The Current - Is Canada ready for driverless cars?

Episode Date: July 15, 2025

A new robot delivery vehicle pilot project in Toronto ignites conversations about driverless cars in Canada and where this country stands in terms of innovation. Our two experts tell guest host Megan Williams what opportunities the autonomous vehicles bring to Canada, and what to look out for.

Welcome to the Dudes Club, a brotherhood supporting men's health and wellness. Established in the Vancouver downtown Eastside in 2010, the Dudes Club is a community-based organization that focuses on indigenous men's health, many of whom are struggling with intergenerational trauma, addiction, poverty, homelessness, and chronic diseases. The aim is to reduce isolation and loneliness and for the men to regain a sense of pride and purpose in their lives. As a global health care company, Novo Nordisk is dedicated to driving change for a healthy world. It's what we've been doing since 1923. It also takes the strength and determination of the communities around us, whether it's through disease

awareness, fighting stigmas and loneliness, education, or empowering people to become more active. Novo Nordisk is supporting local changemakers because it takes more than medicine to live a healthy life. Leave your armor at the door. Watch this paid content on CBC Gem. This is a CBC podcast. Hello, I'm Matt Galloway and this is The Current Podcast. There's a lot going on at this busy intersection just northwest of downtown Toronto. Big trucks, crowded city buses, people on bikes and motorcycles, parents pushing strollers, kids on push scooters. And now there's a new kind of vehicle on the road, robot delivery vehicles.

And as our producer, Julie Kreisler, found out, they are the talk of the neighborhood. Are you from CBC radio? I recognize you. What are you doing? Have you seen those little robots in this neighborhood? The little robot delivery things? I heard they're here and people keep wondering how they can get the robot to come to their house.

It's a big thing. Like, it feels like the future today. Do you know what I mean? My cousin lives right around here. She saw one just on the street and was like, what is that? Yeah, yeah, yeah. And she said she wouldn't have known what it was if I hadn't asked her if she had seen one. Well, that's awesome. Yeah, I haven't seen one, but I've been internet.

Yes, that is the voice of Matt Galloway, who happened to be passing by. And while he hadn't spotted a robot yet, many other people had. Yeah, they're just small small like cooler looking boxes. They're all white. They're pretty slow so I think they keep up traffic but they're kind of cute so I like them. It looks like a small vehicle and it's got advertising on it of course and I don't know I had to like stop and look and then I thought oh a delivery I, yeah, it just was following, a car was following it. The vehicles are part of a pilot project by the auto parts company, Magna International. And yes, if it seems like the future is now, some people were excited and intrigued, but

others a little bit wary. I would welcome them. Please let them know if they're listening right now. I would do anything for them and I could definitely join their empire. Yeah, I need a job too. So if they can employ me that would be great You know it depends on the energy consumption too I mean there are a lot of bike riders in this city and I find that a bit hazardous because often they're not in the bike lane And there's tons of delivery people and I don't want them to be out of a job either if that's what they need to do to make money.

I'm very intrigued to see how they can maintain just how the streets are and everything. So like the potholes, if they're able to see the potholes, go around the potholes. Even construction. Construction is all over the place in the cities. Or like traffic. I know for me when I'm on my bike and I'm like, oh oh snap there's a lot of traffic I'm able to like weave around it. Are these robots able to weave around the cars? Among those raising concerns about this pilot project is the local city councillor. Hi I'm Diane Sachs. I have the great privilege of being the city councillor for

University of Rosedale best ward in Toronto. We are at the corner of Christie and DuPont in my ward, very close to the old Model T factory that used to be here, which is now apparently the depot for these spy cars of Magna's. Well let's start with the fact that the provincial government has given a secret permit to Magna or its subsidiary to run these autonomous vehicles and their chase cars in my ward and a couple of the adjacent wards without the city's permission and they raise serious safety, congestion and privacy concerns. If you see this thing, it is filming your face and storing it and using it for their profit

without your consent and with no clear safety precautions. This thing isn't ready to be in downtown Toronto. The only place it's been deployed in the real world according to what we were told is one car in good weather delivering pizzas in Birmingham, Michigan population 21,000. So we don't know how good their software is. We know they're gonna be not going at the speed limit, so they're slower than the speed limit,

so slowing traffic as they go. They have a chase car right behind them all the time, so they're gonna be rushing through intersections to try to get both of them together. They'll be pulling over Lord knows where. The city staff gave them a list of intersections where it wasn't safe for them to go, and they said too bad.

They wouldn't take reasonable precautions because they expect to make a lot of money using our faces and our streets to practice their algorithm. It's a very busy urban area. We have streetcars, we have buses, we have delivery vehicles, we have bike lanes, we have pedestrians, we have a lot of people with mobility needs. Like the senior just going by us right now. We aren't even getting data to better understand how this would work. We've been given no say, no control. All we know is this thing is not ready for prime time and it's being dumped on us against our will. Well, the concerns might be big.

This pilot project is pretty small compared to what we're seeing in the US. Google's Waymo runs fleets of autonomous cars in San Francisco, Los Angeles, Phoenix, Atlanta and Austin. Tesla launched its first fleet last month in Austin and there are more competitors on the way like Uber. Some are worried that Canada is lagging behind. Stephen Waslander is director of the Toronto Robotics and AI Laboratory at the University of Toronto. He's in studio with me now. Good morning. Stephen Waslander Good morning.

Katherine Blood So what do you make of this Magna Trial in Toronto? Stephen Waslander I think it's really exciting. I think we haven't seen a lot of companies coming to Canada to do their testing here. As you said, it's an early pilot program. I think it covers around 20 vehicles, so it will have limited impact on traffic or anything like that. It's just great to see that they're trying things out on our streets. I hope it comes and lasts and brings big progress. So when we talk about an autonomous vehicle, what does that actually mean? So it means a vehicle that can drive itself without any human supervision. So in the case of these delivery robots, right, we've heard about the chase vehicle.

That vehicle has access to what the autonomous car can see, but really there's just a driver in there that's able to press a button to stop that car in the case of an emergency. When we talk about fully autonomous, we're really trying to solve the entire driving problem wherever the vehicle is trying to go. So it has to be able to sense the whole world around it. It has to make real-time decisions about how to move through the environment. It has to plan its route and it has to safely execute all of the maneuvers that need to be driven.

And it presumably has somebody watching from afar? So in the case of the Waymo vehicles, for example, like Waymo is sort of one of the leaders here in North America for sure, their vehicles are operating autonomously without direct supervision. You know, they're able to pull over to the side if there's anything that they're uncertain about and then a teleoperator can actually dial in and take over the car and resolve the situation. I want to ask you about the concerns we just heard from City Councillor Diane Sacks.

She called them spy cars. I noticed you chuckled a little bit when you heard that and questioned the data that they're collecting. First of all, let's talk with the data collection and why. Yeah. So the primary purpose of data collection is to improve the onboard algorithms for detection and tracking of pedestrians, cyclists, vehicles on the roads. So really they're not interested in people's faces and their whereabouts, they're not interested

in IDing people and tracking them over time. What they're interested in is, is that a person? And what is that person going to do next? And so in any data set I've ever worked with in my research in autonomous driving, we've always anonymized everything about the pedestrian, so facial blurring, we've anonymized license plates and this is sort of a standard within the industry. Now, there's no requirement or law that's in Canada that says anyone working in autonomous driving has to do that, but it's sort of common best practice.

The company running this pilot, Magna, declined to talk to us. But in a statement, they did say that the company does not share this personal information with third parties and does not sell this information. Data is stored securely by Magna in Toronto and managed in compliance with Canadian privacy laws. That's a quote. Now, you've just said you don't really see much problem with it, but I mean, is there any reason why we should be concerned? Certainly, and it's part of a larger problem, right? So we also have security cameras in every shop we ever go into and on street corners, we have traffic cameras, etc. Right? So this is a

larger societal question about what can be tracked and what companies should be allowed to use and resell and who owns that sort of information. So I think it's wrong to single out this particular Magna trial and say they're the spy cars. On traffic and safety issues, Magna told us again, I'm quoting, the delivery vehicles are designed to operate at low speeds up to 32 km per hour on low speed roads only as part of a multi-layered safety system that prioritizes vulnerable road users. How safe are these vehicles on the road? So, I haven't seen any data and I haven't seen them in operation yet.

So, I can't speak too precisely to these vehicles specifically, but the low speed has a huge impact on the overall risk of the pilot. They're always operating slower than traffic. They'll be more of a nuisance, I think, than a safety concern. And with that chase vehicle, their stopping distance would be on the order of a meter, maybe two, at their top speeds. Right? So- Is there a problem with them going too slow apart from the nuisance? Michael O'Neill I think just the nuisance really, right? And you would hope that they would, you know, be able to improve their technology to the point where

they can drive at the regular flow of traffic. Welcome to the Dudes Club, a brotherhood supporting men's health and wellness. Established in the Vancouver downtown Eastside in 2010, the Dudes Club is a community-based organization that focuses on indigenous men's health, many of whom are struggling with intergenerational trauma, addiction, poverty, homelessness, and chronic diseases. The aim is to reduce isolation and loneliness, and for the men to regain a sense of pride and purpose in their lives. As a global health care company, Novo Nordisk is dedicated to driving change for a healthy world. It's what we've been doing since 1923.

It also takes the strength and determination of the communities around us. Whether it's through disease awareness, fighting stigmas and loneliness, education, or empowering people to become more active, Novo Nordisk is supporting local changemakers because it takes more than medicine to live a healthy life. Leave your armor at the door. Watch this paid content on CBC Gem. Hey, how's it going? Amazing.

I just finished paying off all my debt with the help of the Credit Counseling Society. Whoa, seriously? I could really use their help. It was easy. I called and spoke with a credit counselor right away. They asked me about my debt, salary, and regular expenses, gave me a few options, and helped me along the way.

You had a ton of debt, and you're saying Credit Counseling Society helped with all of it? Yup, and now I can sleep better at night. Ha ha ha! Right on! When debts got you, you've got us. Give Credit Counseling Society a call today. Visit nomordets.org. Steve Waslender, I'd like to bring in another guest who's sitting beside you. Bilal Farouk is the Canada Research Chair in Disruptive Transportation Technologies. That's

a very cool title and services at Toronto Metropolitan University. Thanks for being here and listening in. Bilal, how far are we in Canada from seeing more autonomous vehicle fleets in our cities of the kind that have already started in the US? Yeah, so if you look at the kind of evolution that is going on, there's another project going on currently, for instance, in Markham, where smaller delivery

vehicles are doing the same thing. So there's definitely momentum. I would expect that there will be more and more such pilot starting. And I really like the kind of differentiation here, especially in Toronto, where we are working on use cases that are very different than the kind of use cases operating elsewhere. It gives us very good data, very good feel on how these vehicles will be operating on our streets. Toronto streets are very specific. They have their own characteristics.

Yeah, I was going to ask, what are the major challenges to adopting this technology to Toronto or other Canadian cities? For instance, the pilot where is it operating in the area, that area is kind of a medium to high density area. It is not as dense as, let's say, the central business district, but still it has good density, very residential. People are walking everywhere. People are using sidewalks. People are jaywalking, people are, you know, walking into intersection, there are, you know, cyclists as well. So there's enough complexity there, where, you know, these vehicles can learn and develop their intelligence. And based on that, then they can be deployed in more areas where there's a lot more traffic.

Now, Steve, we heard some of the comments of Torontonians on the street and that one woman towards the end talking about potholes and you know things that happen in dense urban, the wear and tear from dense urban cities. But what about weather? What about the Canadian weather? I mean how big of a problem is that? It's definitely a problem, yeah. So we've been actually researching this for the last five years trying to find ways to enable autonomous driving in winter conditions specifically. And what we found is that as the level of precipitation increases, the ability of the sensors to detect all of the necessary pedestrians, vehicles, cyclists diminishes and it diminishes sort of at distance.

And so that means you have to lower the speeds at which you want these vehicles to operate and you have to come up with redundant sensing possibilities. So we've been looking heavily into radar as one alternative. It's lower resolution, but it can see through snow. There's another aspect to this, and that is that everyone's behavior changes in these weather conditions, right? Vehicles slide through intersections, hills become difficult, all kinds of things like this.

You have to adapt all of the deep learning systems that we use in order to build these autonomous vehicles. They all have to have been trained on and be aware of these kind of additional challenges. And so this is why the rollout's been fairly slow. Like we saw the self-driving car companies in the North America started in simple cities, I would call them, like Austin and Phoenix, where the weather's always nice and the roads are mostly straight. Not many snow storms there. Yeah, exactly. And then they went to San Francisco, which was definitely a step up in complexity and a lot closer to our urban environment, but nobody's come up north yet. So, because you lose multiple months where the current technologies would need to be adapted.

Now there's also some criticism around the rules and regulations of autonomous vehicles across the country. Frank Wood is the CEO of the Canadian company Inverted AI. He says they have the third largest vehicle behavioral data set in the world. Listen to him. Well, I in principle agree that Canada's policies are sensible. What really actually happens is that as we consider moving towards on-road deployment, which we are doing, our first option and most cost-effective option is to go elsewhere, where the policies and procedures and so on and so forth are more well-trod and practiced. In effect, it means that we're going to spend in our fleet development elsewhere. Bill, how would you assess the regulatory environment for autonomous vehicles here in Canada? So the elsewhere that was talked about, that is also very specific. Three, four states out of

50 states, you know, that's where they can go. Within Canada, as Steve pointed out, that there are a lot more complexities compared to those three, four states, you know. In that context, you know, there are limits to how much the policies can be bent to accommodate these sort of, you know, vehicles. I think within Canada, the approach that can be best suited would be to look at very specific campuses, very specific areas where these vehicles can be tested in the kind of conditions that are here. And then when we are more ready towards, you know, deployment, we can show,

you know, the proof and then the makers would be much open to adopting it. As you say, this isn't happening everywhere in the United States and in very specific cities, but do you worry that we're losing opportunity here in Canada? I definitely worry about it. It's not kind of a binary sort of on or off sort of switch. There is a middle ground and that middle ground is where we can you know make a difference especially you know right now how things are a car industry is suffering and so on so I think that's where we can bring in innovation that's where we can bring in you know pilots that's where we can bring bring in researchers combined together so that we can create you know a vehicle that is

suited for our environment and that can be scaled in Canadian environment. Steve, what do you see as the biggest advantages Canada has with regards to autonomous vehicle technology? I think we have a really excellent ecosystem of researchers and we have one of the highest trained workforces in the world. So we have a lot of really strong AI capability here. Obviously, you know, we have the godfather of AI, Jeffrey Hinton, in the early days. So I was able to build my own self-driving car back in 2016. Really? Wow. It was called the Autonomous. And we drove about 100 kilometres on public roads, which

was really, really exciting at the time. And the program there within Ontario was amazing. It allowed us to define our own safety requirements. So from an early pilot kind of setting, it was really, really impressive what Ontario was trying to do. Since then, I think the whole world has evolved. So we see really strong regulations in Korea, in Germany, in California, right? That are very clear and very precise

on what reporting is needed, what tests have to be done. And we don't have that yet in Canada. So I'd like to see that emerge more prominently and hopefully a national strategy and not sort of one driven by the individual provinces. Now, what do you say to people who see these robot delivery vehicles downtown? They've heard about self-driving cars causing accidents in the US, and they're worried about having many more autonomous driving vehicles on our roads. What's your response to them? Dr. Sajid Zalzad If we look at the accidents that are happening right now, Waymo reports about less than 0.5 injury accidents per million kilometers traveled

based on these two cities, Phoenix and San Francisco. And if you compared humans causing accidents that are about 2.5 injury accidents per million kilometers. So there's definitely safety improvement compared to what is happening right now. And if you look at the kind of accidents that are happening, they are crashes with minor injuries rather than,

you know, very major injuries, you know, hit the curb or go on a sidewalk, but stop and so on. So there's definitely an evolution in terms of their safety and that should be taken very positively. That should be considered and it is at a scale, at a city scale. So there's definitely improvement and that's what we should look at and it will keep on improving. Steve, what would you say to reassure people?

Bilal's point is bang on. So what we've seen from Waymo is really impressive. They're better than human drivers in the environments that they're driving in. Now, how we make sure that that happens across all of the different companies trying to do that, I think that's where regulators actually play an important role. One of the best things that California did early on was force companies to always report disengagements and crashes. And those are always publicly available. So every year we could track how far these programs had driven, how many accidents there were. So I'd like to see that here as well, obviously. I think that'll help a lot. What we're seeing is we always knew that if we just removed distracted, drunken, and drowsy driving, we would

get significantly better performance out of humans and we're seeing these robots actually do that. Steve, Bilal, thanks so much for speaking with me today. Really fascinating stuff. Thanks for having me. Thank you. Stephen Waslenders, the Director of the Toronto Robotics and AI Laboratory at the University of Toronto. Bilal Farouk is the Canada Research Chair in Disruptive Transportation Technologies and Services at Toronto Metropolitan University. You've been listening to The Current Podcast. My name is Matt Galloway. Thanks for listening. I'll talk to you soon.

For more CBC podcasts, go to cbc.ca slash podcasts.