Can AI Love?

According to Merriam-Webster, artificial intelligence is "a branch of computer science dealing with the simulation of intelligent behavior in computers." Love is defined as "a strong affection for another arising out of kinship or personal ties."

It is difficult today to bypass the raging, Manichean debate in the tech and business communities about the role artificial intelligence (AI) will play in our economy and our society. Will this emerging technology become some kind of terminator, killing all of our jobs? Or will it emerge with a more theological, liberating approach to the human condition? Lost in the conversation is whether it will be a technology we like or possibly even love.

Since Alan Turning, computer systems have been built to automate and accelerate information processing. The internet ended much of the friction to obtaining or contributing to information flow. For the past 50 years, we have designed systems that either augment or replace human middleware in data processing (what we called IT in the good old days). Information technology revolutionized our economies, government and lives but missed the poetry, art and emotion that makes being human worthwhile. You can create the sequel for the cartoon Shrek in half the time but cannot substitute for the brilliant story by William Steig or Mike Myers’s voice.

Thinking about a world where our computer overseers will increasingly be more in charge of us themselves, we should be considering two design questions for AI systems that we may have given short shrift to during the 1.0 computing revolution.

Will AI Reduce The Amount Of Data We Are Trying To Sort Through On A Daily Basis?

One of the greatest promises of AI (a virtual layup) will be the reduction of the great garbage patch of the internet that us mortals find increasingly difficult to navigate. Today, we create over 50,000 GB of data per second, which is stored somewhere on our organizational, personal and social networks.

Knowledge work requires making sense of increasing amounts of data. Shelly Turkle spent the last few decades digging into why more time in front of screens has not made us happier and increases our isolation as people.

In 1996, I was fortunate enough to be working at IBM and observed firsthand the famous chess match between Deep Blue and Gary Kasparov. Ultimately, the computer prevailed, but not due to being a better chess player than the Russian master; rather, it was because the computer wore the human down. Not surprisingly, the advertisements plastered across New York had a photo of Kasparov and the tagline “How Do You Make A Computer Blink?”

Can AI Bring More Human Connections Alongside More Powerful Processing?

One of the fundamental building blocks of AI is machine learning -- the ability for systems to learn from massive amounts of data and identify patterns, allowing systems to autonomously make decisions with minimal or no human intervention.

I am thrilled that computer vision and machine learning can program robots to undertake dangerous or physically exhausting tasks on an assembly line. Can AI better understand and filter what simply is created to anger us rather than bring us together? It would be great if AI systems reduce the acrimony created on social networks, like a considerate friend pointing out when we should not allow someone to annoy us with a warm smile and a pat on the shoulder.

How will AI systems deal with a major provocation? Could software manage a repeat of the Cuban Missile Crisis? Anyone applying AI to wargaming and national defense would do well to read Robert Kennedy’s Thirteen Days and understand what it required to navigate one of the most dangerous and complicated crises of our lifetimes. We will need AI systems that know our adversaries may love their children as much as we do. We need to be able to program context.

AI holds enormous promise to dramatically impact our economy and our society. The technology can be the fulcrum of efficiently creating tailored services for individuals such as personalized health choices and making energy-using systems more efficient to help us address climate change. That is a discussion for another day.

If we are going to interact more often with these intelligent systems along this journey, one hopes they will be like the operating system in the movie Her -- a digital voice we could learn to love and to love us back. If the human is a bridge too far, we can design AI systems to like the staff at Nordstrom’s: always helpful and leaving one in a better mood after interacting with them.