Waymo

These 50 companies exemplify the innovation, progress, and potential of AI.

AI Innovators 2025-Waymo

Fortune 50 AI Innovators

Get insights into Waymo including business profile, stock information, news, and rankings from Fortune.

Waymo Company Profile, Stock Price, News, Rankings

legacy

waymo

Companies

Waymo | Fortune

Waymo Company Profile, Stock Price, News, Rankings | Fortune

Angelica Ang

Waymo to launch robotaxis in Singapore, its first Southeast Asia market, by 2028

/2026/09/17/waymo-launch-robotaxis-singapore-first-southeast-asia-market/

Sasha Rogelberg

Americans are so concerned about autonomous vehicles that Minneapolis is considering requiring drivers to sit inside Waymos

/2026/09/16/autonomous-vehicles-minneapolis-city-council-waymo/

Marco Quiroz-Gutierrez

Uber’s CEO hints at cheaper rides after the company announced 3,300 corporate jobs

/2026/09/14/uber-ceo-dara-khosrowshahi-cheaper-rides-3300-job-cuts-layoffs/

Andrew Nusca

What to watch for as Apple rolls out new Macs

/2026/08/26/what-to-watch-for-as-apple-rolls-out-new-macs/

Google’s founders didn’t market test Alphabet’s name before launching the now $1.9 trillion juggernaut. Here’s the advice Steve Jobs gave Larry Page

/article/what-was-steve-jobs-advice-for-larry-page-sergey-brin-google-ipo-anniversary/

https://fortune.com/2026/09/16/autonomous-vehicles-minneapolis-city-council-waymo/?itm_source=parsely-api

Jacqueline Munis

Alphabet CEO Sundar Pichai’s new $692 million compensation package hinges on the success of two Google moonshots that aren’t making any money

https://fortune.com/2026/03/10/google-ceo-sundar-pichai-692-million-compensation-waymo-wing-success/?itm_source=parsely-api

https://fortune.com/2026/09/14/uber-ceo-dara-khosrowshahi-cheaper-rides-3300-job-cuts-layoffs/?itm_source=parsely-api

Google’s founders didn’t market test Alphabet’s name before launching the now $1.9 trillion juggernaut. Here's the advice Steve Jobs gave Larry Page

https://fortune.com/article/what-was-steve-jobs-advice-for-larry-page-sergey-brin-google-ipo-anniversary/?itm_source=parsely-api

Michael Liedtke

The Associated Press

Waymo, formerly Google's self-driving car project, is coming to Miami—and you'll be able to ride in a driverless Jaguar

https://fortune.com/2024/12/05/waymo-google-self-driving-car-miami-driverless-jaguar/?itm_source=parsely-api

Uber and Waymo launch driverless ride service in Austin

https://fortune.com/2025/03/04/uber-waymo-launch-driverless-ride-service-austin/?itm_source=parsely-api

A scammer used AI to impersonate the U.S. Secretary of State

https://fortune.com/2025/07/09/a-scammer-used-ai-to-impersonate-the-u-s-secretary-of-state/?itm_source=parsely-api

AI has moved beyond the chat window to the physical world. Emboldened by world models and enabled by new hardware, machines can see and act and operate in real environments. The landscape is shifting fast. Cathy Hackl breaks down where spatial intelligence is going and what it takes to compete.

Cathy Hackl, Founder and CEO, Future Dynamics

Here in the Real World

Conferences

That that's a hard, that's, that's a hard channel to follow, guys. OK, uh, I am a hyper specialist in spatial computing, gaming, and simulation, so that's what we're gonna be talking about. I am gonna keep it in the real world, but we're gonna talk about AI and everything happening in that space. That was, let's go. See, um, I think many of us can agree that the next, the next AI revolution won't happen in chat windows. It's going to happen in the physical world. Obviously, many of you guys are fixated in large language models and agentic AI, but I'm gonna take you guys a little bit further when we talk about expanding computing into the physical world. Um, how many of you have seen this slide before? Did you see some people, OK, uh, I like showing this slide because it kind of sets the tone because you've heard a lot of terms being thrown out here, right? Agentic, you've heard physical world models, all sorts of things. Just kind of level sets kind of where at least Jensen and where Nvidia thinks this is going. I think most people agree generative AI, you know, no longer necessarily the talk of the town except for IPOs. Agentic AI is the darling of the moment. Um, and then where we're heading is physical AI. You heard a lot about physical AI here on stage. Um, I focus specifically on that agentic to physical AI moment. So, LLMs, um, I love that we had a panel here on robotics, uh, humanoid robots, and, and thinking through, you know, who would you trust? Would you trust to have a robot in your house? If I ask you right now, if you can have a, a robot at home to fold your laundry, how many of you would, you know, raise your hands? Most people, would you give that robot a knife? No, would you have that robot take care of your newborn? No, right? Because it hallucinates. So, a lot of what is happening right now is LLMs can predict the next word, but what we need with physical AI is AI that can predict the next action. And this is where a lot of this is slowly heading. For that physical AI moment that we're heading into, we need a completely new AI stack of sorts. It's going to need a body, we need an interface, we will have tons of new hardware. And here's what I'm here to tell you, the moat will no longer be in the models. Alright, maybe some specific models, but the mode is moving away from models. We're going to have tons of hardware. I'm going to get into that in a little bit. And also we're going to have a moment of data scarcity depending on the models that are needed to get us to that new physical AI moment. We're also going to talk about connectivity. So agentic AI, you know, no tech no tech conversation would be, you know, would be OK if I didn't mention agentic because everyone wants to talk about gentic. I see agentic as the nervous system of where we're heading with physical AI. Agentic eventually will need a body, right? Whether it's a hologram, a humanoid robot robot, or something, it's gonna need some type of body, and it's gonna need some type of interface beyond just a chat window. Right, I see that as the connective tissue, the nervous system that gets us into that next phase of physical AI. Uh, this is a, this is something that I love to show because I think most people are not even understanding kind of where this is headed. This is from my friends at Niantic Spatial. Um, if you talk to anyone working in deep tech and in deep AI, we will all agree that in order to get to that next stage of AI, you need new models and new architectures. And some of these, these new models and architectures are foundational models, so world foundation models, and then also large geospatial models, which are at the top there. Most people don't, don't really think about large geospatial models. I am going to talk about that, but as you go up in this pyramid, there's more data scarcity. And this kind of explains, this is also from my friends at Niantic, that explains a little bit where the models are. This is a great slide to take a picture of, if you want to better understand where the models are heading and who's doing what. So world models, show of hands, how many of you here know what a world model is? I mean, you've heard the terms, OK. All right, but pretty much, I'm not gonna kind of dig into it 100%, but I will tell you it's about simulation, but it's about simulating the next frame of reality. Right now, you have tons of capital. Slowly going into this area, so Doctor Fayei Lee, who many of you guys know, she's the godmother of AI. I'm known in tech circles as the godmother of the metaverse. I know it's not sexy to talk about the metaverse anymore, but I feel a certain sisterhood with her. Uh, she's working on this. You've got Jan LeCun, who used to run AI at Meta, now, uh, running his company. He raised, get this, most of you guys know this, but he raised $1.1 billion seed round. That's unheard of. Right, in the age of AI. So, a lot of money flowing into world models, because they can predict the next, uh, the next part of reality. It is about simulation. So, most of what's happening with world models is creating simulation, and there'll be simulation in virtual spaces, also simulation in the physical world. So, you've got a lot of agents. You might have seen a lot of work around agents, uh, doing different types of modeling and simulation. Uh, you also start, start to move from language to physics. That's why a lot of the money is going into physical AI into world models. It's because we're moving away from language to physics. OK. And anyone that has kids recognizes this Roblox. You know, we all work for Roblox. Roblox is actually entering the world models space. You've got Waymo also talking a lot about world models, and you also have the company Runway, which many of you know for video. They're also now a world model company. So, geospatial AI, this might be the first time you guys ever hear about geospatial AI models, but this is a very important part in order to have those humanoid robots, uh, autonomous vehicles, drones in large quantities, because it's going to allow creating that layer of AI that connects, connects us to the physical world. So AI that understands language is powerful, but AI that understands place is transformative. Place in real time. There is something called the ground truth advantage. So, a world model can predict plausible outcomes, you know, what might, what might happen in a street corner, but when you have real-world data, that actually tells you in real time what is happening, that gives you an even better outcome. And this is the type of data that is irreplaceable. And there's, there's going to be a huge, and I'll tell you this, there's going to be a huge economic incentive towards more real world data. So for example, the information that Ryan might have from real estate or booking might have about the physical world, that they're sitting on a treasured trove of data. That leads us into spatial computing, where I spent the last decade. It's expanding computing into the physical world. And why are you starting to see spatial pop-up everywhere? Because robots need to have spatial intelligence. They need to be able to understand the physics of the space, whether you're a car, a robot, a hologram, a drone. You need to understand the physics of the space. Here's where we get into smart glasses. Is anyone here wearing like the Meta Ray-Bans? I see a couple of people I get asked this question all the time. Why in the world are all these companies rushing towards new wearables and new devices? It's because slowly, someone, we need to replace the mobile phone. And for several reasons, they want to replace the mobile phone, because whoever owns the hardware has access to that real-world data to train those newer models to get you to physical AI. Right? So, whoever owns those, those wearables has access to that. And also, if you add to that wearables as the interface and you add neurosymbolic AI that actually helps to reason, you can start to eliminate hallucination. The reason many of you guys would not give that robot a knife or have him take care of your newborn is because this technology still hallucinates hallucinates. We need to get to a point where AI is actually getting everything right with extreme precision, and in order to do that, you need spatial computing, you need geospatial models, and you need world models and LLMs. So, slowly, everything's going to start to move from the cloud to the edge. Right now, there is, the, the connectivity does not exist for us to be able to power millions if not billions of devices altogether. That is being built. The infrastructure is being built, but it's no longer, it's not here yet. Um, I love that we had Peggy Johnson here. I actually spent some time with her in Doha at an event there and had a chance to talk to her about humanoid robotics and the big rush that's moving towards humanoid robots. And this is not me, this is Jensen Huang saying humanoid robots are going to be the next consumer market. Just for a show of hands, how many of you agree with that sentiment? Alright, alright, so there's a lot of skeptics. Uh, we'll have to see really what happens, but I do truly believe we're at a like kind of a, a dawn of new hardware. And this is the big thing, intelligence is slowly entering the physical world, and it is becoming spatial. So, the spatial understanding that we have, that we gain as children, for example, you wouldn't put your newborn, you wouldn't put your 3 year old in front of the stairs and be like, hey, go at it, right? Slowly, children learn to go down the stairs and understand the physics. That's kind of where we're at right now, with a lot of these robots and a lot of these devices, is they're starting to understand the physics of the space. Right? So, we're moving from language into understanding physics, and that's the only way we can actually get to physical AI. A holograms is also some of the work I've been doing for a long time. When we talk about agenttic AI, like for example, when we were talking about, you know, talking to an agent about booking something for you, eventually you're going to want to see that agent, and that's going to come in the form of potentially a hologram or some device. Also from a communication standpoint, we will stop eventually doing phone calls necessarily and start doing hologram calls, which is very sci-fi, but it's truly being worked on once we have the right connectivity. So, the infrastructure is currently being built. The networks are being built to be able to power these systems. The interface is arriving. And this is not me trying to replace the mobile phone. This is everyone in Silicon Valley. For example, Apple is going to be bringing AirPods with cameras. Why are they doing that? Because once they have cameras, they can start to map the physical world. Right, Google's back in the market with glasses as well, so the moat is moving from the models to the hardware. Why? Because that's how you get the training data for the newer models. So I truly do believe that the next advantage and where a lot of the opportunity is going to lie is in building for the physical world. Thank you.

The dream of autonomy has become reality. With robotaxis now navigating real traffic, Zoox is showing how AI can master one of the world’s most complex systems: the city street.From breakthroughs in edge intelligence to safety engineering, the Amazon-backed company hopes to redefine how and where the world moves. Which problems are still to solve? And how will the company compete with more entrenched rivals?

Your Autonomous Ride Has Arrived

Good morning. Thanks for being with us, Jesse. We don't have a ton of time, so I want to tell you from the jump. I'm going to try to get at least one question from the audience, so please start thinking of that now. Um, let's do a quick show of hands. How, how many folks in the room have driven in a or rode in a Waymo? Anyone? A lot of people? OK. Has anyone been fortunate to ride in a zoos yet? One? Anyone else? No? OK. So let's, let's just start there, um. How, how should we think about that disparity right now and how I understand it's the early days, but how you close that gap. Yeah, absolutely. So you know Waymo has been at this for quite a bit longer than Zoos, and we also took a little bit harder path of building our own vehicles from the ground up. Sounds a little bit harder, a little bit harder, not that, not that what Waymo did was easy. It's also incredibly hard, and they're doing an awesome job, um. But we did take that harder and longer path, and we actually are pretty proud because we think we've been able to close the gap quite a bit, but of course there's still a gap. So we just started taking the public off of our waitlist in San Francisco over the last couple of weeks, and we still have a very big waitlist. We are open to the public in Las Vegas and we have been since September with the public launch over there. It's still a preview service and it's free, but it's going really well. We've had over 100,000. and we're actually excited. Just just last Thursday we crossed 1 million miles in the Robotaxi. That's the actual robotaxi, you know, no driver on public roads just this year calendar year 2025. So it's coming, it's going super well, but obviously we have more work to do to expand our geo fence, add a bunch of vehicles, and then take everybody off the waitlist in San Francisco. When, when are you able to start charging for these rides and, and also when, you know, talk to me about the expansion. Both in other markets and sort of what you can do within your current markets going forward. Yeah, so it's a little complicated, so we need approval on both the federal level and the state levels. And so for example in Nevada we already have approval from the state, but in California it's a little bit more complicated because you have to work with the CPUC and there's more steps there so. Moment we are also waiting on the federal exemption, the Part 555 exemption from NHTSA, which we're working with them on as we speak. We expect to get that hopefully early next year and what that would give you is the ability to start charging. That gives us the ability to start charging from a federal perspective, and then we can quickly start charging in, for example, Las Vegas. Uh, and then in California there's a couple of additional steps with the CPUC. We're working with them as well, so charging in San Francisco would be coming later in 2026 in California as opposed to earlier in 2026 in Nevada. I was that's a good, that's a good point, uh, time to point out that you are a subsidiary of Amazon. Amazon typically likes making money, so I was going, I was going, you know, safe to say, and um. I was going to ask how confident you feel in sort of the long term runway you have, um, but it sounds like revenue coming in next year will be a start. Absolutely, and I want to be clear, I mean, the amount of revenue will make next year won't really pay for Zoos. I mean Zoos is pretty expensive and you know we're a few 1000 people and building a lot of robo taxis, but it's definitely a start and a proof point. And then you know over the next few years it'll start to be a really interesting business because the revenue you can generate from a robotaxi is quite a bit more than the expense to run a Robotaxi, including amortization of the bomb cost as well. So we're really excited about that, but it's still several more years before we can operate at a scale where you know that's financially interesting to Amazon. Um, let's talk about the technology because part of the reason, um, there is that gap we talked about for now with, with a company like Waymo is you've built your own vehicles, um. Can you give the the sort of queerest um Uh explanation on why you did that and why you're confident that's still the best path forward. You call them purpose built vehicles, right, versus retrofitting an existing car. Yeah, absolutely. So cars have been designed over the last 100 years for humans to drive them. And so all of the choices, their shape, their architecture, what components they have and don't have, they were all designed for human drivers, and you can modify them and add a little bit more redundancy and you know maybe remove the steering wheel, maybe not, but you're still ending up with an architecture that's far from optimal for a Robotaxi. So we're able to create a much nicer customer experience inside the vehicle with carriage seating and features you don't find in a car, much more room inside and a much more compact. Externally, which is great. We're also able to add a much bigger battery. This is really good environmentally and economically because you don't have to be charging the battery multiple times a day like you like you do on a retrofitted car. You're also able to have much better safety and redundancy. So not only can you be safer, but you can also ensure that if a hardware component fails, you can still take the customer to their destination, and these are all things you can't do on a retrofitted car. So in the future, and I'm going to come for a question in a second, anyone has one in the future. In your mind, someone would choose a zoos for a ride over Awamo or um we can talk about that company, Tesla, um. For what reason or reasons? Oh, it's just a much better experience, right? So, so your, your time in the vehicle is dramatically nicer. You have much more space. If you're with friends, it's dramatically more social because you're facing each other you're facing each other and, and you know, trying to put multiple people. In a regular car is not not very nice. And even if you are by yourself, I mean, you know, you have that, that front seat in front of you, it's just, it's just again it was very much designed for, for human drivers. So when you do not have that constraint, we can build something much, much nicer, and that is exactly what we've done. Um, do we have any questions? Otherwise I can Yeah, we have one here um just wait for a mic please and tell us who you are and. Where you're uh Where you work up. Oh yeah, right here they're coming. Hi, my name is Ashley. I'm with Microsoft. Um, my question for you is, since you're a subsidiary of Amazon, why focus on RoboTaxi and not just the last mile delivery component? Like, it seems a pretty crowded field, so why not just go into the space where we know Amazon has been a dominant player. Well, we don't really think that Robotaxis is a crowded field. There's really only one company out there right now in the US that's doing Robotaxis, and so that doesn't seem very crowded to us, and there are actually no companies doing a purpose-built roboTaxi anywhere in the world yet. So we think there's a really exciting market for us from Amazon's perspective. It's not that they wouldn't like to automate some aspects of package delivery. They're obviously working on that. But the economic opportunity and the opportunity for customers to create this whole new category of transportation is actually much more exciting and even more financially compelling than simply taking something they do today and saving a bit of money so um. We'd love to save money, obviously, but that's not as exciting as creating this whole new market, and the addressable market for moving people around cities is just profoundly huge, and the union economics of Robotaxis seem like they're going to be very compelling. So that's why Amazon and Zoos are really excited about that. Is there any um Delivery component on the roadmap, not explicitly. We do talk to Amazon last mile as they're kind of, you know, working on their autonomy plans, but we're quite separate, and we've we've been instructed all the way from the very top to stay laser focused on moving people around. It's harder to move people around than packages. In terms of what you have to do with the vehicle, but on the other hand, there are also challenges with automating packages, right? You have to get the packages in the vehicle and get them out of the vehicle, so it's actually not as straightforward as it might seem to do automated package delivery. Um, as an innovator, I'm curious about, you mentioned earlier, no one in the world is doing purpose-built, um, autonomous vehicles right now. Does that, that, that looks like a huge opportunity, but does that ever make your confidence waiver that it's the right approach? Not really, especially lately. I mean, I think it was. Very fair to be skeptical, you know, a decade ago when we were starting Zoos, but I think at this point we have now a lot of data from from the field from actual customers. I mentioned we've had over 100,000 people riding Zuxes, and these are not people we know. These are just people, and the reaction has been actually much better than we expected, which is great to see because we obviously thought people would love it, otherwise we wouldn't have built this thing, but seeing how much. People prefer being in the zoos over any other form of transportation that can get them around cities is super encouraging. And so our view is that people aren't doing this not because it's not a good idea, but because it's just really hard. It takes a lot of time. It's very cross functional. It's expensive, but I do think over time this is, this is going to be a much more popular way of doing urban transportation. I think I saw another hand. We have one up here if we can get a mic please on this side. Thanks for your patience. Just tell us who you are and uh who you're with. Hi, this is Ursh from Adeline. Um, I have like a two parter. How do you model the economic impact for zoos of zoos and What would you say Zooks is doing in 10 years from now? Is it still just Robotaxis? If not, how does the portfolio of products and services expand from kind of the foundational stuff that you've built today? Yeah, so the economics are super interesting. It's actually one of the reasons why we started Zoos in the first place is, uh, you know, cars are used 4% of the time. And that means that 96% of the time they're just sitting around taking up space and depreciating. Not only is that economically wasteful, it's environmentally wasteful as well because there's a lot of stuff that goes into a car. It's not just driving cars that's bad for the environment. It's making them as well. Making a robotaxi also bad for the environment. But if you have to make way fewer of them to serve the same number of people, that's a win. And so when we look at the economics, it's a, it's a really exciting win for customers because you know you only have to pay for the vehicle when you're using it. You don't have to pay for parking, insurance, depreciation, all that, and it's great for us because one vehicle can charge for itself all day and all night long. So those unit economics, obviously you don't have to pay a driver, are really compelling. Now it takes a lot. Of capital and a lot of work to get to that point, but we're very close to it and that's very exciting as far as 10 years from now, you know, look, I think you could hopefully assume we have multiple types of vehicles, different form factors, and I would say 10 years from now if things are going well, you know, we should also be in many countries, not just the United States, and we'll probably be in some other areas, you know, maybe we'll be working. On packages or other things like that, it's not that we're not interested in expanding. We're just trying to stay super focused on, you know, robotaxis for people because that's already pretty hard, and we believe that, you know, we can create a multi $100 billion company just moving people around if things go well. We have a lot of work to do to get to that point, but we do believe that's possible. Um, I do have one, question about Tesla. Um, you, you made a little news last year, maybe accidentally, um, you were asked about full self-driving in Tesla, and, um, and you said something along the lines and I'm paraphrasing, so please correct me of, well, it, it doesn't fully work and in that you meant, you know, you still need a driver in the car. Um, Elon has. Made some sort of promise which he's known to do or said that they will be removing drivers test their safety drivers from some of their rides that they're currently offering. I'm curious if your opinion has changed at all um based on. Um, what they're saying they're going to be doing soon in taking a safety driver out of the vehicle. Yeah, I mean they've been saying similar things for many, many years. I think their FSD is by far the most advanced. and safest driver assistance system in the world. I mean there's nothing that's even close, and it's a remarkable technical and product achievement. It's also pretty clear from the recent data that if you have an attentive human driver supervising it in many scenarios it's safer than, you know, just a regular human by themselves. So that's all wonderful. My point is that the. Difference between a supervised system and an unsupervised system, which means that it can be, you know, safer than a human with nobody watching. That's a, that's a huge difference. And so our view is that, you know, having the extra hardware, you know, active sensing and and really designing a system from a systems engineering perspective very rigorously is very important if you want to remove the driver, and we continue to believe that. Would you take a ride in a Tesla with no driver in it? the taxi. That's an interesting question, um. Maybe for a few miles, a few miles, OK, um, just to see what it's like. One final one for me and we'll try to squeeze in one more, uh, what. What AI advancement in the last year or couple of years had made has made the biggest sort of difference in Zu's technology? Yeah, I mean. It's incredible. It's happening in the AI field. When when we started Zoos, we did not anticipate, you know, AI broadly being this capable and you know, sort of so close to something AGI-like, and we don't think you necessarily need to have AGI to drive safely in cities. That's clearly the case. But since, since there is so much progress in that direction, we've been able to really automate a lot of our pipelines. One of the most important. And things we do is how quickly can we improve our software and validate that it's safe? We've been able to use these modern techniques to really speed up and improve the quality and precision of these offline pipelines that help us validate the software. We're also online using a lot of these modern transformer-based techniques, for example, to take in all of the sensor data and really just understand what's going on in the world rather than having to have separate pipelines. For cameras or radars or liars, it's also really improving the way we do prediction and motion planning. So we're really moving to kind of combine prediction and motion planning into some of these newer models still with some of the guardrails and costs associated with some of the traditional techniques we're really kind of doing a hybrid approach to really guarantee safety while leveraging a lot of the advantages of these newer techniques. So it's a very exciting time. And you finished right on time, Jesse. Wow, let's get a round of applause for Jesse Levinson. Thank you for having us here. Thanks so much.

James Manyika, Senior Vice President, Technology and Society, Google
Interviewer: Alan Murray, Chief Executive Officer, FORTUNE

Brainstorm A.I. 2022 - Can A.I. Help Solve The World’s Biggest Problems

Can A.I. Help Solve The World’s Biggest Problems?

Brainstorm A.I. 2022 - Can A.I. Help Solve The World’s Biggest Problems?

last september, our next guest announced Google's $25 million initiative to help organizations use AI in support of the U. N. Sustainable development goals, reporting straight to the Ceo James Manyika is tasked with assessing the impact of Google's tech on society. He also advises the White House on A. I policy in his role as vice chair of the US National Ai advisory Committee. We'll sit down with this season tech executive and techno optimist to explore how AI can be leveraged to address the world's most critical problems. Please welcome James Manyika, senior vice president of technology and society at Google for a conversation with fortune. Media ceo Allen Murray. Thank you James. Great to be with you. Good to be with you the last time we did this you had a different job. That's true. That was what five years ago it was pre pandemic. Well, uh congratulations, fascinating position you're in. Google has such huge ambitions in a I mean in the last couple of months I've heard talk about predicting natural disasters, dealing with climate change, healthcare, uh broad remit and healthcare, uh writing code et cetera. What is google's Ai mission? Well, first of all, I mean I think you have to go to the to the founding of the company. I mean at the end of the day it's a computer science information company. And I think if you look across all the things that Google is trying to do. I think they fall into a couple categories. # one a huge focus on actually advancing the field of AI so we can talk about the extraordinary number of breakthrough research and results. In fact a lot of the excitement we're all having now about generative models can be go back down to a particular set of research papers that came out of google, the classic attentions, all you need that opened up the space around transformer architectures that have led to these large language models. So the advancing the field itself is one of those goals. I think there's then there's also the practical goal about how do we actually bring these technologies to everybody. So through products, tools, infrastructure services, yeah, basically, you know bring this to the world and in the case of google is doing that in several ways, one is through google's products. So hopefully you've all noticed that search go better uh you know, google photos is much better etcetera, but also new products, but think about also the other companies if you like that google's birthing that would not have been possible. But for Ai so think Waymo think more fake, which is now a new drug discovery companies only made possible by the breakthroughs in AI So that's one said, I should also mention though Alan that I think google are also quite committed to this idea of how do we help others make use of these technologies, whether it's in on societal issues, whether it's climate change, flood forecasting science and discovery is another part of I think what google is trying to do and with ai embedded that deeply in the entire company. What's your job like how does that work? You report directly to the ceo you have responsible for a I had a responsibility for ai what does that mean? Well, so you know this is a dream job for me, alan I did my PhD in and robotics a very long time ago and to see the field come before you became a consultant, you're being very careful not to mention your former employer because even though Mackenzie, so if you look at some of the workouts doing at Mckinsey Global Institute looking a lot of the impacts of these technologies and air on business and the economy, I'm sure we'll get some of the business implications. But I think so my role is to connect the ideas and the research across all of google and alphabet and work with the team's leading that work to advance if you like. Our most ambitious research and innovations which include a lot of ai a lot of quantum computing and all these big computing infrastructure initiatives to make sure that they're being useful to people to society and that we're doing it responsibly and tie all the pieces together all the pieces together. So so you said when we were talking earlier, you said the last year and a half have been phenomenal year and a half for a I explain why well air is already making progress, but if you look at what's happened in the last couple of years I mentioned the transformer paper which has now led to this foundational uh, models that we're now talking about faith was talking about them. Look at what's happened with, for example, innovations such as uh, you know, uh, if Like diffusion models that have led to all these generative capabilities, I think the reason the last year has been interesting that I think we've, we've come to realize a few things. Alan one is these, these technologies have now become multi model so you can use them from text to images, texts too. Uh, code text, all these rights of application areas and similar. We've also discovered that in fact when you build, when you scale them up, you start to get properties and performance that goes far beyond what was possible. Before I'll give you a very good example. Uh, if you take translation, everyone loves Google translation, it works pretty well. And if you notice it says, you know historically cover covers about 230 ISH languages, it does probably 40 or 50 of them extraordinarily well before this last year and a half, that was kind of the story. Now we know that with these very large models, you can actually take languages that were historically called low resource languages where we didn't have the equivalent datasets to train them and you can translate those two. So this year, for example, because of these Large models, we added another 25 languages spoken by 300 million people like Bengali or Uganda, which people thought you couldn't really do those. And now we've just announced in fact we're gonna aim for 1000 languages. Why not? Yeah. When we were talking earlier about that, you said the science implications of this are particularly big. Can you give some examples? Yeah, I think the science arena fascinating things. You've seen what's happened in biology, protein folding. Our team at Deepmind basically went from, you know, this is a 50 year grand challenge where, you know, computational biologists didn't know how you can go from amino acid sequences to understand the structural proteins and every thought this was in theory Possible. So we went from zero between in the 18 months between, you know, the summer of 2020 from showing that you can do this with a I techniques to earlier this year and June predicting the protein structures of all 200 million proteins known to science. And we actually open That data set. So now you now have like, I think close to 600,000 biologists who are accessing these protein structure data set. So that science that's biology, you can go to physics. Uh if you if you care about fusion for example, it turns out one of the biggest challenges your infusion, high energy physics is how do you do plasma containment? Uh if you go to the Topamax in europe and other places are building. So it turns out you can actually use ai to stabilize uh you know, the plasma structures in physics, you go all the way to quantum computing. You probably saw last week we announced an incredible result in quantum computing, which shows that you can actually even create gravitational wormholes. So you can go in all these fields in science and I didn't mention mathematics. So we're starting to be able to show that AI can actually help you help us make extraordinary breakthroughs in this foundational fields that are going to be incredibly, what's the business model behind all that work in science? Uh Well, if you take an area like drug discovery in biology is straightforward, you hopefully we'll discover something extraordinary therapies and drugs we've now launched. I mean for google though. What? Yeah, I mean you you you talk about this almost like google is a university. Oh no it's not, it's a business. So the business model for us is I said come back to how we're doing it, we have a cloud infrastructure that's now ai enabled google cloud which is making extraordinary I think uniquely distinctive uses of AI. We also have products, core products, search and other things that are now, I think that's better, but also creating new products. So you're gonna see a whole range of new products that are only possible through a I come from, we have such knowledge in this audience and my knowledge is limited. So I really want you all to engage if you have questions for James or comments or something that you're particularly interested in hearing about, raise your hand and we'll get a uh we'll get a microphone to you right away so you can ask but but James, you're as you described your job, you're responsible for monitoring the interaction and the effects on society of all these massive changes in technology faithfully was talking about that a little bit earlier. How do you do that? I mean it's well to be perfectly my job is not so much to monitor but to work with our teams to make sure we are building the most useful technologies and do it responsibly. So to predict it in some ways to see it and to see the problems that may come in advance. Yes, but also make sure being we're building the most useful things now to be perfectly clear to your point, these technologies have come with extraordinary range of risks and challenges, right? We know that sometimes their own work is expected right? The leader to issues of bias and toxicity in their outputs and quite frankly non factuality. So that's a problem. There's also other issues that kick in such as for example the possible use and misuse of these technologies which you can imagine all the kinds of you know miss uses this could be put to. Then you've also got the kinds of things that we alluded to this, where we have to think about the impact in the labor market and the jobs. So there's a whole range of things we have to be thoughtful about. And that's why you probably noticed that in the case of google, google has been I think extraordinarily thoughtful about not rushing to throw these things out into the world without actually doing proper research and fully understanding what they're capable of. And these days, what at the top of your personal list of concerns? I mean when we were talking four years ago there was still a lot of concern about, oh this is going to destroy jobs, take my job away. I'm not gonna be able to work. Well we're 3.7% unemployment. Everybody who wants work can find work somewhere that doesn't seem to be the issue. But what are the issues that are top of mind for you? Well, I'll tell you the ones that I think a lot about. I think these issues to do when when they technology don't work as intended. So the, you know, the toxicity bias and so forth. And I think we're making progress on those things. I'm happy to talk about that. But I also worry about what I think of as the use misuse questions when is it gonna be appropriate to use these for what and quite frankly how we regulate that. I actually think, you know, this may surprise people, I think you'll find that, you know, I actually, you know, many of us embrace regulation because we're gonna have to be thoughtful about when is it appropriate to use these technologies? How do we put them out into the world so that we're not only, you know, not to throttle them. I don't think we should do that, but make sure we're using them in most useful ways. The most appropriate way with sufficient oversight and thought into them and that confidence that there will be somebody to rein in the inappropriate use is absolutely, there's a question uh back there in the back Yes. Mark Anderson Ceo of pattern computer. First of all, I want to say how cool, I think is what you guys are doing. You know, we we are involved in the same world in a tiny way, but google is a bright, shining light and I'm really impressed and I appreciate all you do. So that's a heartfelt thing. I will mention that we now have demonstrated explain ability. So, um I'd love to talk to you about that. We've actually done it. Berkeley Labs has kind of certified it for us. So I think we're good and it makes a huge difference. The I was inspired by the title which I assume you chose for this conversation and it had something to with the world's largest problems and we worked on that pattern. Um we're creating five new novel drugs for treating the top five cancers every month. So um we care a lot about this kind of stuff. There's a question I have a really personal question, I have to say that first. So when you is a little bit like the last question, but I hope more piercing when you personally look for what to work on, I think that there's this world of Ai discoveries can be broken into incremental and Pat Collison thinks it's too incremental or major. We call them pattern discoveries. Do you see that difference and how do you navigate that? Well, I do see the difference that I appreciate the question. I just think that the the kinds of questions in science and biology and in health that you're describing related drug discovery are such huge important issues for society. So think about those, think also about what's happening with climate change and sustainability. One of the things that didn't explain for example is it turns out that in fact you can do a whole range of things in climate sustainability and and the people affected by those, whether it's around flood forecasting and those kinds of solutions that are incredibly useful and important. So I think the big questions are really, really, really important and I think there's an interesting intersection here between businesses like google and others pursuing those opportunities, response and making sure everybody benefits from those solutions, but at the same time being responsible. You know, often this question of AI and responsibilities kind of framed as a either or in my mind, being responsible about AI involves both, capitalizing on all the amazing things that it can do for people and for society while being responsible. It's gotta be both. We have to do both. That's fascinating. Thank you James so much. Forget helping. Uh we didn't get to the business to the business application issues, but we will throughout the conference, you've really given us a great framework to think about what's happening in AI. Thanks for taking the time. Great to see you again.

If you're lucky enough to live in these cities, you can hail a driverless taxi today.

Driverless vehicles are here, but growth is slow

Tech

While you can hail a driverless taxi in a few cities, it will be some time before the vehicles are truly commonplace.

just a year ago, I don't think a lot of people were just expected to see vehicles with no drivers behind the steering wheel and it's happening now Cruz is building an old electric autonomous fleet. We have about 200 cars registered with the California D. M. V. And we use them across two fleets. Um one which is our fully driverless fleet, no driver behind the wheel, not even a safety driver, It operates our public service between 10 PM to six AM across 70% of san Francisco. And then the second fleet is our testing fleet which runs all hours of the day and the full seven by seven of san Francisco Momentum is the leading autonomous driving technology company, committed to create a better future for smart mobility through our race to Ai technology recently with the kind of projects for Texas with our partners, the ability to deploy 60 active in Shanghai, Shanghai is one of the largest city in china. There is huge potential to sell automated vehicles in the city and also the people to customers in shanghai. They are very living to accept new technology when no one is the program we have where anybody can just download the app, sign up, take a ride. We've done thousands of rides that way, tens of thousands of rides and we've been focusing a lot on kind of the phoenix metro area in san Francisco and now we've taken that exact same software, the exact same cars and we're trying to deploy it in downtown phoenix, it's going great and it's a really good sign for how things are going for us that we're able to take what was built in one city and use it in another city without a whole lot of additional work, 80% of the writers, they use our service twice or more after the initial track. So we see some thickness of that services. We also try to include some of the cool technologies into the rock taxes for example, they can already choose music, they want to play in the way of after they started the tax on top of that for our company. We also use that as a very good communication to tell people what the functionality was, the benefit of autonomous driving. So that in the future when we began to deploy our solutions to many of the mass produce the vehicles. They are the first group of people who is going to buy our products. You know, we can see that Waymo pace here is really accelerating, but that doesn't mean that there's gonna be a date where, you know, every city across the world has autonomous vehicles. It's not something where we just push it to the app store and other everywhere. We need to get cars everywhere. We need to, you know, develop the match for the city. So it will be a little more gradual than what people might expect other technology, but it's really picking up the pace, the origin is a game changer, it's a vehicle built from the ground up for autonomous driving. We do not assume that there's going to be a driver if you look at it, the sort of internal cabin is just really huge and very comfortable. Um, and that's what's very exciting about the origin and that by leveraging autonomous technology, we can actually maximize the experience that passenger can get from a vehicle. But I I believe, as I mentioned, this is a huge game changer. And the transportation industry, our vehicles are very good in driving comfortably because they are really good in estimating sort of physics and making sure that we break comfortably. We turn in a comfortable manner and at the same time they're also always looking out for risk for inclusion For Pedestrians. And so we are achieving amazing levels. There's lots of evidence that humans are not perfect driving right. There are some more around 40,000 deaths in the United States every year related to automobiles, drivers can be texting on their phone, they could just be distracted, our cars are not distracted, They are paying attention. They have a 360° view of what's going on and they are attentive because that's the only thing that programs to do is to watch out and drive safely Every year, there's more than 1.3 million people died in the accident and 90% of the accidents are likely to be caused by human error. Therefore we believe our Australian technologies can help us to increase the safety and ultimately save more lives. Like any new technology, there is a period where people need to become more familiar with it. However, I find it hard that anyone would not want to leverage sort of autonomous driving at some point in their life. So I'm actually extremely excited about what's going to come up and in the next few months and beyond. So I think the future ahead of us is gonna look very differently than what we're used to and it's happening now. So I'm super excited.