Here’s why we shouldn’t fear artificial intelligence
The rise of artificial intelligence is both celebrated and feared. At once a testament to human ingenuity that promises to revolutionize the way we live and work, AI threatens to make untold numbers of jobs obsolete. PwC, for example, estimates nearly 40 per cent of jobs in the U.S. are at risk due to automation, especially in the transportation, manufacturing and retail industries. Earlier this year, senior deputy Bank of Canada governor Carolyn Wilkins warned of job losses and greater income inequality stemming from AI and robotics, though she expects these technological developments to eventually create more jobs than they replace. The disruption caused by AI has also been on the mind of Daniel Pink lately. As the author of six books on work and human behaviour, Pink was somewhat early to the discussion about the impact of automation on white-collar work. In 2006, he wrote A Whole New Mind: Why Right-Brainers Will Rule the Future, which outlined the skills workers need to stay employable in an increasingly outsourced and automated era. Canadian Business spoke to Pink, who will speak at the upcoming 2017 GREAT CEOs Speaker Series in Mississauga, ON about why he’s ultimately optimistic that adoption of AI will be positive for society and the future of work.
You recently gave a talk on “surviving” in the age of automation, which implies a threat. So what exactly is the threat?
My argument is we are in a shift from one economic era to another. I’m in my early 50s, and I grew up in the American midwest. People of my generation were raised to believe a certain set of rules about how to get ahead: Get good marks in school, go to university, and pursue a profession that’s going to give you some amount of economic security. In terms of skills and abilities, what employers wanted were logical, linear, spreadsheet abilities. Today, those skills are still absolutely necessary, but they’re no longer sufficient. Now it’s abilities that we’ve overlooked and undervalued that matter most: Empathy, design thinking, inventiveness, big picture thinking. There are some forces in the world tilting the skills in favour of these more artistic, empathic, big-picture skills.
Okay, what forces?
One is offshoring of white-collar work. In the short run it hasn’t had much of an impact, but in the long run it’s going to have a huge impact. Think about a place like India. If a tiny portion of India’s population reaches college educated, upper-middle class status, say 5 per cent—that’s like 65 million people. Think about that. That’s double the population of Canada, competing with Canadians for jobs. The other threat is automation. In the last generation, machines replaced our backs and our muscles. Now artificial intelligence and machine learning is replacing parts of our brains—the logical, linear part. You can go online and get an uncontested divorce for like 200 bucks, for example. If I go to a lawyer, it’s going to cost 2,000 dollars. And the lawyer isn’t doing anything with an uncontested divorce. So this nudges us into a fundamentally different economic era, where the central economic actors are not knowledge workers but creators and empathizers. People who can do stuff that’s hard to outsource, hard to automate and that augments machine intelligence rather than replaces it.
What kinds of jobs are you talking about?
When you talk about it at the level of professions, it’s really hard. In the same way that 15 years ago, we didn’t think there would be jobs for search engine optimizers or social media consultants. It’s more about the content of work. Let’s take something like a financial advisor. That person needs knowledge of asset allocation, but the truth of the matter is a lot of that kind of thing has become commoditized. If that financial advisor is going to survive, what is she going to do? She needs to have greater amounts of empathy, understand where her client is coming from. That person needs to be very good not at the reductive task of problem solving, but at the more creative, complex task of problem finding. Can you see around corners? Can you identify issues that your client doesn’t face now but might face in the future?
You’ve mentioned empathy a few times. Why will that be more important in terms of employment?
Machines have a very difficult time—actually, human beings are not that great at empathy to begin with. Machines are even worse. There’s some really good research out of Harvard showing that the fastest growing job categories in the U.S. required math skills and social skills. Jobs that require just math skills are easy to automate. They’re routine jobs. But the jobs that require math skills and social skills—communication, working with others, all of these things you’re supposed to learn in kindergarten, those end up being valuable because they’re very hard to reduce to code or assign on a spec sheet to people living overseas.
There’s a huge a focus on teaching kids to code at a very young age in North America. Is its importance overblown?
I don’t know, but learning how to code without social skills is not a good recipe. Number one, having good social skills helps you understand what people want and need. You could be an expert in coding something that is irrelevant to people’s lives. Second, in the design of most complex software, it’s a collaborative process. You have to be willing to work with other people, and have the give-and-take of saying, “My code isn’t as good as yours. Let’s use yours.”
There’s a lot of debate about how many jobs will be lost to AI, and the displacement it will cause. How concerned are you about that?
This is not the first transition we’ve made. The same thing happened when North America went from agriculture to industry. This is just another one of those transitions. But in every one of these transitions there’s been hew and cry about how jobs are going to be lost, there will be widespread unemployment, this scorched earth of despair, and it’s never come true. What it does signal to me is that it underestimates human ingenuity. When I was growing up in Ohio, there was talk of moving from a manufacturing economy to a services economy. My father was like, “That’s nonsense. We’re going to have an economy where everyone is running around giving each other haircuts? That’s not an economy.” He didn’t envision this whole world of e-commerce consultants, or Facebook consultants or even something like massage therapists. If past is prelude, it’s not going to be this widespread devastation that a lot of people predict. I don’t know for sure. My saying that represents a leap of faith. It’s not because I’m a crazy optimist; it’s just that’s how it’s usually been.
But surely there are industries and people who will lose as AI becomes prevalent.
Oh, there’s no question about that. There’s going to be huge dislocation. There was dislocation when we went from manufacturing to the information age. We did a terrible job dealing with it. That’s why you have cities that are completely hollowed out. We didn’t have the social safety net to keep people from falling into despair, and we didn’t have the mechanisms to help people retool. Here’s what we should be concerned about: The most common job in most states in the U.S. is truck driver. If you have self-driving cars, you’ll have a lot of people out of work. That is a huge issue. At some level, it’s a slow-moving crisis, and we don’t respond well to slow-moving crises. We need to look at this the same way we look at Hurricane Harvey or Irma. We say, “This is terrible, we can’t let people suffer,” and we respond immediately. But the possibility that we could have one million unemployed truck drivers in 10 years, we don’t approach those problems with the same sense of urgency.
So why are you still optimistic?
To continue with this example, the chances are that self-driving cars will lead to a whole array of industries and professions that you and I can’t envision right now. In general, it’s going to be a net positive to the economy. But there are going to be a lot of people suffering. And the United States has done a terrible job dealing with that. We’ll need a structure for continuous learning, to help people retool and re-skill to make their way.
Daniel Pink is speaking on The Cascade Effect: How Small Wins Can Transform Your Organization on October 19 at the Mississauga Convention Centre as part of the 2017 GREAT CEOs Speaker Series.