Ed Clark, chair of the Vector Institute, explains the opportunities and challenges of an AI driven world


More and more, artificial intelligence and machine learning are being woven into our lives. Fast cheap computing power, processing mounds of data and analyzing it to predict and solve problems.

AI is your Google search. It’s what you see in your social media feed. It can scan for tumours, discover new drugs, optimize heating and cooling. And yes, AI can eliminate jobs.

Toronto is the largest AI hub outside of China and Silicon Valley, according to Forbes magazine. That, in part, is due to the Vector Institute, an artificial intelligence research institute invested in deep and machine learning.

Right now, Vector, which is funded by the government and the private sector, is championing the use of AI to boost small and medium sized businesses.

I sat down with Vector’s chair Ed Clark, who is the former CEO of TD Bank, to talk about the opportunities and challenges of an AI driven world.

Howard Green: Great to see you again Ed. Thanks for making time.

Ed Clark, former CEO of TD Bank, is the chair of the Vector Institute for artificial intelligence in Toronto, which is the largest AI hub outside of China and Silicon Valley, according to Forbes magazine

Ed Clark: Well, thanks for having me, Howard.

You once told me that you were good at making the complex simple. Can you do that with AI and machine learning?

I can try. The concept, in one sense, is just analyzing data and people have been doing that for ages. The difference with AI is that it can have massive amounts of data that it can analyze, but with machine learning it doesn’t start with the proposition of what we think we should find interesting, it lets the data tell you what’s interesting. So let me give you an example. We have a professor Alan Aspuru-Guzik, who moved to Toronto to join Vector. He is a chemist. He has a robotics lab that 24 hours a day, seven days a week is looking at different molecule combinations to understand how they can come together and what they would reproduce. He lets the machine learn. That’s the power. It brings together all the data and lets the data tell you what the right thing is.

With the help of that cheap, fast computing power I mentioned.

Absolutely. That was the barrier.

So Ed, when you were about to leave the bank, somebody said to me, ‘hey, this guy, he’s not going to retire, go play golf. He’s going to stay in public policy,’ which you did. You were an unpaid adviser to the premier of Ontario for a while. Now you’re the chair of Vector. Why are you so interested in AI?

I was originally trained as an economist and the thing that interested me was how some countries grow their economy and make their people a lot better off, and some do it less well. And to me, what AI represents is an opportunity for us to actually make Canadians better off, significantly better off. Because if you take a look at Canada, we don’t have that kind of growth in our productivity per workers as some other countries do. But we have some of the keys to making a great knowledge economy and using that to make people better off. So when I look at the knowledge economy I say, what are the components. They are having a great education system, having a great health system, being willing to take immigrants. But then you’ve got to take that combination and apply our learnings to make people better off. When we set up Vector, we said, yes, we have to be great at academic AI, but we also have to say that we are going to help Canadian institutions. We said institutions because it might be companies, but it may be health care workers or governments in some way. And third, we have a unique ecosystem here where we have more AI startups in Toronto than anywhere in North America. But in Canada, our history has been we get these startups, then they get sold and we lose them. And we said, can’t we help these firms actually grow to become world competitive companies headquartered here in Toronto? So that’s the mission, and I get really excited about making sure that happens.

What about Canadians? Do we trust it (AI) because so many people think, you know, it’s just going to wipe out my job?

yeah. So two points. One is, I do think people worry about AI. It’s hard to understand. They worry, ‘Oh, the surveillance economy. People are using my data to make me want to buy stuff or do things like that.’ And so they have legitimate concerns, So one of the things that we’re trying to do is work with small businesses and say, ‘how could you make yourself better?’ So an example I cite is a mushroom farm. The guy’s trying hard to be competitive because the machines that they used to gather mushrooms damage the mushrooms and don’t work very well. They installed AI and all of a sudden. Wow, they’re getting tremendous crops. They can hire more people and say, ‘Well, I can grow here because I’m very competitive.’ So there’s simple things like that, that AI can improve.

So is this competitive pressure — global competitive pressure — that is going to force small and medium sized enterprises to adopt this?

Absolutely. You know what the Chinese say all the time is, ‘Yes, you can beat us at the academic AI. We don’t care. We’re going to beat you at the applied AI, and that’s where the money is and that’s where the jobs are.’ The example I always use is to say AI is like electricity. Well, if there was a country that said ‘Electricity was coming out, we’re not going to use it. We don’t like it.’ That country is not going to be very well off today. And AI is going to be a component of most businesses in one form or another and the people who are better using it are going to be better off.

So obviously, I’m not telling you anything you don’t know here, but electricity can be great — it’s fueling what we’re doing now — but it can also burn down your house. AI can be optimized for good, but it can be optimized for bad.

Yes, absolutely. And we have lots of professors at Vector who worry about exactly those things. Things like explainability — let’s understand that bias. Are you introducing bias? That’s true in the medical field. If you’ve got a database that has only Caucasians in it, will it accurately predict for non-Caucasians? Those are issues that people are quite aware of and are working on. It’s not perfect. But the fact is, if you had a baby that was going into cardiac arrest and you didn’t have AI, it could die. Now we have predictive models to send messages to the doctor to say, ‘That baby’s about to go into cardiac arrest. Here’s what you do.’ And that dramatically changed the outcomes.

So that’s being used here right now?

Donate by a teacher at Vector working with Sick Kids to save lives.

How do you think about your grandchildren’s future in a world so dominated by AI?

Well, I think like electricity, there will be goods and bads and we will figure them out. But I think if I have a worry — and that’s where I am spending my energy — if AI only makes Google and Facebook and Apple better off, I don’t think that’s a particularly great outcome. But what I want to do is, how can we teach the mushroom farmer, the miner, the steelworker to improve? How can we make our firms more competitive using AI?

Listen here or subscribe at Apple Podcasts, Spotify or wherever you listen to your favorite podcasts.

JOIN THE CONVERSATION

Conversations are opinions of our readers and are subject to the Code of Conduct. The Star does not endorse these opinions.



Leave a Comment