In short
Secret Leaders Podcast Episode Notes
Episode Title
Meet the CEO Bill Gates backs to beat Tesla: Alex Kendall, Wayve.AI
Podcast Overview
- Hosts: Dan Murray-Serter and Chris Donnelly, serial entrepreneurs with extensive experience in building, selling, and scaling companies.
- Focus: Insights from influential business figures, exploring their journeys and lessons learned in the business landscape.
Guest Introduction
- Featured Guest: Alex Kendall, Co-founder & CEO of Wayve.AI, a British startup valued over $1 billion, aiming to revolutionize autonomous driving and challenge Tesla.
Key Discussion Points
- The Autonomous Driving Landscape
- Current Reality:
- 1.2 million road deaths annually, primarily due to human error (99.6% of accidents).
- Cars are dormant 96% of the time.
- Wayve.AI’s Vision:
- Develop an AI that allows cars to make autonomous decisions, enhancing road safety and efficiency.
- Transition from hands-off to eyes-off driving, ultimately creating a second living room in vehicles.
- Competitor Landscape
- Differentiation from Tesla:
- Wayve does not sell cars but licenses its AI technology to various car manufacturers.
- Focus on an AI platform adaptable to different vehicle types, enhancing vehicle functionality without the need for complex infrastructure changes.
- Safety and Public Trust
- Public Perception Challenges:
- Misconceptions about driverless technology often stem from fear of the unknown.
- Safety as Priority:
- Emphasis on learning from past incidents in both autonomous driving and other industries (e.g., airlines).
- Societal Impact of Autonomous Vehicles
- Urban Transformation:
- Envisioned future includes pedestrian-friendly cities without car parks, reduced traffic congestion, and increased green spaces.
- Improving Quality of Life:
- Autonomous technology designed to give back time to users, allowing for enhanced productivity and well-being.
- Future of Work and Employment
- Job Disruption:
- Autonomous driving will disrupt existing jobs (e.g., truck drivers), but potentially create new roles that require upskilling.
- Historical Context:
- Comparison to the Industrial Revolution, emphasizing the need for proactive education and adaptation strategies.
- The Role of AI in Daily Life
- Embodied AI:
- The potential to automate physical tasks across various sectors, significantly impacting global productivity and sustainability.
Key Takeaways
- Technological Advancement:
- Wayve.AI’s approach seeks to simplify autonomous driving, moving away from high-definition mapping toward a flexible AI model that learns from its environment.
- Community Engagement:
- The importance of community acceptance and regulatory frameworks to promote the adoption of autonomous technology.
- Vision for the Future:
- A world where autonomous vehicles become a norm, improving road safety, efficiency, and overall urban living conditions.
Conclusion
- Final Thoughts:
- Alex Kendall expresses optimism about the role of AI and autonomous driving in shaping the future, aiming for significant societal benefits and a robust technological landscape in the UK and beyond.
Contact Information
- For further inquiries or details, reach out at: hello@secretleaders.com.
Sponsors
- Sponsored by Wise Business and Vanta.
---
This markdown file summarizes key points and discussions from the podcast episode featuring Alex Kendall, highlighting essential insights into the future of autonomous driving and its impact on society.
Written by AI. May contain mistakes. Listen to the episode to check what was said.
Transcript
Automatic transcript. May contain errors.0:00We want to give the car the intelligence to make its own decisions and deal with things it's never seen before. AI is the way to solve that. Today on Leaders, Alex Kendall, co-founder and CEO of Wave AI. What our system's able to do is you can sit back with your hands off the wheel and initially you'll need to monitor the system. Point-to-point drive from when you're parked to driving out to the road to navigating to going through roundabouts, traffic light intersections, motorways to your destination and park yourself. And that point-to-point navigation is what our AI can enable. The company alongside Tesla leading the driverless vehicle industry and betting on the future of transportation being autonomous.
0:35When we started eight years ago, that was totally different from anyone else in the industry, including Tesla. A couple of years ago, Tesla has pivoted their approach to the same as what we're doing. And so we share the belief that an AI approach is the way to solve this. We want to be on a large number of different car makers' vehicles and fleets, whether that's delivery or robo-taxi fleets, and enable people to choose what kind of vehicle they want. Alex is the man building a safer and more efficient way that we'll all commute to work. transport goods and ultimately run the world. I think the number one thing for us to continue to be very cognizant of as an industry is safety.
1:08Sadly, there have been incidents with the company's autonomous technology in the past. And if I look to the airline industry, if we can collectively learn from those as an industry and make sure that we put procedures and technology in place to mitigate those risks in the future, I think that's the thing to continue to keep an eye on. But then the next step is going to eyes off, where you're not just with your hands off, but also eyes off and it can give you time back. This is a good thing for society as we see the benefits of AI bringing in transportation, bringing more safety, sustainability, accessibility.
1:37I'm pretty convicted that autonomy will be here. I think the question is around timing and how quickly the adoption is from initial experiences through to it penetrating. I dream of healthy streets where you've got no car parks blocking the road, but you've got great architectural interaction between footpaths, vulnerable road traffic like cyclists and shopfronts. you've got much more space opened up for green space that improves the well-being of cities and then the improvements on adjacent industries whether it's healthcare or even just connectivity of people or freeing up the choice of where people want to live I think all of these things can be opened up too.
2:15Alex welcome to the show. This is great. Thank you you're already excited which is a good start. So some of my friends are quite terrified about a driverless future so what are of the misconceptions that you think can put that to bed when i see people get out of our car and the magical smile on their face from the experience i think it's one of those ones where self-driving's not really accessible today you've got to go to a couple of affluent cities like shanghai or san francisco to experience the technology and so it's natural when people haven't seen it before it's it's you know there's going to be a fear of the unknown but when you're in the vehicles and when you experience what it's like to be driven autonomously.
2:56It's such a magical experience and I think the challenge here is that we need to take this technology from small-scale pilots and geofenced operations into something that really scale around the world. And then the other thing is it's so tragic how many of us know someone that's been impacted by road safety and I mean you look at the statistics, unfortunately over 99.6 or 7 % of accidents are due to human error on the road. There's about 1.2 million people around the world lose their lives every year due to road accidents and orders of magnitude more injured. And I think we've got the opportunity to really drive those numbers down to zero through automation.
3:35So these are the kind of things that I think should make us feel very optimistic about the impact this technology can bring. And it's It's one of those ones you've got to go for a ride to see it. And I can't wait to make that accessible to everyone around the world. I accept your invitation. Is that how it works at the moment? How does one experience the magic of autonomous driving in somewhere like London at the moment? Well, right now, we're the only vehicles driving around in London. We do it every day. I'm in the car most days, all around the city. And we drive actually across the UK. We're not yet open to the public, so we can go for a drive.
4:08but I can't wait till it's at the point where we can open up to more people. Tesla is obviously, I guess, the best known brand in this space, right? They really pioneered very originally with a very specific view as well about how to do driverless and what an autonomous future looked like. But unlike them, you obviously don't sell cars. So can you explain for our listeners, what is the key differentiation point between what you do and what Tesla does? I founded Wave eight years ago and we started the company with the belief that end-to-end AI or embodied AI would be what makes self-driving possible.
4:43The challenge with self-driving is how do you make something that has the intelligence to deal with diversity and share vast scenarios that we see on our roads. If you think about when you start watching for these things, it's interesting just how many things you notice. Like I was driving the other day and there was one of the mounted policemen on a horse on the road or you see scenarios like there was a a car driving with a one of those roadworks temporary traffic lights on the back of it and all kinds of things that are maybe not immediately obvious how to drive but the diversity of these kind of scenarios is enormous and so fundamentally our belief is that AI is the way to solve that it's not a system that's going to be told how to drive where in a certain certain parts of the world you've got mapped out areas where the car's told how to drive but instead we want to give the car the intelligence to make its own decisions and deal with things it's never seen before.
5:35If you remember during COVID all of the footpaths were widened to allow for social distancing that's an example how the world's constantly changing and so I think for to develop a system that's truly safe and trustable we've got to give it the onboard intelligence to make those decisions. So that's really where we're coming from as a company and when we started eight years ago that was totally different from anyone else in the industry including Tesla. A couple of years ago Tesla has pivoted their approach to the same as what we're doing and so we share the belief that an AI approach is the way to solve this.
6:06But how we differ is our model is we're producing an AI model that we are licensing to fleet and car manufacturers around the world and so we're enabling this experience, this autonomous experience for all kinds of car manufacturers. I mean so I think in general there are going to be vertically integrated robotics and autonomous vehicle manufacturers that build the vehicle and the autonomy stack themselves, but the vast majority of the market have amazing vehicle platforms, amazing driving experiences, but can benefit from the power of a platform that we can provide around the industry. So that's really how we differ.
6:45We want to be on a large number of different car makers vehicles and fleets, whether that's delivery or robo taxi fleets, and enable those vehicles and enable people to choose what kind of vehicle they want, what kind of driving experience they want in every country around the world. And we believe that through that kind of platform play, we can build this to a much greater scale and to a much higher level of safety. Yeah, I guess interesting because Tesla obviously started as a, you know, the whole thing was, we're a technology company that makes cars. And so when people first got in them and saw the way that that could be different, it was truly innovative and super different for everyone.
7:23Whereas the rest of the automotive industry are car makers and brilliant ones. So they're not trying to pretend to be AI companies or technology companies. And I imagine there are maybe some exceptions, but most of them don't fancy their chances of trying to also become AI companies, right? What's your favorite car brand, Dan? That's a good question because I'm doing tons of research because I'm having another daughter. So I'm about to change my car. But I've been driving a BMW for the last few years, and I do like it. I'm considering a Volvo because I'm about to have a second daughter, and I'm quite into the whole dad banter vibe of being a dad that gets a Volvo.
8:00And the new Volvos are bloody amazing. So I'm oscillating between the two, but I've been actually, you know, this is the right time to ask me, so I've actually test-driven about seven different brands recently. Nice. Well, it's really interesting to see the diversity across the car industry because we've had most of the CEOs for drives in our car right now and to understand and look at this technology. And of course, we're working with a number to get this out there in their products. But there's such a diversity, you know, some brands are at the moment, given what's going on in the world, really looking to build efficiency and cut costs.
8:37But there is a large number of brands that are looking to innovate and really drive the future and of course autonomy is is sadly so rare today and yet to I think in the next couple years we're going to see it become a a major part of vehicles around the world and when that happens I think the car that you choose right now that the fact is you talked about about a lot of brand or maybe different aspects of comfort or safety I think a lot of that is going to be driven by what is the autonomous experience and ultimately how intelligent and smart and customizable is that car to helping you with with your life and helping helping give you a safe mobility experience and i think that's really going to come down to come down to the intelligence and the capability of the ai that drives these cars yeah i suppose actually really interesting to this exact point i want a bmw because i like how it looks on the outside i think the inside's okay i've never been that fussed about it but like it looks great on the outside i just think it's a lovely looking car also you know not like a porsche or like something too flashy which i think would just get keyed and stolen in london not range rover where you can't get insured all that stuff right so that's that interesting factor but volvo have these stats of no one's ever died in an xc90 and so when you're a dad making a decision for a car you're like that's very compelling i don't really care about the look and feel of my car as much as i care that no one's ever died in this car that's an incredible selling feature that maybe doesn't matter at all in a world of wave AI right yeah and one of the really interesting things about making this platform approach possible it really comes down to the generalization of the AI so generalization means how what's the AI's ability to adapt to new environments it may not have seen before so for driving we see that in driving in different countries driving with different vehicles or driving with different use cases if you're doing point to point navigation or if you're dropping goods off or something like this.
10:36And one of the interesting things is that our AI is so flexible, it can work with different sensor setups around the car. Are the cameras mounted behind the windshield, on the wing mirrors? And that gives automakers the flexibility to build cars in the shape and the industrial design that they want. And so there's still going to be a lot of brand differentiation through this technology, particularly because of the flexibility that we offer. but the more interesting thing is what is the driving experience it creates and maybe um maybe to introduce that what we're going to see in the future of course today we have the concept of driver assistance and today what you see on most products is very basic levels of autonomy like a lane keep assist feature when you're on a highway or a you know some cars might have a smart parking feature where it helps you do a parallel park but they're very nuanced with specific scenarios and they're also quite brittle um you know if you're on a highway and you lose the road marking the system won't work and you have to take over things like this.
11:32What our system's able to do is point to point drive from when you're parked to driving out to the road to navigating, to going through roundabouts, through pedestrian crossings, traffic intersections, motorways to your destination and park yourself on that point to point navigation is what our AI can enable. So you can sit back with your hands off the wheel and initially you'll need to monitor the system and it'll drive like that with your hands and feet off and you can, it can drive in a way that gives you a higher level of safety and confidence. But then the next step is going to eyes off where you're not just with your hands off, but also eyes off and it can give you time back.
12:11I mean, this is going to be super impactful for people's lives, being able to spend time doing more things, but also just improving the road safety by removing the incidents that are caused due to driver fatigue or driver error or something like that and i think um going to the eyes off experience where portions and and eventually the whole of the drivers is one where you don't need to pay attention to the road that's going to be such a such a fundamental change in the way we drive and really open up so many possibilities of what you can do inside the car okay so you've called uh autonomous driving obviously the space race of our generation so what is actually Wave AI's goal?
12:49What is like the vision as you see it? The vision is to, I mean, our vision and statement as we've defined as a company is to propel the world forward into the age of autonomy. And how I think about that is that the idea of autonomous intelligent machines for me is the next wave of tooling. If you think about what's driven us forward as a human race. It's our ability to build tools from the original wheel to a calculator to an x-ray machine. And all of these tools enable us to do so much more and live a much more prosperous life. And so for me, intelligent tools are just the next stage of that.
13:28When we have machines that we can delegate tasks to in a way that they do very complex tasks in a way that we can trust and that actually adds value. Now, I think we're first going to see this with autonomous driving because in autonomous driving we have really reliable hardware supply chains, vehicles around the world. There's a great business and safety case for doing this, but this is just the tip of the iceberg. Beyond cars, we're going to see this grow into manufacturing robotics, domestic robotics, other kinds of intelligent machines that can really assist our lives where we choose for them to do so and allow us to delegate tasks that we want to do to free up our time for where it matters most.
14:08I think this is going to be one of the biggest technological changes we go through in our generation. And what makes it all possible is the embodied AI, the AI that can interact with the physical world to make these things possible. And that's exactly what WAVE's building. We want to build the embodied AI that has the ability to generalize across all of these use cases and enable people to see the benefits of intelligent machines throughout their lives. So as a dad of, you know, seem to be two daughters, with a wife, mum, how does this future change each of our lives? So for the kids, some of my age, and some of my mum's age, like how is life changing for them in the next five years?
14:54Oh, it's going to be profound. I mean, think about all the products you use, the way that they're built and manufactured. I think there's a huge opportunity to improve the efficiency, the sustainability of those supply chains in driving? Think about the way you move people and goods around the city? Or, I mean, what are some tasks that you do that maybe you don't enjoy, maybe you wanted to delegate? I'm sure some come to mind. Yes, there's quite a few, you know, but email is always the top of the list, isn't it? Yeah. So, I mean, email I'd put in the realm of cognitive AI, as opposed to embodied AI.
15:27So cognitive AI is AI that deals with knowledge. and of course the chat GPT and the large language models that you see everywhere today, they're incredible. I mean, these systems will for sure be able to accelerate and automate a large amounts of email communication. They already are today. But the other question is for physical tasks. If you look at all of the work done around the world, about 65 % of global GDP is physical labor and physical tasks. So it's clear that if you think about bringing the benefits of AGI to the world where it can really support most of the work that we're doing, the vast majority is not today cognitive tasks but physical tasks.
16:10And I think there's a real opportunity there to bring this technology and to really improve our ability to work with all these kinds of physical tasks. So a few years ago I invested in a company that was doing robotics called Figure AI. obviously you've got Tesla that has Optimus what you're saying here sounds like you've also built the system that a lot of these robotics will use is that right so if companies like you know BMW or whoever are making their own robotics you will be the operating system for them is that is that my understanding correct well it's it's interesting to look at the other other forms of of robotics outside of autonomous driving I mentioned why I think autonomous driving is is going to happen and it is happening now, we're through the trough of disillusionment.
16:55There was a bunch of hype about it in the 2010s, another sort of winter about this. But now with what we're seeing with AI, we're able to build products that can really meet the standards of safety that we need. So autonomous driving is happening now. There's regulation in place, there's global supply chains, and there's the right demand and technology to do it. If we look at industrial robotics, it's interesting in automotive vehicles were opened up with what is referred to as a software defined vehicle. So a car that you would have bought five years ago would have had upwards of tens of maybe sometimes even 50 or 70 computers in it doing each individual discrete function.
17:38One for infotainment, one for windscreen wipers, one for every sensor, every camera had a computer on it. And they're all often quite different companies, right? Exactly. It's very fragmented. And what's happened in automotive is there's been this consolidation into software-defined vehicles. We have a dominant central computer that can be over there updated with new software and can learn through data and enable these autonomous applications. And that's incredible progress that the automotive industry has made. If you look at industrial robotics, that hasn't really yet gone through that opening up of the platform.
18:10And so there's still that to go through. And then if you look at humanoid robotics, I think they're at the very start of where self-driving was in the 2010s. There's going to be, I think there's a peak hype cycle right now. But what we've learned in embodied AI is that to build general purpose embodied AI, it's fundamentally an AI problem. limits, not one where you want to create a very constrained environment and invest in a ton of expensive hardware for these kind of systems. Instead, you want to find a way to expose it to large amounts of data so it can learn from general purpose applications.
18:42And what we're currently seeing with humanoids, I think there's still a few of those learning cycles to go through. But no doubt this will be a huge opportunity in the future. And we think there's a real opportunity to take this embodied AI from automotive, all learning it's had about how the physical world works and to bring it into these other kinds of applications as well. And there's going to be so much co-development from both of them. It'll learn from each other and just ultimately get more intelligent. Okay, so let's talk about practically where we are right now. How are cities going to change?
19:18What does it look like for a city? Is a city going to become the most convenient place to live again? Well, initially it won't change. Our technology doesn't require anything of the infrastructure to change. We don't need 4 or 5G connection. We don't need different road layouts or road rules. We've built our systems with the intelligence to drive in a very human-like way. And so initially we'll see... But without the 99.7 % error rate. I should say in a way that, yeah, exactly, in a way that can deal with mergers and all of the interactions that you need, but do it in a way where it sees 360 degrees around the vehicle dozens of times a second and is constantly alert and switched on for hazards.
20:03So that's where we're at today. But what's going to happen is we see autonomy become not just the minority, but the majority of road transportation. I mean, changing cities is a large capital expenditure. So I think it's going to take time for some of these things to evolve. But there's tons of opportunity. First of all, I would say that I think it's really important that we continue to have multimodal transport. So I don't think autonomous vehicles are going to be the only way to move. Autonomous vehicles or autonomous cars? Well, both. I mean, if you think about trains, micro mobility, biking, walking, I cycle to work every day.
20:40I think these kind of applications are still going to be prevalent and necessary. And would you like your bicycle to be autonomous? I enjoy the cycle. And I mean, if it was autonomous, what would I do with my time? It's somewhat hard on a bike to sleep or read a book or something like that. So maybe I'm fine. Maybe I'm fine keeping cycling myself. Probably just sit in anxiety for a whole hour. Yeah, but I mean, there are so many opportunities once you do have prevalent autonomy to start to think about how we can you redesign a city. Everything from, I mean, one of the reasons why speed limits are low is for safety.
21:18When safety is improved, can you move faster? Most of the congestion, actually the road network is typically over provisioned for the cars that need to drive. Most congestion is due to driver error, whether it's crashes or even just inefficient driving from people. Actually, even crashes themselves sadly cost about one and a half percent of the UK and US's GDP. So there's a ton of impact on that. You can economically quantify on the on the health care system on just just on on on life that that's something that we should be able to drive to near zero. Look at what the airline industry has done through advanced automation technology and safety.
22:00I think they've really made huge grounds over the last couple of decades. And I think we need to do the same in road transportation. Yeah, I suppose that's interesting where everyone has a record, almost like a black box, like in airplanes, right, of what actually happened, which is very different when two people crash, there might be a differing opinion, and then they've got to consider their insurance, and maybe there isn't a camera that can even prove what actually happened so how would you actually train anything that way how can you really learn from that mistake yeah i mean it was really unfortunate we've we've we've had like one incident on the road in the uk where uh sadly we were stationary at a red light and someone uh ran into the rear end of us and uh um you know we pulled up the rear camera in the car and looked at what happened and sadly they were on their cell phone distracted and didn't didn't see our car stationary in front of them so there's a bunch of opportunities like this of course to just just make road driving safer and more efficient i mean there's the whole accessibility part of this as well many people can't drive um and being able to give them the freedom of mobility through a vehicle that can drive itself and take liability for driving is going to open up so many opportunities in a world where um every so let's take london every car in the whole entire city is autonomous okay i know appreciate you like multimodal so obviously there'll still be biticles and everything else but I assume there's no good reason why you don't envision every car should be autonomous as part of your vision unless I'm mistaken maybe like some old Bugatti or something that can't integrate but still otherwise so if every car in the city is autonomous what does that do to transportation time efficiencies and then also I'm interested about emergency services?
23:46Yeah, I mean, just in terms of transportation time, it makes a lot of new interesting models available, maybe lower volume, higher frequency public transport, maybe customizable public transport. Transportation time should go down because a lot of the congestion is due to, as we're talking about, human error. Do we still need traffic lights? Possibly not, because the cars will be able to communicate to each other and more efficiently go through through traffic lights another interesting statistic is in London car parks take up 10 times the size of Hyde Park and cars sit stationary about 96 97 percent of the time I mean most of the time cars aren't used right and so there's an opportunity to take cars from three or four percent utilization to 70 or 80 percent utilization and then why do they need to park they can continually be doing work for us and that opens up so much space can we turn those car parks to green parks interesting okay so in that vision um you've sort of got a i guess you know a taxi model going right where you can just lease your car and is that model if 97 of the cars are sitting dormant not being used and stuff is that i'm going to make traffic a lot worse or is it just be so efficient that actually it doesn't really make a difference i think there'll be um i mean as uh you know as you see um products and services become more accessible of course utilization goes up, but the efficiency factors and safety factors should be more compelling than that.
25:13And so overall, I think we should see net benefit to transportation in cities. But yeah, the the it's going to be quite profound. And and then, of course, there's other forms of robotics outside of the vehicles themselves that will impact everything from delivery to just in general home life. But I talked about hands off and eyes off as a driving experience for personally owned vehicles but then after that then of course you get to driverless where you don't actually need a steering wheel you don't need anyone in the front seat and the vehicle can actually can actually just just drive itself how far away do you think we are from that well you can go in an experience that's driverless today in shanghai or san francisco where you can order a car it'll turn up there's no one in the front seat it's remarkable if you haven't tried it yeah that's way more you definitely should yeah i mean but i guess there's a big difference between we can do it to cultural acceptance because i feel like obviously the safety net in a human's mind anyway is as long as there's a steering wheel and brakes i can take over if i need to yes so i think if we look at what's going to take it from those couple of cities to really worldwide i think there's uh there's three factors to talk about of course there's uh technology regulation and and um public acceptance and uh firstly if we look at regulation last year we were able to legalize this here in the UK.
26:34We worked with government to help write the legislation that legalized autonomous driving. That's awesome. It's going to take about another year and a half to actually implement that legislation, but we will see the legal framework in place to actually run driverless services. We're really closely working with the UN to extend this legislation worldwide, and we're co-chairing one of the industry committees in the UN to be able to get that done. And so that's going to really pave the way from a regulatory and legal standpoint. So that's in a really good spot. Then you have on the technology side, of course, we need to make the systems more affordable, more scalable, more intelligent.
27:12And that's what we believe our approach can enable. Moving away from these systems that we were talking about that have to follow high definition maps, operate in a geo constrained area or rely on vehicles that have equipment on them that's too expensive to manufacture and millions of unit volume to a system that can drive with the kind of lean hardware that we have and have the intelligence to drive based on what it sees for itself and drive in places it may never have been to before. And so where we're at as a company is that we've been building this for eight years and our technology is able to now scale all around the world.
27:49We have cars in the UK, Germany, the US, Japan, they're driving all around the world and now it's about productization. It's about integrating this into vehicles with our automakers and working with them to get the data and the experience to scale up performance to the point where we can validate those levels of safety from hands-off to eyes-off to driverless. So it's coming and we're on the sprint to go down that in the next couple of years. So obviously begs the question then, so many jobs obviously are related to driving, so truck drivers and cab drivers being the obvious ones. I mean Uber comes under quite a lot of fire for encouraging all their drivers to essentially make themselves replaceable.
28:32And they all know that there's a self-fulfilling prophecy that is inevitable, but it's happening. So what is your take on job disruption with this kind of technology? It's a really tricky question and one that I think is important for not only us as an industry, but I think the entire economy to look at when it comes to AI. If you think about analogies like look at the Industrial Revolution in the in the 19th century I mean that happened over about three generations So there was time for generational learning of upskilling beyond the jobs that used to exist to the jobs that will exist in the future And I think that kind of disruption is natural right many jobs that existed 56 years ago if you think about draftsmen and architecture that's now replaced by computer aided design CAD software There's a lot of evolution like that that pushes up the quality of work we do to the higher level.
29:24Even in our team, we've pushed super hard the use of generative AI tools for everything we do from analysis to coding and things like this. And my expectation is that our team is now working on higher level tasks that bring more impact. So these tools should ultimately provide us leverage, but they will disrupt a number of jobs we have, just like other forms of tooling have throughout humanity's history. The challenge here though is that this is going to happen over a couple of decades, not over a couple of generations. And so the time is much more compressed. And so opportunities for moving into new jobs and re-education need to be addressed much more proactively.
Read the full transcript
30:03And I think that's something that collectively we have to find a solution for. But I think overall this is a good thing for society as we see the benefits of AI bringing in transportation, bringing more safety, sustainability, accessibility, and other fields as well. I think this is ultimately going to be a good thing for society and something that we have to collectively work through. I presume it's one of those things that's going to happen anyway, right? It's one of those classic things of, that is the vision of the future, so why not be part of building it? Yeah, that's right. I mean, more practically, if a country chooses to slow it down to protect legacy jobs, another country where it's in global competition will push this forward.
30:47And so ultimately, it is going to be a net benefit for people's lives. And so I think we've got to, you know, be as eyes wide open as we can and tackle these problems head on together. OK, so your PhD in Cambridge, it focused on robotics, right, and automations. So take me back to the moment where you have the spark of the idea that turned into Wave AI. Oh, man, it's, I'm not a, I don't believe in eureka moments. You know, the story of Archimedes jumping in a bath and seeing water spill over and running naked down the street to talk about how he's discovered the laws of density. I don't think that's the case.
31:28I think. You went running down the street naked in Cambridge talking about automation? Maybe in my shooting days. But no, look, autonomy, hard problems are not discovered overnight or in a eureka moment. It's sustained effort over time that allows you to solve things that are truly hard, truly deep tech. And for me, this has been a journey of a lifetime and it will continue to be. I was passionate about building, figuring out how things work in engineering from a very young age. And growing up in New Zealand, I was very curious about adventure and even building robots and creating intelligent machines.
32:06I spent a lot of time building drones at my family's home and flying around the paddock chasing our sheep and things like this. Of course, at university, studying mechatronics engineering at Auckland. I was immersed in this stuff. And so coming to Cambridge was amazing, man. And I got to move into this place that was like a castle, spend time with people, not just in engineering, but cross-functionally were at the top of their fields and just have some time to think and learn and build in this area with very little, with just so much openness. And it was interesting because at the time robotics was completely approached with a non-AI approach.
32:51Everything in robotics was was about trying to simplify the problem and build an infrastructure to make these robots work. And I was fortunate enough to be inspired by a couple of papers and people. And I saw what was happening with deep learning and deep learning was. I mean, it's been worked on for many decades, actually, you know, starting here in the UK with many, many years ago. but ultimately it was those kind of methods that I was able to look at and say, okay, these aren't just for computer vision, these can apply to robotics too. And actually, end-to-end learning is when you get the right data around it, the right learning, and you have the GPU compute, this is eating up any problem we think about.
33:39And we've seen this time and time again from image recognition to game-playing agents with alpha go to protein folding, and of course robotics is going to be the same thing. And when I was thinking about these ideas, I was pitching them to many people and outside of academia, they were they were really dismissed and laughed at by the industry at the time. And so to maybe come full circle to your question about what started Wave, it was a conviction that the robots of the future would not be handheld with infrastructure, but they would be empowered with the intelligence on board to make their own decisions.
34:10And the way to do this was with end to end learning. That was a fundamental belief I had. And so then my question was, OK, I'm passionate about this technology. I want to build it. How's the best way to do this? Many years it was in academia as a PhD student. But then there were some limitations. I as one person or a small team with with my colleagues, we couldn't build robots. We couldn't build simulators. And and then I was I was fortunate enough to get the opportunity to go over to Silicon Valley and spend some time with a startup there. And I had no idea about the world of venture capital and startups and got exposed to it.
34:44And was like, wow, there's the ability to take these ideas to VCs and raise money and go build your vision. And so that's really what sort of landed me in there. And then it's been the most roller coaster of eight years of journey working on these problems to be able to get to the kind of performance you on the road now. And I'm so proud of what the cars can do. I was in the US and Japan recently and seeing our AI, the same AI, drive in both those countries. They're on different sides of the roads with very different driving cultures and deal with the complexities of these environments with no infrastructure, no HD maps, just an AI, a single computer, cameras around the car.
35:26I mean, that's just incredible. So in 2017, you're basically building this in a garage in Cambridge, right? A few friends, like, you know, paint that kind of picture. yeah i mean so uh how we started was um it was interesting i was uh coming towards the end of my phd i was writing up my thesis but this was all gathering steam and so i uh you know actually we had a couple of uh car companies at the time for a phd you didn't want to offer ridiculous amounts of money to buy out the technology but to really shoehorn it into data labeling and what's ridiculous amount of money because i'd say a billion dollars is a ridiculous amount of money So what's your, if that's not, what's your metric?
36:05Well, a PSG student on a 12 ,000 pound stipend a year, it was millions of dollars. So it's quite strong resolve to say no. Yeah, I mean, I've always prioritized adventure and learning over, you know, I think most deep tech entrepreneurs will say this, like we're not doing it for financial returns. It's all passion for the technology. and definitely did not put that into your pitch to SoftBank. Well, no, I mean, that's the resource to go and to go and build this kind of technology and to bring it to the kind of ambition that we've been talking about. And so what happened was, at the same time as I was writing my PhD thesis, I managed to go and meet a few VCs.
36:54Actually, one pinged me, who was an amazing guy, Mike Dempsey, from New York, who was actually reading some of my academic papers, saw it and we got in touch and we were able to bring together a seed round of£1.6 million. And with that money, we rented a house down the road from the university, I got together some of my lab mates and we started building our first prototype in the garage. It was quite fun actually. We used to stay in some of the bedrooms. The small bedroom was a server room where we had about at its peak about 60 GPUs running there with the window open all the time with fans blowing the air out and then the middle bedroom was the boardroom we used to call it and then the main big spaces where we had at the peak we had about 16 people working from this house and our neighbors loved us because during the day that we were we were out of there during the night but the kids around it loved seeing the cars being tested but one day the council saw us trying to build a startup in a house and said go get a real office and so we were also bursting at the seams at that stage but that's when we took it to London and took it to a real office and have started to grow up the company.
38:02Just eight years on from that you raised a billion dollars just over a billion dollars actually. How monumental did that feel? It's a milestone it's a UK milestone right it's the most amount of money that's ever been raised in around it's quite a phenomenal statement uh it was uh it was it was it was a huge uh honor for our team and it's one of those ones where financing grounds are typically a lagging metrics of success and not something to celebrate in itself because they're a means to go build what you're doing and um i think uh you know we certainly see the opportunity and see where this is going internally at Wave, our whole team, we ride in the car, we see what the technology is doing, we see what's possible.
38:47And what I'm grateful for is we got to spend the time with Masa and the team at SoftBank and the level of excitement and conviction they have for robotics and AI, I mean it's their entire focus and has been for some time. This was just a great partnership that came together. And so not only them but also Microsoft, Nvidia and Uber. I think part of the lesson I've learned through all of our financing rounds is it's not just about the capital, but it's the right kind of partners that you need that aren't just investing because it's the hyped thing at the time, but have the kind of conviction to really engage with you, get in the weeds and building a company is so much more than the financing.
39:30But having said that, I think showing that it is possible to raise that kind of financing as a British startup, which for us gives us the escape velocity, meaning we don't need to be absorbed by an American big tech company, but can go build and create this product, launch it and go build a large generational public company. I think that's something that really inspires our team and I hope to see more of it. So we've got a particular fascination on the show with Massa Sun. I've had so many of his funded guests on here from SoftBank. We've had Lionel Barber, obviously wrote his biography and spent a lot of time with him.
40:06There's always a great story with Masa-san. So have you got any? What was it like meeting him?
40:14He is such an exceptional individual. The work ethic, the level of integrity and respect that he brings is, you know, right up there. Is he intimidating?
40:28I wouldn't say intimidating. I mean, he does have the most... did you know about him enough like at the time of like going to meet him there's sort of like folklore around him oh yeah i i i deeply prepare for every conversation right okay for that one for example i spoke to a dozen other ceos that have worked with him okay um it's uh it's one of those ones where good bad and ugly yeah of course i mean i i i think that um there's a right and a wrong way to partner with SoftBank. And I think that the partnership that we've brought together is one that's just firing on all cylinders and is built for sustainable success.
41:09And I also think it depends on the entrepreneur. I think being able to work with and partner with Masa, I mean, some of the things we've worked on together have been amazing. We've gone and met most of the CEOs around the world of different car companies. We've worked on growing the business, but then also ensuring that I can drive away of capital allocation in our company that continues to make the right investments towards um you know towards the future these kind of things are really important yeah that's the thing i can kind of imagine right he can just be like do you want to meet the CEO of Mercedes and BMW or whoever and then just take you there and then if those meetings go even moderately well that's him being like blank check for the next thing as when it happens if you need it well let me let me tell you the story of when I uh when I first met him um it was uh this was uh this was this was an incredible experience.
41:59This is before his investment. We were, you know, we were starting to starting to work together and we got the opportunity to go to, I won't say where, but to go meet one of the one of the biggest car company in the world, one of their CEOs, and we met them together. We we pitched things and, you know, it was, of course, a really excited response from that CEO for what we're doing. And and after that, we we flew to London and managed to get in one of our cars from the airport and drive autonomously into central London. And that was such a cool experience to be able to go there, work together on, of course, winning business and then to be able to experience our product.
42:39And it was the first time we'd really done any highway or at that time it was quite late at night, nighttime driving. You're nervous? Well, it's been an interesting journey where the most exciting thing for me is seeing a car do something new for the first time and being able to have a conversation. I've most of my meetings tend to be in in in the car while we're driving autonomously and being able to have a conversation while I'm watching what the car does. And early on, of course, when we are still learning, I used to be very nervous about what how the car would behave and how the AI would behave because it's still learning basic concepts.
43:17But today, the performance is so awesome that often you can have a conversation and you forget that you're being driven autonomously. and then every once in a while you see something that's really interesting to watch some multi-agent negotiation where you've got different road users trying to um push their way through you or something like that and and then you can of course pause the conversation and look at what's happening around it it makes it really entertaining next time we can uh we can do a conversation in the back of the car yeah yeah sounds good okay so um like masa is he good at small talk like how do you spend like that many hours with someone like that or is it all just business business business It's business, it's vision.
43:51I mean, he is one of the most visionary guys. And I mean, you can read in the books about him, about the kind of passion he brings. But he also has deep focus. And this is something I've seen with a number of different, you know, incredible, iconic tech leaders that I've been able to meet. But then the other really interesting thing is that Master is such an amazing networker. I mean, a lot of his early relationships were built from bringing, bridging Silicon Valley tech leaders with Japan. And even today, if you look at the relationships and meetings he's had with people around the world, he is someone that just really has unparalleled access.
44:27And so for us, it was a perfect partnership to bring together for our Series C financing. And how do you get to the number? You know, is that a master driven thing or is that an Alex driven thing? It's both. It's one of those ones where, of course, we have a we actually we just brought on board on on board our first CFO, which is awesome guy to work with in Max and has given us, I think, a great more deal of insight into into our financial projections. But of course, we had we had those and there's various trade offs of do you incrementally raise rounds? And I think one of the things I've looked to do as a deep tech company is making sure we have the capital on our balance sheet.
45:05That means we don't need to build a next investor demo, but we can focus on what the long term product vision is. And so for me, the opportunity to find an investor that's willing to take venture risk at this level of scale and really back me as an entrepreneur. It's the most magical feeling for this kind of horizon. and so it was really exciting for us and means that we've been able to of course give our customers the confidence that we're not just going to get this product shipped but we've got the capital to really grow it and we can optimize long term and continue to drive this AV 2.0 vision that we're building so this is really how it all shook out and as a geek raising a billion dollars plus what was it like meeting Bill Gates?
45:50oh I mean he's very similar right like in terms of his curiosity, his, I think, intensity and drive. I mean, every time I've met him, he'd be the kind of guy when you're sitting opposite and he's got his notebook and he's furiously scrolling notes and sort of listening and learning about what you're saying. And then he'll come in with a sharp statement. And first time we met, 15 minutes into the meeting, he cut apart my business and picked up two or three most pressing issues as someone completely coming in fresh. And it's just this amazing ability to study and learn. And I think that's something that I really admire.
46:30I got the chance to go to go visit his house in Seattle as well. And he's just yeah, I think I think what he's built is really inspiring. And and being able to to learn from these kind of people is I think something that it's one of those things that it's when I go back to New Zealand and talk to my friends back there, it's quite hard to relate to these kind of experiences. But I do. It's one of those ones where I feel so privileged to do what I do every day to learn from those kinds of people. And I hope to build a business that can earn the right to be in that echelon. I certainly think we've got the opportunity to do that.
47:09Yes, I think raising a billion dollars, you know, that's more than most companies have ever done, including, you know, Microsoft's, Apple's, etc. There just wasn't that kind of fundraising environment ever then. So how do you sort of sincerely put together a plan to spend that, right? Is it, you know, even imaginable or does stuff sort of just get broadly put out there? Like you said, you know, it's just so we don't have to raise again for a long time because too much money can be a problem. Yep, yep, it certainly can. And if you look at the companies that have been, let's say, overcapitalized, I think the biggest failure mode is letting that money destroy your culture or spending unnecessarily or losing focus.
47:50And this is one of the things that I mentioned that I, you know, I prepare very in-depthly for, you know, important meetings and things like that. And I studied a lot before meeting SoftBank and meeting MasterSan. And one of the things that was very clear is that raising this kind of money, you need to be very careful about how you deploy and allocate that capital. And one of the things I'm really pleased with at Wave is that we've been very disciplined since that financing and still got the majority of that money still, almost all of that money still sitting in our bank account. And we're now starting to ramp spending in a way that because now the market's opening up and we have the opportunity to go and go capture I think and become the platform across the automotive market but being being very disciplined around that being gated with our spend and you know being very deliberate in that spending plus also I mean this this culture may you know starts with me but it really needs to go across the company and making sure that we are taking the right decisions and that everyone's empowered with transparent knowledge to be able to make cost versus speed versus quality trade-offs that of course is the guts of all good engineering.
49:07And then having this frugal culture from the beginning, I remember when we spoke of the house we started in, we used to nickel and dime every set of cutlery we bought from Amazon or something like this for the house and making sure that we are really being still scrappy at the size and building in a way that's efficient and cutting through red tape, that culture still remains. They're having a high ownership culture, high empowered culture, and of course being very deliberate about where we want to scale up, which we're starting to do based on commercial traction. These are the set of cultural principles that I think are putting us in good stead, and I'm certainly really happy with how our team has responded to the super capital.
49:53I think we've used it as a compelling force to go out and win the market. I mean, just because you're a good student, engineer, innovator, and then essentially entrepreneur, that doesn't mean that you are necessarily going to be an incredibly capable leader for a team of three, a team of 10, a team of 100, a team of 1 ,000. So how have you had to personally adapt your style to meet each challenge? Oh, I've learned a lot of things the hard way on that journey. But yeah, it's interesting. I mean, I look at formative things in my life, like I had the chance to captain field hockey, a sports team, you know, through high school and university.
50:37And I really look up to high performance sport for inspiration of excellence. And I learned a lot from those kind of experiences, or even during my PhD, It was interesting. The very first conference I went to when I was talking about these ideas of end-to-end learning, it was actually in Santiago in Chile, of all places. I went out there and managed to pitch some of the research I was doing. It was one of those conferences where about a few thousand people. And what I noticed was there were actually, at the time, there were a handful of people building similar things. This was the very first semantic segmentation system that was done with the end-to-end deep learning, the ability to figure out what is where in an image.
51:22And there was many, many of us sort of pitching these ideas. But one of the things I learned is the ability to tell stories, to be able to put out demos online, open source software, to be able to. A lot of my best ideas I get when I actually give a speech or a presentation because you're forced to think about why you're doing something and explain it to people. And so I actually think storytelling and building a narrative is a core skill I remember learning then. And it's such a critical part about building a company like this, because when you think about, I mean, we're eight years in and we're just getting started.
51:55So to be able to explain, rationalize and inspire that kind of journey is, I think, a really critical part. and then there's aspects of you know I do think of myself as a as a servant leader and and you know a lot of what I do is set up to unblock amazing people to go and teach me something and drive the company forward but learning when and how to to you know drive disruptive opinions to to shape a team I think that's that's one of the key things that that certainly I've learned over time and it's changed like I remember the first year when I was doing a lot of the coding to every year, my job certainly completely changes.
52:35And that's one of the beauties of getting involved in a startup and going through one of these growth journeys is that as soon as you're good at something, you've outgrown it and you need to learn a new skill. I wouldn't trade it for anything. Amazing. Okay. So been meaning to ask this question for a while, waiting for the right moment. There's never a right moment, but I'm sure that's been asked you a lot. the moral question of how to design a system with a very challenging moral philosophy scenario like grandma is driving the car five-year-old child is in the road decision has to be made by human grandma human grandma may choose to sacrifice their own life to save the child but what does a machine do in that scenario?
53:22And part two, how does that scenario change if grandma's also driving granddad and a couple of other people in the back? Broadly speaking, this is called the trolley problem. And there's some great research on it. One of my favorites is the MIT moral machine from MIT. You can look it up. But essentially, they survey and look to understand what society's viewers on these kind of problems. But the first question I ask is, How many times in your life have you encountered a situation like that when you're driving? Never. That's the main point to make here is that these kind of scenarios are typically so unlikely that they are very important academic discussions, but not of practical importance to us as an industry today.
54:11because the reality is is that to build great road safety the requirements are really a few stages removed from that and and what we design into our systems is that every point in time during the drive we should be able to initiate a minimal risk maneuver which means that there's always a third option which is to safely stop the vehicle with minimal risk which might mean pulling over or stopping in lane or something like this and so we never we build the systems to never get into a situation where you're making a decision like that. Now, from an engineering perspective, even if in some extremely unlikely scenario that you did get into that, there's a lot of uncertainty around, can you actually determine what are those two things?
54:51And often you end up with principles that are, you know, don't do anything erratic, maintain current course of action, remove kinetic energy, slow down and just try and try and stop. And taking erratic decisions like that under extreme uncertainty, extreme unlikelihood is likely a bad engineering decision. It's not to say that as this technology improves, of course, with millions of cars driving around the world, there is some non-zero statistical chances will happen. And so there again, I think we need to look to society's morals, expectations and principles like the ones I shared. But yeah, it's a fascinating topic to read up on from an ethical standpoint, which I'd point to the academic literature.
55:31But practically speaking, we want to design the situations that design our autonomous vehicles significant to their situation. It's interesting because practically, obviously, you want to do that because who wouldn't want to do that? But when there, you know, there will always be a scenario where you have to answer that question in reality, whether that's with regulation or whatever, because there are doomsdayers and naysayers whose job is to bring up these things and make sure that they're included in policies for companies. So at that moment of having to make that decision, is the right approach for you as a company to you know lean on other moral philosophers and say well you know based on what they've said that's our assumption because we're not moral philosophers we don't understand how to make that decision or is it to employ people like that who do literally sit around thinking about this challenge that that so that that's a good question i think um because it's a hard thing for you to answer but it's all your responsibility so sort of like no one's expecting you as a CEO to necessarily have the depth of a moral philosopher who does literally think about that as a career, but still your responsibility to get to a framework that can answer it, right?
56:42It is. And I think we need to draw on a number of different things, right? There is the philosopher's writings and studies, academic literature, there's regulation, there's legal liability. And we need to understand all of those and have the right expertise in our team to understand all of those and then of course on the engineering side you have the the need to be able to understand what's in the data and learn from it and actually then build a system that complies with those so i do think there's an expectation for us to have expertise across that space and ultimately bring bring together the product that makes the right choices across all of those dimensions it's i mean one thing to reflect on is robotics is such a multi-stakeholder problem it's not just AI engineering, but it's data simulation, public policy, hardware engineering, vehicle manufacture.
57:28And then, of course, you've got commercial operations, insurance. I mean, it's truly a cross-functional technical challenge. And we mentioned before, it's one of the reasons why I call it the space race of our generation, because I think it is one of the greatest technological problems that brings us together. And I guess, you know, other side of the doomsday scenario people do get worried about how hackable the stuff is right almost back to the point of uh steering wheels you know why do you want the steering wheel i mean i'm pretty certain that's almost like a stupid point because if it was hacked then steering wheel wouldn't help you very much either but it's one of those things that normal consumers like myself that's where your head jumps to right how hackable is it and therefore is it possible to revert to a fully manual mode just in case well i mean i i love our head of security he is at the forefront of all of this and and there are some really well established security frameworks to work with now especially for automotive there's been a lot of thinking around connected cars and and how to build the right protocols about to communicate with them but also give them the ability to ensure that everything's got the right level of encryption and ring fencing that when it actually operates on the car.
58:42So there's many different threat vectors when it comes to security. You can think about direct access to the vehicle, fault injection or even externally. Of course, it's been documented that some people have tried to adversarily behave around autonomous vehicles, even externally to the vehicle, just through pedestrian and other vehicle behavior and things like this. And so there's a lot to manage there. And again, it's a multi-stakeholder problem. But from an engineering perspective, yeah, there's some pretty robust techniques we have now to ensure the cybersecurity of a vehicle. And it should be that expectation that that's a core pillar of our product.
59:20And I think that's certainly an existential threat to the industry that we need to continue to take very seriously. So earlier in our conversation, you mentioned San Francisco and Shanghai. So how far behind is the UK and Europe from the US and China? I think it's two dimensions on the first dimension in terms of product availability clearly we're behind you know we are doing closed testing in our company but there's no public service here so that's that's certainly behind but from a technological opportunity I think what we're building is is truly next generation it it breaks through the glass ceiling that all of the those products have and enable something that can truly scale to the world so it's it's one of those ones that retrospectively it was quite good for us to start outside of the bubble of Silicon Valley.
1:00:07But today I really see Wave as a global company. We have offices around the world from Germany to Japan to Silicon Valley and we work with automakers around the world and global talent. And ultimately this technology is not just going to be in one country and we want the platform to exist worldwide. So I think we have an opportunity to leapfrog and that really comes down to the incredible talent and innovation that we see here. What I would love to see more come out of the UK and Europe is the ability to take this innovation and build it into products and create, you know, large-scale companies that can attack trillion-dollar markets.
1:00:44And I hope to see more of that. But I do think Wave has a really, really great role to play in making that possible. So what's actually stopping it from being everywhere right now then? Is it law, regulation? Is it people just adopting it and being comfortable with it? All of these things are coming together. So regulation is either there or in some markets coming together over the next year or two. The technology today that our AI is at the ability to offer driver assistance, and it's an integration and homologational compliance exercise of getting it into vehicles. And we're working with our partners on doing that as quickly as we can.
1:01:22and then it's an exposure to data and learning to grow this up into driverless and so all of these things will come I can talk about the topological order of them but there are many different things the critical path here is really taking our AI and integrating into consumer vehicles so it can be sold and then learning from that experience to grow it to driverless operations and then again launching those vehicles and of course regulation and public engagement and acceptance needs to needs to come along the way but uh you know it's it's it's happening and i'm i'm uh i've got deep conviction that all of those things are uh progressing actually really rapidly uh together and um we're going to find ourselves in a world very soon with this product launched around the world how how soon do you think it will be for me in london to be using wave so first that you'll have the ability to purchase a vehicle and with that vehicle to have a assisted driving experience can you say the types of brands that we would be able to use or not yeah not yet not yet but but watch that space a number around the world of leading brands will have that very soon great i mean and you're gonna have to tell me privately so i won't buy the wrong car.
1:02:38I'm literally about to buy a car. So you get the right software update. And then look, shortly after that, our software will enable robo-taxi experiences and delivery experiences. So that's the order of sequence that things are coming. But it's soon. It's coming in the coming years. Yeah, as in two to five years? I'm not going to put a date on it. I can share the order of things that are going to happen. But we're working as quickly as we can. And actually, of course, we've got a very clear internal plan. And over the last year, we just keep pulling those dates in. It's accelerating. Okay. So how does global trade change then with all of these kind of updates, right?
1:03:18So if autonomous trucks make shipping completely dirt cheap and lightning fast, how does that impact other stuff? Do we still have conversations from Trump about the Suez Canal and Panama? Are they still important? How does everything change? I think it further improves mobility and interconnected networks by creating making them more safe more accessible and more efficient it's interesting to think about how autonomy will affect manufacturing where today manufacturing is often located in low-cost labor regions around the world but with autonomy there's an opportunity where it may be better to locate it around low-cost electricity regions and of course to be more co-located with local supply chain which can improve the efficiency and reduce the carbon miles that things are taking around the world so i think we're going to see broadly speaking a lot of and through efficiency maybe there's decentralization of this as well maybe you can get more local personalized experiences through more intelligent machines and you're not needing to spend millions of dollars of capex to build a production line to produce at mass scale, but actually we can manufacture much more customizable products through autonomy.
1:04:33So there's a bunch of opportunities like this, but yeah, ultimately I see it as really opening up and creating a whole new set of possibilities. And with that, I think new products, new jobs, new experiences, all of these are going to come along the side. In 20 years, driverless technology, obviously you could take us to many different visions of society. What's the best case? What's the worst case? Oh man, I'm all immersed in the best case, which I think is great. Do you not have discussions about worst case at the company, even when you're like five beers deep on a team night out? Look, I think I'm pretty convicted that autonomy will be here.
1:05:16I think the question is around timing and how quickly the adoption is from initial experiences through to it in a trading. all walks of life, particularly in ground transportation. So I think that's coming, but we can argue around the timeframes. I think the number one thing for us to continue to be very cognizant of as an industry is safety. I think that rightly so, autonomous machines are held to a higher level of safety than we are as people when we drive. And I think making sure that we have systems that have the intelligence and our ability to actually simulate and prove that that level of intelligence exists so that we can feel really good as society, as a technology producer and as regulators that allows us to launch this technology.
1:06:09I think that's the thing we need to get right because, you know, sadly, there have been incidents with other companies autonomous technology in the past and and you know if I look to the airline industry if we can collectively learn from those as an industry and make sure that make sure that we put procedures and technology in place to to mitigate those those risks in the future I think that's the thing to keep the keep continue to keep an eye on okay won't blame you for not giving me the negative fearful side fair play do you want to share I mean what's what's your negative view? Well, I don't know.
1:06:44But instead, I'd love to know the opposite side. So give me your most optimistic, amazing view of what it's like for your average person in London in 20 years in a world where Wave has been completely phenomenally successful, achieved all your goals and driven this autonomous revolution. What does city life look like what does life look like in britain as a result of that oh yeah let me let me riff on a couple of points i mean i dream of healthy streets where you've got no car parks blocking the road but you've got great architectural interaction between footpaths vulnerable road traffic like cyclists and shopfronts you've got much more space opened up for for green space that improves the well-being of cities.
1:07:34And then, of course, you've got mobility that all of the lost time that we have from parking, from looking after our vehicles has really shifted towards autonomous, efficient, shared services. And ultimately, when we're in cars, it could become a complete infotainment experience where we can or we can rest or we can actually use that space to take meetings or to consume content to ultimately get that time back. And then the improvements on adjacent industries, whether it's health care or even just connectivity of people or freeing up the choice of where people want to live. I think all of these things can be opened up to.
1:08:18And then outside of autonomous driving, I think in our houses, I would love to have more support for some of the domestic tasks at home through automation. and then I hope that not just wave becoming a company that a trillion dollar company that is able to open up this whole market but I hope to see more deep tech success stories that come out throughout the world you know building in this in this industry and and beyond that's the that's maybe one picture of success what would you add well I guess what I'm interested in what about you know the the daily wins for someone like myself or you know my grandma what would life be like for us, what kind of ways would our lives be improved?
1:08:59Well, my grandma just she's 93 years old back in New Zealand, and she just lost her driver's license. And so she would be able to still get around, get around her home city without the ability to drive herself. She'd be able to get in, she'd be able to talk to the driving AI and ask it to drive where she needs to go and still have that ability to move around. um uh you know sadly she was involved in a road accident a few years ago and um uh you know not having that ever happen and actually having a world where road safety is is uh you know it's just taken for granted that it's it's um it's near perfect uh and i think this is this is going to be a fundamental change for the accessibility of elderly people yeah i think that's really interesting and obviously disabled you know you mentioned like mobility disabled people people who can't drive don't have access that in itself is a better world amazing dude incredible amazing vision really really interesting i am fucking thrilled for you i think it's awesome and also as a brit i'm so proud that this is happening here in london uk success story i think europe really needs it and there are quite a few um green shoots of positivity that the uk needs at the moment that can truly bring out the best of what we're really good at in Britain.
1:10:20I think there's been a lot of negative PR over the last couple of years about Britain's place in the world and how Britain's bearing. And, you know, we had the founder of Zilchin last week, you know, another UK unicorn. And we were talking about, you know, actually, statistically, we still got some of the best universities in the world. We've got some of the most amazing AI companies here. We've got incredible talent. We just need more incredible visionary leaders who are building here and creating success stories like that, because I really feel like that creates momentum for the next generation and helps us build life-changing and meaningful businesses here.
1:10:54So thank you for being part of that revolution. Oh, cheers. I appreciate it. And I can't wait to see the wave keep growing.
From the publisher
In this episode of Secret Leaders, we sit down with Alex Kendall, co-founder & CEO of Wayve AI — the $1B+ British startup taking on Tesla in the race to build a driverless future.
From raising $1.05 billion from Microsoft, SoftBank, and NVIDIA, to reimagining cities with AI-powered vehicles, Alex shares how his company is changing how we live, commute, and move goods.
- Why do 1.2 million people still die on roads every year?
- Why are cars asleep 96% of the time and what if they could work for us?
- Could AI make driving safer than humans ever could?
We dive into how Wayve’s embodied AI competes with Tesla, why public trust is still the biggest barrier, and how your car might soon become your second living room.
--------
Sign up to Wise Business banking:
https://wise.com/uk/business/
Join Vanta and receive $1000 off:
http://vanta.com/secretleaders




