

A few months ago, an artificial intelligence called AlphaGo, designed by Google to play the ancient strategy game Go, soundly defeated the world’s best human player. It did so, as pointed out in the Wired article by Cade Metz in large part due to a series of moves that no human player would have ever considered.

For decades, scientists have struggled with the task of developing AI that thinks and learns like a human being. A big part of that challenge has been that we still understand so little about the way the human brain actually works. Relatively recent advances, such as functional magnetic resonance imaging (fMRI) have given us better insight, but teaching a computer to think like a human has eluded us, for the most part. One of the things that has always separated us from computers is our creativity, our ability to “think outside the box.”

The unprecedented moves made by AlphaGo, however, may suggest an opportunity for both AI and human development. One of the many promises of artificial intelligence has always been that computers have the capacity to retain and process data far faster than the human brain. So, the thinking goes (if you’ll pardon the pun), that if a computer could learn to think like a human, AI would have the potential to instantly solve problems that would take humans years or decades to muddle through. AlphaGo showed us that AI might not just be able to find the solution more quickly, but it may be able to find solutions that a human would never imagine.

With complex, seemingly insurmountable problems such as climate change looming, perhaps an AI can see a way out that a human could not. In the field of education, where teachers endeavor on a daily basis to present information in new ways, to reach every student, AI may have the ability to differentiate in novel ways that can benefit students and teachers alike.

The fact that AIs are besting the world’s greatest players at chess and Go signals, to some, the beginning of the end of the human mind’s reign at the top of the table. However, perhaps it merely suggests that we should stop trying to force computers to think and learn like us. Instead, we may have arrived at the day when we can begin to learn from each other.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s