Uber’s $1 trillion plan for the post-driver world

Updated Mar 25, 2026, 11:53pm EDT
Business
PostEmailWhatsapp

In this article:

Title icon

The News

Uber’s next trillion-dollar opportunity: the care and feeding of robotaxis.

CEO Dara Khosrowshahi sees a future of fully autonomous cars that need to be serviced, charged, repositioned, kitted out, insured, financed, and washed. Even if the job of driving cars goes away, “these are very, very sophisticated machines that need lots of tender, loving care,” he said on the inaugural episode of Semafor’s Compound Interest podcast. “All of the jobs other than their driving have to be done, even more so, in an autonomous world.”

Khosrowshahi envisions a world in which Uber is the wrap-around servicer for self-driving fleets “owned by big financial institutions, the Blackstones of the world” in major markets, plus smaller one- or two-garage entrepreneurs in the suburbs. “We think this is another trillion-dollar-plus opportunity,” he said, and asked: How do we make it safer, cheaper, and available to more people?

His challenge is to turn Uber into a driverless world’s “mission control” without disappearing into the background, becoming a back-end B2B platform rather than the $30 billion verb of a brand name. The danger, he acknowledged, is customers deleting their Uber apps and instead telling their agents, “Get me a car.” Guarding Uber’s place as the consumer “gateway” will require it to do battle with AI agents and bake in customer loyalty, leaning hard on its Uber One membership.

It will also require a blend of competition and collaboration with companies like Waymo, which partnered with Uber to launch robotaxis in Atlanta and Austin. Waymo has more recently gone at it alone in Dallas and Orlando instead of partnering with Uber, a shift that has unnerved investors and contributed to a 25% slide in Uber shares since October. A similar friction may play out on the food delivery side of its business, as restaurants like Chipotle and McDonald’s look to steer customers to their own apps.

“This is not going to be a black or white ecosystem,” Khosrowshahi said. “Whether it’s a Waymo or a WeRide or a Baidu or a Pony.AI or a Wayve, they will have a mix of direct channels and indirect channels.”

The future is wide open enough for everyone to have a piece. And the savings from replacing human drivers can probably be shared generously enough to keep partners happy: Khosrowshahi said he’s comfortable keeping Uber’s 20% take of bookings where it is and that the “vast majority of the economics” will go to the rest of the ecosystem. His focus on growing the pie rather than increasing Uber’s slice is a go-along-to-get-along strategy. So is using Uber’s capital to help partners grow; the company has invested in Pony.AI, WeRide, and freight-focused Waabi, among others.

The question is whether Khosrowshahi has bigger plans for Uber’s capital. A protégé of M&A machine Barry Diller, Khosrowshahi begged off when we asked: “I think the best M&A strategy is not to have to buy anything.” But he seems committed to Uber as a platform for flexible work, even once the driving is being done by computers. Could he buy Instacart or Fiverr and roll up the gig-work economy? He clearly has ambitions in travel (if not into space) and is still on the board of Expedia, where he was CEO before coming to Uber in 2017.

In this article:

Title icon

Transcript

Liz Hoffman:
Welcome to Compound Interest from Semafor Business. I’m Liz Hoffman, Semafor’s business and finance editor, and with me is Rohan Goswami. Hey, Rohan.

Rohan Goswami:
Hey, Liz. How you doing?

Liz Hoffman:
I’m great. This is our first podcast. We’ve been covering the forces behind business in our twice a week newsletter, on the web. We call people and talk to them all the time.

Rohan Goswami:
Harass them.

Liz Hoffman:
Yeah, harass. It’s the gib. I’m excited to get this podcast going because there’s just certain conversations that you want to spend a little more time with or tilt the lens on a little bit. So this podcast is going to be a great place to have these important conversations about the ways that business is changing and these sort of really radical but often kind of unseen ways being pushed by forces on Wall Street, being pushed of course by tech, of course by Washington. But to try to take a look at those businesses through the people running them. So I’m really excited to get going.

Rohan Goswami:
Totally. Who do we have this week?

Liz Hoffman:
We’re talking to Dara Khosrowshahi. He’s the CEO of Uber. And for most of its existence, Uber was the way you called a cab. And now it is maybe more than any company that kind of hit it big out of Silicon Valley’s app era, trying to figure out what it is in Silicon Valley’s AI era. Is it the way that I summon my Waymo? Is it the way that I will kind of rent out my own car? Rohan, do you have a car?

Rohan Goswami:
I am a car owner, not in New York, but I am a car owner.

Liz Hoffman:
Okay. Well, I finally bit the bullet a year ago and I have a car which spends most of the time on the street. Is it the way that I rent that out when I sleep if it turns into an autonomous vehicle? Or is it like the kind of operating system, the kind of mission control if you believe the big ideas people out there, it’s just going to be everyone is going to have an autonomous car and the sort of human element of driving gets replaced by all the software and services and insurance. And really what Uber is, is sort of a demand aggregator. So kind of like what does that look like here?

Rohan Goswami:
And it’s funny, this comes as there’s a real fight among all these operators to be the app that you open for a ride, order food or whatever. At the same time that AI agents might come along and just abstract this whole thing into some bot boiler room.

Liz Hoffman:
Yeah. And then there’s a question of what happens if human drivers are no longer needed. Certainly Uber has had a fraught relationship with its drivers. Most of that was before Dara got there. Travis Kalanick was not known as a people person, I don’t think. And it was all a little bit dilutive to the brand and that was some cleanup work that Dara did on the way in.

Rohan Goswami:
A little bit.

Liz Hoffman:
Yeah, but if it does happen, remember Uber is a marketplace that connects people with cars and people who want rides. If half of that, sort of, supply goes away or changes, and it’s already changing, theoretically every car can be summoned anytime and is making deliveries at 3:00 A.M. to bodegas.

Rohan Goswami:
Yeah. I don’t know that anyone wants to drive around my 2006 Volvo Station wagon, but I will say this is a vision, and give Elon credit for this, that Elon laid out, even if he hasn’t really yet executed on it, what, 15 years ago?

Liz Hoffman:
He has been on a rolling two-year clock of everything being autonomous and has not delivered. But no, I think Uber’s right in the middle of that, and this is a business that was itself incredibly disruptive to the taxi industry.

Rohan Goswami:
Sure.

Liz Hoffman:
And is now kind of having to remake itself. I’m excited to talk about all that with Dara who spent his career at Barry Diller’s Media Empire, then Expedia.

Rohan Goswami:
Dara is probably at his heart just a media guy like all of us.

Liz Hoffman:
That’s why he was such a great guest for the pod.

Rohan Goswami:
Precisely. Our inaugural guest. Let’s take a short break and when we come back, we’re going to pick it up with Dara.

Dara Khosrowshahi:
Thank you for having me.

Liz Hoffman:
Thanks for coming. Just to start, Uber, for most of its life, was a way you called a cab basically, and it became a way to get food delivered. What is it now and where is it headed?

Dara Khosrowshahi:
Well, to some extent for the consumer, what we want to be is just your operating system for daily life in a city or the suburbs, et cetera. We want to give you your time back. And part of the reason where people earn their time back on Uber is you don’t have to do the driving, you could do some work while you’re being driven from point A to B. You can get great food, groceries, makeup at Sephora, same day delivered. So we are about increasing the convenience of your life and bringing your time back to the consumer. We do it by building this essentially next generation, on demand, logistics system. There’s never been a logistics system in the world, the size and the scale and the ubiquity of Uber, but the goal of that system for consumers is to help you live your day more conveniently and more easily and give you your time back.
And then of course, we’ve got nine and a half million drivers and couriers for whom Uber is a platform for flexible work where they can work when they want, where they want, on their own terms, so to speak.

Liz Hoffman:
And how long does that last? What’s it like to run a company where you know that eventually most of those humans are going to go away?

Dara Khosrowshahi:
Well, eventually it’s going to take a long time and hopefully it lasts for a long time. But at the same time, like you’ve said, technology is going to change and has changed historically the nature of work. And I think with AI and physical AI and autonomous, whether it’s a desk job, like ours, or whether it’s a job delivering or working in a warehouse, the nature of work is going to change. And I think that people talk about the drama of machines replacing humans, but more and more what you’ve seen historically is that machines complement humans and they take, it, call the easier jobs and the more rote part of the job’s over. And then humans adjust what they do to be more value add, et cetera. So you could imagine a world where, if you look at a big chip manufacturing plant, there are people around, they’re not doing the work of, let’s say, manufacturing the chips, but they’re making sure that the plant is working, they’re overseeing the systems, et cetera, and the processes.
And you could imagine a world where you’ve got autonomous vehicles everywhere, but taking care of the vehicles, recharging them, making sure that the system is working as designed, so to speak, becomes the work of the people and the driving and the delivering more and more becomes the job of the machines.

Liz Hoffman:
You launched Uber Autonomous Solutions a couple days ago, which I think you’ve described as kind of mission control for AV, for autonomous vehicles. What does that actually mean on the road and what does Uber’s business look like going forward, what’s in 10 years?

Dara Khosrowshahi:
Well, if you think about the Uber driver today, sure, the part of their job that you see is they’re driving the vehicle back and forth, et cetera, but there’s so much more that they do. First, they’ve got to buy the vehicle, take care of the vehicle, recharge it. They are deciding where to position the vehicle, et cetera, repair. They take care of insurance. We work with them on the commercial side as it relates to insurance as well. And all of the other jobs, other than their driving, have to be done even more so in an autonomous world. The fleet management, again, the repair, these are very, very sophisticated vehicles and machines that need lots of tender loving care. We are building that ecosystem at scale to eventually be able to operate in the 70 plus countries at the scale of the 40 million trips a day that we do today, building that whole ecosystem, a lot of which is taken care of by individual drivers, that is essentially what Uber Autonomous Solutions represents.
These are technologies that companies have been working on for 20 plus years. Billions of dollars of investments have gone into this industry. Uber Autonomous Solutions is about speeding time to market with the data that we have, et cetera, and then allowing these businesses to scale with fleet management, software, et cetera, so that you can come to market and you can be profitable, faster than any other alternatives if you are on the Uber network.

Liz Hoffman:
I finally caved last year and bought my first car in 15 years. And somewhere between the 2006 Honda Civic I owned and the new one, they turned into computers and there’s no autonomous part of this. This is just a car and everything is a computer.

Dara Khosrowshahi:
Well, our job is going to be to convince you to get rid of the car in five years.

Liz Hoffman:
I live in Brooklyn. I would love nothing more than to spend that 60 minutes that I spend every week moving my car for street sweeping. But is that an acknowledgement that Uber might not, in a truly AV and AI world, BV, consumer-facing consumer brand that it is today, but that it needs to plug into partners and find more ways to make money even when it isn’t the one delivering the car?

Dara Khosrowshahi:
Well, listen, I think that to some extent is true of us today in that we’re not the ones doing the driving. We depend on an ecosystem of nine and a half million drivers and couriers to help power the ultimate service that people recognize as Uber as a verb even. So I don’t think that is going to change. The umbrella brand will still be Uber. I see where the world is going more like if you think about Uber Eats, right? When you order from Uber Eats, you are ordering from Uber Eats, but often the brand that you’re looking for is a Chipotle or maybe that local restaurant that you love and it’s Uber Eats that is helping deliver that brand and deliver that service to you. And the two live in conjunction, right? They are different brands. And listen, Chipotle loves their direct customers as well.
They want you to come to Chipotle.com, but Uber Eats is another way for them to capture consumers, to get their brand out there. And in autonomous worlds, yes, of course we will be working with partners like a Waymo or a WeRide or an Avride, and they will have their own presence and they will have their own brand, but I think Uber and the AV ecosystem can really work together to create a win-win.

Rohan Goswami:
You guys have really managed to do, to your point about turning yourself into a verb. That’s kind of the holy grail. Only a handful have really ever done that. Google, Venmo, Photoshop, arguably.

Liz Hoffman:
I’m old. Photoshop means something to me. Xerox did it, right?

Rohan Goswami:
I would not agree with that assessment of you being old, Liz, but fine. The point being, there is a not inconsiderable amount of brand equity, goodwill on your balance sheet. If the future is a little more anonymous and it looks a little bit more like an operating system, the backend of moving people, things, food, assets around.

Dara Khosrowshahi:
Well, I think we will be more the front end, which is we will have, just like we’ve got drivers today, 10 years from now, might be a robot driver. We will be a gateway to essentially mobility of all kinds, delivery of all kinds. And we have historically been a supply-led company. You associate us with a consumer brand, but the consumer brand has grown because we have essentially wired up all of the mobility and now delivery supply in the world. And as we wire up that supply, the service becomes faster, it becomes cheaper, it becomes more ubiquitous, it becomes available not just in the center of New York City, but in a Brooklyn and a Westchester County, et cetera. And as the service becomes faster, cheaper, more reliable, more ubiquitous, our consumer brand expands it as well. And I think what autonomous promises is safer streets, but an enormous expansion in terms of the supply of safe mobility to everyone, it is going to bring prices down eventually.
Early on, it’s very expensive, and it will make our logistics system more ubiquitous around the world, which I think is only going to grow our brand. Anytime there’s technology disruption, and AI is definitely a disruptor, companies have to adjust. Do we have to adjust? Absolutely. But the potential for continued expansion and growth of this business is enormous and it’s our job to take that potential.

Liz Hoffman:
When you think about the wages right now that go to the driver, do you have a sense, and pick your timeframe, what that savings is and then how it gets split between a cheaper ride for me, a higher margin for Uber, shared with, obviously, partners like Waymo and others. How does that break down?

Dara Khosrowshahi:
Honestly, it remains to be seen. If you look at our services overall globally, we do over $200 billion of bookings on a run rate basis, and the platform generally keeps about 20% of that take. 80% goes to the ecosystem, whether that ecosystem is the driver, courier, restaurant, et cetera. My view is that that 80%, if you look at the economics of AVs at maturity or drone delivery, et cetera, that 80% is where the vast majority of economics go. We think that as a platform, kind of a 20% take rate compared to other platforms is a good, healthy number. And for us, the goal of autonomous is how do we expand the market and we think this is another trillion dollar plus opportunity. How do we make it safer? How do we make it cheaper? How do we make it available to more people?

Liz Hoffman:
I think you’re talking about multichannel ecosystem and that Chipotle has probably 10 different ways I could get Chipotle, some direct, some through lots of partners. You launched a partnership with Waymo in a couple places. Waymo has expanded into some new places without you as a partner. What’s your sense of the experimentation they’re doing? Does that unnerve you?

Dara Khosrowshahi:
The ecosystem as it’s building out, it’s not a surprise at all to see a Waymo expand using their own channels and work with us in certain markets as well. And what we’re seeing is that the hypothesis behind Uber’s value as a platform is that you can drive much higher utilization of your fixed asset if you use our platform. And the evidence that we’re seeing in terms of Austin, Atlanta, et cetera, is the number of trips per vehicle per day that we are able to drive through our platform is very, very high and can earn that 20% take rate and more that I talked about. So I do think that this is not going to be kind of a black or white ecosystem, whether it’s a Waymo or it’s a WeRide or a Baidu or a pony.ai or a Wavye, they will have a mix of direct channels and indirect channels.
But what we’re seeing is that the value that Uber brings, both in terms of demand and then all of this ecosystem that we’re building around fleet management, repair, even helping partners with the in-car experience, that the value there is very significant. And this is going to be an opportunity for the Waymos of the world, for the Ubers of the world, for the We Rides of the world, this is a market that’s going to get much, much larger.

Liz Hoffman:
Yeah. How long before I go into Robotaxi and it picks up the podcast that I was listening to at home?

Dara Khosrowshahi:
That’s going to happen much sooner. That’s actually a relatively easy technical challenge compared to the technical challenge of driving safely in all environments, weather, et cetera. You’ll get that. That’s an easy one.

Liz Hoffman:
While you still have the people, Uber launched a program last year that pays drivers to do AI data tagging, sort of blocking and tackling for third party AI companies while they’re waiting to get rides. And it’s a little grim, but I understand why it makes sense for you and perhaps for them. How big a business is that and what could it be?

Dara Khosrowshahi:
I think it’s great. A way of looking at Uber is, we are a platform for flexible work opportunities. And the first work opportunity that we went through, went after, it was driving, then it’s delivering, then it’s shopping, and this is just another work opportunity that we are finding our flexible workforce. And there are some people who love doing this, whether it’s in between rides or when they go home, et cetera. And some of the roles are becoming much more specialized, finding a PhD who speaks French and understands physics as well. So it’s really, it represents our expanding the scope of our, call it flexible labor ecosystem, and it happens to provide more people more work opportunities, and that is absolutely one of the missions that we have to expand the kind of work available to the general population.

Liz Hoffman:
And how big a business is that for you today?

Dara Khosrowshahi:
It’s growing. It’s one of our fastest growing businesses. It’ll hit the hundreds of millions we think this year. And then beyond that, we think this can be another billion-dollar business for us. The role of whether it’s training the agents or we even have drivers driving around, measuring cell phone signals, et cetera. There’s a lot going on, going to shopping shelves and taking a real-time inventory on supermarket shelves. The kind of work is quite expansive, and I don’t see that ending anytime over the next 10 years. After that, who knows?

Rohan Goswami:
I mean, one thing that is funny to me, and Liz and I were talking about this earlier, is that it costs less on Uber to send a package across town, from Brooklyn to Manhattan or what have you, than it does a person.

Dara Khosrowshahi:
One of the reasons is that packages are not sensitive to how quickly they get there. So what you may not be seeing behind the scenes is the packages get to drive together. So usually we’ll take one package from you. Liz may have another package. We batch that together. That’s one of the benefits of the network that we have and the enormous amounts of supply and demand that we have together so that one driver can drive four or five packages and the packages are much more patient than human beings, go figure. They deliver the packages at a lower cost as well. That’s what’s going on. It’s really a bunch of AI that is driving batching technology to drive the cost of each package delivery lower.

Liz Hoffman:
It’s not that the driver charges more to interact with a passenger. There’s not some discount for human interaction based into that that should unnerve us all.

Dara Khosrowshahi:
Listen, there’s some drivers who actually like driving inanimate objects more than human beings, and that’s one of the powers of Uber. I still remember after COVID, there was a driver who I was talking to who had switched over from driving people to delivering food, and you can make more money driving people, generally. There’s a higher premium there. And I asked her like, “Why don’t you just go back to driving people?” And she said, “Well, Big Mac isn’t going to give me COVID.” So every driver, they can decide what kind of work that they want to take on. And that’s one of the beauties of our platform in terms of the flexibility that we offer.

Liz Hoffman:
I remember Uber came in for some criticism at some point that you could put your driver on quiet mode or something and that understandably ruffled some feathers, but I always worry about it on the other side, which is that I am a chatty passenger. I like to talk to people. And I always wonder if there should be a way to match with someone who that does not annoy.

Dara Khosrowshahi:
I’m glad that you’re concerned. Some passengers kind of don’t care. I remember when I drove, I was always nervous as well. There’s this feeling out your passenger, do you want to talk? How was your day? And you see if they’re talkative or not. But listen, we’re all human beings and we figure it out one way or the other.

Rohan Goswami:
Do you still have time to fire up the driver app and do a few laps around and pick people up or no?

Dara Khosrowshahi:
I make time for it, but I don’t have nearly as much time as I’d like to.

Liz Hoffman:
Got to stay close to the product, Dara.

Dara Khosrowshahi:
Yeah.

Liz Hoffman:
Let me ask, if every car is autonomous in the future, to your point that there’s this fixed appreciating asset that most of the time isn’t doing anything, is there a world in which my car is out delivering to bodegas at three o’clock in the morning and I’m making some money on that? And by the way, Uber is the wraparound service by which that happens. And what does that world look like?

Dara Khosrowshahi:
First of all, I think it’s quite likely that 10 years from now, every new car sold will have autonomous capabilities, L4 autonomous capabilities. I think that that is very much a likely outcome. It’ll be very good for Uber. I do think that the cars are going to be cheap enough where someone like you, you’re probably not going to want strangers in your car. And the amount that you can make for your vehicle driving around at night is just not going to be that much. So the ecosystem that we see is that, in the big cities, for example, there will be very large fleets that are going to be owned by big financial institutions, the Blackstones of the world, et cetera.

Rohan Goswami:
Good, good, good.

Liz Hoffman:
We’re going to find out if Elizabeth Warren listens to this podcast.

Dara Khosrowshahi:
They will own and finance the fleets. We will kind of build that whole operating infrastructure, the fleet management tools, the recharging, et cetera. And so in large markets, you will have institutional supply. And then on the outskirts of those markets, in a Westchester County, Terrytown, New York or that area, you’re going to have smaller entrepreneurs. There may be two people with a garage that bought 20 cars that are running a smaller fleet management operation, and it’s going to be quite entrepreneurial. Hopefully a lot of those people will be our drivers. And then I do think there will be a minority of individuals who kind of send their cars out. It will certainly be something that you can do, but I imagine it to be a little bit like a Airbnb where you’ve got these super hosts and then a lot of Airbnb’s volume sure comes from individuals, but there are these superhosts as well.
You will have superhosts, in the big cities, there’ll be institutions. And then in the outskirts, in the suburbs, you’ll have smaller fleet managers who are entrepreneurs, who are building business for themselves.

Rohan Goswami:
You’re a finance guy, not an engineer by training, but how do you imagine cars look like in that world 10 years from now where everything is autonomous? What do they actually feel like to be in them?

Dara Khosrowshahi:
There’s a tagline. It’s a ultimate driving machine. I think you’ll move from an ultimate driving machine to an ultimate passenger machine, so to speak. You won’t have gas pedals, you won’t have steering wheels, you’ll have a lot more room in the back. You’ll have specialized vehicles for different missions, so to speak. I think that there will be a lot of two-seater vehicles because the majority of rideshare actually is either individual or two passengers, and then you’ll have vehicles. I do think that the bus networks, et cetera, of cities are going to become autonomous. And so you’ll have a population of vehicles that will be tuned for the individual, and then you’ll have populations of vehicles that will be much, much cheaper and built for mass transit. And I think it’s going to make for a much, much better logistics ecosystem in the cities.

Liz Hoffman:
It’s interesting. One thing that tickles me, I was talking to someone who makes these cars, and I was like, “Why does the Waymo have windshield wipers?” And he was like, “Well, for right now, it’s because we don’t want to have a different assembly line that doesn’t put on windshield wipers.” But he said, “But there’s actually people who get nervous if the car that they’re in doesn’t have windshield wipers, even if it doesn’t need it, that there’s obviously still an emotional connection to having a steering wheel where you expect it to be.” And, I don’t know, I wonder how quickly you think that part will be to change.

Dara Khosrowshahi:
Well, I do think that one of the work areas that we’re thinking about is that internal environment and whether or not people will kind of want their phones and that’s an ultimate kind of personalized ecosystem that you go through, or there will be pretty big AV systems, et cetera, so that you can move your Spotify podcast from your phone to a high fidelity AV ecosystem as well. So there’s going to be a lot of customization going on, and I think windshield wipers probably are going to see their last days, is my guess.

Liz Hoffman:
Your stock was among those hit this week with this kind of little known research firm put out this very cinematic dystopian view of not that far in the future where AI first pushes everyone into the gig economy, which is probably good for you in the short term. And then consumer demand just goes to zero and no one has the jobs to buy all the stuff that AI is producing. The caveat that nobody knows anything right now, does anything about that scenario read true to you?

Dara Khosrowshahi:
I think the scenario was just a bunch of bad stuff happening without any of the positives, and I think it’s just highly unlikely. Of course, there will be negative externalities associated with this enormous change coming up. The nature of work is going to change. You as an individual have to become very knowledgeable about using AI tools, and if you don’t, the individual who is, is going to outcompete you. So I do think that there’s this drama of the black and white, which is AI replacing people. AI is going to augment people. Machines have always augmented people over a very long period of time, and they make those people more capable. And the question is, who is going to make that transition to the AI native worker? And I think it’s AI native workers who are going to win, and I don’t think it’s the AIs themselves that are going to win.
I think humans and AIs together are the better solution. That’s certainly what we’re trying to build, and I think that’s a future that certainly I want to believe in.

Liz Hoffman:
You’re making this millennial very nervous, Dara.

Dara Khosrowshahi:
Do you walk out of your house without a phone?

Liz Hoffman:
No.

Dara Khosrowshahi:
Right. We are augmented human beings. We’re augmented by this thing.

Rohan Goswami:
True.

Dara Khosrowshahi:
And whether we know it or not, it’s already happening. There’s a machine in each of our pockets. That machine is making us more productive, more communicative. There’s good things about it and bad things about it. If you’re having dinner with your family and picking this thing up, that’s bad. But the majority of it is good. So this is, we’re just moving up on that augmentation path. This journey has been happening for a very, very long time. It’s usually slow and you don’t notice it, but I know very, very few people who walk out of their homes to work without one of these things. So it’s happening to us whether you like it or not.

Rohan Goswami:
In my youthful naivete, I prefer your much sunnier and optimistic forecast to Satrinis. But still, since September, you guys hit all time highs back then and your stock’s been down a little bit. What are these investors missing?

Dara Khosrowshahi:
The public markets don’t like uncertainty. And the fact is that we are going through a period of uncertainty. There’s a lot of risk as it relates to AI and the changes and the competitive effects that it’s going to have on lots of companies, software companies, physical world companies like ours. So I do think that we have an enormous opportunity ahead of us. It is a trillion dollar plus market. I do think that if we get it right, transportation is going to become ubiquitous, and we have the potential of actually growing at an accelerated pace, but we have to get it right. So it is that moment of the market picking winners and losers. Risk is definitely increased, but ever since I’ve been at Uber, we are a company that has had lots of question marks. We’ve executed through incredible periods of crisis and come out better for it, and then some.
This is one of those moments, and I think if we put our heads down and execute and innovate and build as fast as we’re building and build with this AV ecosystem, I think we’ll be winners and whoever makes a bet on us during this period of uncertainty is going to be really, really happy. It’s my job to make that happen, and I definitely intend to.

Rohan Goswami:
There were also some rumors that you were actually interested in picking Expedia back up. If you want to bite on that, was that true at all? Was there anything? But also more broadly, it seems like you guys are covering transportation. You are in 70 plus countries. Do you think about whether travel bookings more broadly tapping into how people move around? Is that part of Uber’s future five, 10 years from now?

Dara Khosrowshahi:
I’m still on the Expedia board. Love the company, love that team, and they’re doing great. Travel is a very big use case for us, right? Because we are ubiquitous in so many countries, the international traveler especially uses us, both for mobility and delivery as well. If you get to your hotel and room service ain’t that great, Uber Eats is going to be there for you, we are wiring up different travel opportunities. For example, in the UK, we’ve got cars, we’ve got buses, but we also have trains and we even have Uber boats moving up and down the 10s. So you can imagine us wiring up more things that move. And with travel being such a large part of our ecosystem, it is something that we are going to build. There will be more travel options on Uber, but we’re doing it through partnerships and stay tuned as it relates to who those partners are going to be.

Rohan Goswami:
I mean, is there a world where I’m sitting in my apartment and I decide I want to go to London on a whim and I open the Uber app and say, “Take me to House of Parliament.” And Uber will pick me up in an autonomous car, plop me on a Delta flight and send me across the pond. Is that a world that you envision if you’re talking about being an operating system for mobility?

Dara Khosrowshahi:
If you get bored of your job and want to work in our strategy team, you’re welcome. Absolutely.

Rohan Goswami:
This job is too fun. I get to talk to people like you.

Dara Khosrowshahi:
Yeah, exactly. And listen, the power of AI, too, it’s actually, it has historically been difficult to build out here is the code that can respond to the real world because lots of things happen in the real world. There are delays in every single one of those handoffs, but with AI and these larger models, what you described, that vision of kind of door-to-door travel through different modes of travel and having a player like ours who has access to all of the inventory and all of the prices and can stitch together that end-to-end journey for you, that is absolutely going to come true. It’ll be through a combination, for us, of partnerships and then direct relationships with the ecosystem, but it’s definitely a vision that we think about and it’s a vision we love to bring into the world.

Liz Hoffman:
How do you think about M&A though? You came up in Barry Diller’s Empire, which has been put together and remantled a couple of times. How do you think about it?

Dara Khosrowshahi:
I think the best M&A strategy is not to have to buy anything.

Rohan Goswami:
Said the former CFO of IAC?

Dara Khosrowshahi:
Listen, it is. Deals are hard. The majority of deals fail. So my job, number one, is to build out a path for organic growth, and we’ve grown over 20% for, I think, four or five years now, that continues for the next three to five years where I can actively plan for. 10 years from now is much harder to plan for. My job is to set that reality so that any M&A that we do is essentially optional and on top of that, we tend to look at M&A in areas that are adjacent to us or M&A doing the same thing that we do, expanding into different countries. For example, buying one of the leading food delivery players in Turkey, which has been an enormous success. So M&A will always be a complimentary strategy, but my mainline job is to make sure I never need to buy anything, and the business keeps growing at very, very healthy rates.

Rohan Goswami:
Tech deals also look very different today than they did 20 years ago. Now people are aquahiring, and taking stakes. I mean, is that something, is there a world where you would consider, maybe no specific company names, taking a 10% stake at a very favorable terms in a partner of yours? Is that how you think about M&A at all?

Liz Hoffman:
We’re fresh off the Meta AMD deal yesterday. We got warrants coming out of our ears, Dara. Sorry.

Dara Khosrowshahi:
Yeah. That seemed like a complex deal. No, but we are, for example, investing in our AV partners. We have investments in Pony, WeRide, Waabi. Many of these partners, we are helping with our capital and helping with our data capabilities, data collection capabilities, et cetera. There definitely are investments that we’re making in the ecosystem, and that will be a significant part of our activity going forward.

Liz Hoffman:
One other sort of big business model pivot that I see everywhere and I’m super interested in is that every company wants to be a membership club. You see it in subscriptions, you see it. And of course you have Uber One, but these sort of ecosystems that monetize loyalty. I’m curious, why is that better than just me giving you money and you giving me a ride? And how do you think about building that community, I guess?

Dara Khosrowshahi:
Well, Uber One has been an enormous success. It’s one of the fastest growing parts of our business. We’ve got over 46 million members now, growing well over 50% on a year-on-year basis, and close to 50% of the overall spend on Uber now comes from members. And the way I look at membership to some extent is we compete against players who are rides-only players or delivery-only players. We have both, and our membership typically is priced at the same level. I think we look like Netflix, which is, you pay a certain amount and we just have more content than anyone else. And the player who has more content is a player who over a long, long period of time wins. And winning means getting you to use my services on a frequent basis, hopefully for the rest of your life. And membership, it is a trade for a consumer, which is, I’m going to pay some money upfront.
It kind of creates stickiness for you in that ecosystem, you’ve already paid. But what we see out of membership is that the use of our services, both in terms of the amount you spend and over a long period of time, in my old job in the banking world, there was, I think Sumner Redstone said content is king. And we’ve got the content.

Liz Hoffman:
And we talked about this a little bit earlier, but I want to revisit it, which is that does that loyalty become more important in an agentic world where, to the extent that I really want to fly United over Delta, I’m going to make sure that I book that directly or just become less valuable because it is all happening behind some curtain where I’m not actually making the call.

Dara Khosrowshahi:
I think my instinct is it becomes more important. The danger of an agentic world for players like ours or any other retailer or platform company is the agent gets in front of our platform. You call OpenAI, ChatGPT, “Get me an Uber.” And that removal, one level removal of our brand could pose a threat. But if to the extent that you are an UberOne member and you get a discount on that, and we know where you’re going, if you say, “Get me an Uber to work,” we know exactly what that means. I think a combination of ubiquity, personalization, and membership will solidify your place so that when you use that open AI agent, you don’t say, “Get me a car to work,” but you say, “Get me an Uber to work.” And I think membership plays an increasingly important part of that.

Liz Hoffman:
And get me a chatty driver except when I’ve just come off a red eye, know all of those things, right?

Dara Khosrowshahi:
You won’t need to tell us that. We’ll hopefully know that.

Liz Hoffman:
Amazing. I wanted to ask, you know just huge questions about AI over the travel industry, which is sort of the archetypal job that was put out of business, the travel agent, by the rise of the internet. How are you thinking about all those booking platforms and where they should be spending their time?

Dara Khosrowshahi:
By the way, you might be surprised to know, but there are plenty of travel agents around, right?

Rohan Goswami:
The high end.

Dara Khosrowshahi:
Yeah, high end. Exactly. So it’s the agents, the Expedia does the simple stuff, the less value add things, and then the agent moves on to higher end categories. It’s exactly that road of travel. What we see, first of all, is that AI is enormously powerful for the planning part of travel. I’m looking to go to Europe, maybe Southern Italy, build me an itinerary, et cetera. That AI content is quite rich and quite differentiated and can be quite valuable to the extent that it’s sourcing consumers. The booking part has been more complicated. I think the good news for an Expedia is that they actually have a very, very strong business in terms of third party bookings. They power a lot of other sites. It’s a private label business that is the fastest growing part of that business. And I think there is a world where actually that business accelerates and powers all of the AI agents of the world, which would really put Expedia in good stead.
But there’s a lot of change coming and there’s a lot of execution that they have to get right, but I actually think they’re quite well positioned.

Rohan Goswami:
As much as I hate to cut it short, we’ve got to take a quick break. We’ll be right back with Dara.

A SPONSOR MESSAGE

The View From

Rachel Oppenheim:
Welcome to What’s Working, the show within a show featuring Compound Interest’s season partner, Amazon Business. I’m your host, Semafor Chief Revenue Officer Rachel Oppenheim, joined by Amazon Business VP of Technology, Doug Gray.

Welcome Doug.

So a question for you. I know the answer to this because Semafor is actually an Amazon Business customer, but for those listening, tell us the difference between Amazon and Amazon Business.

Doug Gray:
That’s a great question. More than 10 years ago, we took what people love about Amazon, the millions of products, the low prices, and the convenient delivery options, and we created Amazon Business for organizations that are looking for the same simplicity in their business buying. And the difference is that businesses need more than convenience.

They need a business-relevant selection. Bulk buying capabilities, the flexibility to add multiple buyers, budgeting and purchasing controls, approval workflows, invoicing options, and visibility into spend across all of their teams. Amazon Business gives organizations of all sizes, from small businesses to Fortune 100 companies, innovative tools that help them buy smarter.

Now, what makes it smart and what really drives me these days are the many AI-powered tools that we offer. For example, Spend Anomaly Monitoring. It’s an intelligent tracking system that monitors spending and proactively sends alerts to teams so they can make adjustments accordingly. All these tools are designed to help businesses discover savings, track spend patterns, and deliver insights that help organizations make smarter purchasing decisions.

Title icon

Liz Hoffman:
On a more serious note, we’re obviously at a moment of huge tensions and potentially generational change in Iran. You are Iranian, your family left before the revolution. What do you hope happens there?

Dara Khosrowshahi:
I hope there’s regime change because that regime has been absolutely terrible for the people of Iran. I view it as a regime whose goal is to drive a religion versus helping the people of the country. In the end, Iran is a country first and a people first. Whatever religion that they want to participate in is up to the individual. And so I think it’s a regime that has not earned its keep. I hope that regime change comes from the inside. And obviously you see all of the protests and all of the unrest because it’s a regime that’s serving themselves and not serving the people. I think regime change driven by countries, driven by Western countries especially, doesn’t have a good track record. It’s easy for me to say, here in the safety of the US, because people are being killed and they’re being killed at astounding numbers in Iran, but I’m hoping there’s a change.
The Iranian people deserve better. They deserve to be part of the world. They deserve to be part of the economic reality that’s changing so much. It’s a talented, educated workforce and they deserve better.

Liz Hoffman:
Have you talked to the president about it?

Dara Khosrowshahi:
I have talked to the president about it, but very quickly. I’ve thanked him for his support of the Iranian people and his concern about the Iranian people and the fact that he cares matters.

Liz Hoffman:
What’s your peak AI use case?

Dara Khosrowshahi:
It’s lots of things, but I vibe coded my own to do app. I’m a very organized, anal person in many ways, and I always use these third party apps. And my app is not as slick as the others, but it’s set up exactly as I want it, and that is the power of AI.

Liz Hoffman:
What’s the first thing on it after this podcast?

Dara Khosrowshahi:
After this podcast?

Liz Hoffman:
First thing on if it’s not classified or board confidential.

Rohan Goswami:
No, no. He can talk about board confidential if he wants to, Liz. Let him speak. Come on.

Dara Khosrowshahi:
It is pretty cool though. Again, it’s not as feature rich, but it’s exactly what I want. And that’s the power the AI bring you.

Liz Hoffman:
Cool. Wait, can we see it? Is there something? Can we see your phone?

Rohan Goswami:
Oh, yeah, can we see it?

Dara Khosrowshahi:
You can see my entire roadmap and life on it.

Liz Hoffman:
How do you think we got all of our secrets, Dara?

Dara Khosrowshahi:
Yeah. Exactly.

Rohan Goswami:
Ask people to show us their phones. You guys have been in the, not just package business, the freight business for almost a decade now at Uber Freight, but it’s not a big part of your business. It’s not growing as fast as the other parts of your business. How do you think about that longer term?

Dara Khosrowshahi:
I view freight as our first shot at getting into the end-to-end logistics business. We do... with Uber Eats to some extent, we’re doing on demand local last mile logistics. And one thing you may or may not know about Eats is if, for example, you order an Apple iPhone, you can get it delivered to your home, and that’s an Uber Eats courier who goes and picks it up from the store and gets it delivered to your home. So to some extent, already we’re doing third party, last mile delivery. What freight represents is first mile delivery, from the warehouse, from the factory to the warehouse. And then there’s this middle mile that we haven’t penetrated yet. Our vision is to build an end-to-end on demand next generation logistics network powered partially by truckers moving large trucks from factories to warehouses, smaller trucks from the warehouse to the store, and ultimately individual consumers from the store to the end consumer.
We’re the only company that’s taking it on. Freight is part of that ecosystem that we’re putting together. It’s a big challenge, but an enormous opportunity for us to build that end-to-end logistics stack, and it’s something I’m really, really excited about.

Rohan Goswami:
Sounds like you’re firmly committed to it.

Liz Hoffman:
Speaking of long distance logistics and freight, is there anywhere for Uber to play in space?

Dara Khosrowshahi:
I have not yet gone there. Let me do my job on this earth and I’ll let them take space for now.

Liz Hoffman:
Someone’s got to book all those cargo loads. I want to be able to go to space tourism. You can book that.

Rohan Goswami:
Lake Como is going to be flooded if we can... Suddenly everyone can afford to just pop over there because AI makes everything so cheap to get around.

Liz Hoffman:
Yeah. We have to find cooler, more exotic places for you to vacation.

Rohan Goswami:
Precisely.

Dara Khosrowshahi:
We’ll get you to the launchpad. The rest is up to the rocket company.

Rohan Goswami:
What do you make, broadly, of the panic around SaaS businesses, the selloff?

Dara Khosrowshahi:
Because the market is so afraid of uncertainty, they’re treating everyone similarly, the whole software market similarly. And there are, if you’re a thin UI layer on top of, let’s say, systems of record, you’re going to have to earn your keep. You’re either going to have to go deeper, embedding yourself inside of company systems or figure something else out, go broader so that you’ve got more context in terms of what people are looking for. I think, as it relates to the enterprise, it’s easy for me to vibe code a BS to do app for myself. It sounds cool, but it’s pretty easy. Enterprises are much more obsessed with getting things right. And so I think the pace of change in enterprise, especially large enterprise, is going to be slow. So I think where you want to look for the tip of the spear is really small businesses, SMBs, and what they’re doing.
And I think the market is probably overreacting short-term, but long-term, these trends are real and long-term the software companies are going to have to add more value or they’re going to disappear.

Liz Hoffman:
What would it take for you to... I assume you use Workday or Salesforce, what would it take for you to switch?

Dara Khosrowshahi:
It goes to what I was talking about, which is we’re not working on a Workday and Salesforce, but there are software companies who will remain unnamed who are buyers of ours where we’re spending like five, 10 million bucks a year on, and we’re looking to build that internally. I don’t know if it’s going to work or not. Maybe we’ll graduate to the big stuff, maybe we won’t. I just don’t know. They are all building AI layers on top of those systems of record. So I think if they change fast enough and they continue adding value and there’s a lot of value that they add today, they’ll be more than fine.

Liz Hoffman:
We’ll have you back in 10 years when Uber has this big business being the sort of SaaS provider for the fleet that Blackstone owns and then Blackstone decides to start vibe coding that internally and stop paying you whatever they’re paying you.

Dara Khosrowshahi:
First, we’ll have an AI summarize everything I said and compare it to the reality and then we’ll have that conversation.

Rohan Goswami:
You don’t even just show up. You can just send your AI agent. Liz and my AI agent will meet you. They can just talk and they can disperse it all out to the other AI agents.

Liz Hoffman:
Yeah. At some point we’re going to have to have you holding up a copy of today’s paper, like as a hostage video so that we know that it’s you and not some fake podcast guest. Listen, Dara, this was a lot of fun. Thanks for coming.

Rohan Goswami:
Thank you so much.

Dara Khosrowshahi:
Thank you. Take care.

Rohan Goswami:
He is always such a good interview. I don’t even know where to start, but what’d you think? What jumped out at you the most there?

Liz Hoffman:
I think there is real uncertainty around ... And he was pretty honest about it, but what is Uber? It still wants to be the front end, he said. He pretty clearly said he doesn’t mind being B2B, but he does not want to be B2B. And that is ultimately going to create some tension with those partners of which Uber is sometimes a shareholder. I think for now, sort of let a thousand flowers bloom and Waymo can go out and do some of its own stuff and do some partnership stuff without a lot of feathers being ruffled because there’s so much capital going into the space and there’s so many questions around it. But on Semafor’s media podcast, Mixed Signals, a couple weeks ago, they had the former CMO of McDonald’s who was saying, there’s like 23 different ways to order McDonald’s. And McDonald’s is not agnostic about that.
They really you to use their app or walk into their store, but they kind of live in reality. And as this stuff shakes out, that 23 will probably winnow and what Dara’s talking about with the multichannel distribution is just messy. And there’s going to be a lot of economics that have to be shared. It’s not insurmountable, but it’s clear that he thinks a lot about that.

Rohan Goswami:
Well, what we didn’t talk about at all is that in the US, they’re duopoly and Lyft is their main competitor and every one of us switches between those apps. And if I’m Dave Rischer, I’m sitting here thinking, “This is my moment to leapfrog.” Autonomous is an opportunity, the way they did with Citibikes, to actually undercut Uber, whether it’s on price or whatever other terms they can come up with, and take the lead for the first time ever. Because your point about winnowing, it’s not like Waymo has to just be partnered with Uber. They can do it themselves. They can go to Lyft. They could go to anyone and find a new way to get to consumers. They could do it in the Google app if they wanted to.

Liz Hoffman:
And you have to think that that is actually, it’s comforting to see Dara be as front footed about this if I’m an Uber investor because the underdog always has to take moonshots. They need the widest distribution of possible outcomes, which means they got to be trying a lot of stuff. And it’s often the incumbent that’s a little slower to think about it. And I take him at his word. Most CEOs actually don’t really want to go through the mishigas of large scale M&A. But as you’re thinking about it, the conversation we had, he’s talking about consolidating the gig economy. Does he buy Instacart? Does he buy Fiverr? Someone is going to aggregate that long tail of gig work, right?

Rohan Goswami:
Yeah, you have to.

Liz Hoffman:
Do they buy Hertz and is it become the way that I get my rental car? Because the most annoying part of a rental car is, how do I get to the place where I picked up the rental car? I don’t want to go to JFK.

Rohan Goswami:
And to be fair, you can get a rental car through Uber now, but it’s funny, Uber is weirdly, I think, briefly, going to be at the same moment they were at the very beginning where they were trying to acquire customers or retain customers, but they also need bodies in there, and they need metal in there. They need cars, they need people. And so yeah, I mean, if there was a moment to do a deal like a Hertz or a Fiverr or an Instacart, it’s now.

Liz Hoffman:
We’ve written a lot about large scale M&A and why it’s happening now. And actually, I think it is sort of fundamentally AI driven, but not quite for the reasons that people think. It isn’t like my business is going away because of AI, I got to go find another business. It’s like, actually AI is going to transform my business and I want to put that AI on top of the biggest possible business that I can. So you’re seeing these big consolidations with an eye towards, okay, let’s smash these things together and then put the AI layer on top. But obviously he has big ambitions in travel. You can tell.

Rohan Goswami:
Maybe not space, although I am hopeful to see an Uber rocket. Maybe not an Uber rocket, but like an Elon X-Uber collab.

Liz Hoffman:
Dara has gotten a lot of runway out of being the nice guy in tech and came in following just like an utter collapse and customer and investor confidence in what Uber was before. And I don’t know that competing in the Edge Lord space game is going to be value for him.

Rohan Goswami:
Maybe not.

Liz Hoffman:
The other thing that stood out to me was the autonomous solutions business that they just launched, what we talked about, is something that Uber built for itself and is deciding to commercialize. And Uber famously was kind of built on top of a business that Google had that he was just kind of putting on the internet, which is Google Maps. And eventually Google realized, oh no, we should charge for this. But Uber got a long way into its development by kind of taking something that a company was doing sort of for itself, sort of for its customers, not thinking that hard about how to commercialize it and built a giant business on top of that.
AWS was famously kind of the cloud business for Amazon before they sort of externalized it. And so you have companies looking around saying, “What is this thing that we have that we could commercialize?” And he was saying Expedia does a lot of that, that the white label for a bunch of independent engines. So you can kind of just see his kind of product brain working in interesting ways. And I think the AV solutions is the first really outward sign of that.

Rohan Goswami:
But you joked about this, that 10 years from now we’ll be talking to him about Blackstone’s fleet of autonomous vehicles. Again, I guess we’re reporters, we look for tension anywhere, but there is a tension in a business model when you’re so reliant on third parties, whether that is a driver. And what he may say may be true, but in New York, they’re all professional drivers. Whether that is a driver or a company, at some point if he’s vibe coding his own to do list, really what is to stop Blackstone from going, “We don’t need this.” Is it just the wrapper? Is it just the people? Is it the bodies? What’s the moat here?”

Liz Hoffman:
I think there’s a way to think about Uber as a wrapper for the AV mobility revolution. That is not a crazy pitch certainly to investors, certainly to customers. I don’t know. I mean, his comments about Workday and Salesforce, I tend to think that the SaaS selloff is, if not overblown, at least totally indiscriminate. The Blast Radius is four times wider than it ought to be.

Rohan Goswami:
There’s no thought to it whatsoever.

Liz Hoffman:
And that’s why I’m glad we asked him about it because the thing that most investors and reporters do not spend a lot of time is finding the person at Walmart who procures the software and asking them, “What is your appetite to switch?” Because the answer is zero. My appetite is zero.

Rohan Goswami:
So much cost, so much time.

Liz Hoffman:
And all you have to do is miss one payroll, right? All Uber has to do is miss one driver payout, and any money they would’ve saved from getting off of Workday just goes away. So I actually think he’s pretty clear out about that.

Rohan Goswami:
Yeah. I mean, perhaps uncharitable SAP was briefly Europe’s largest company off a business model of just that. It’s too hard to switch. And that’s kind of the bet Uber is making. They can get enough people on Uber one, get enough people on their platform. At some point, again, Ubers used to be subsidized. They used to be seven bucks to get across Manhattan. Now they’re not. And people, if they stick with it, they just stay. People are very resistant to that change.

Liz Hoffman:
I remember those good old days, but even at the time you knew it was being subsidized mostly by SoftBank, right? And you’re like, this cannot possibly last.

Rohan Goswami:
Well, you were smarter than me. I was in my teens.

Liz Hoffman:
No, I’m older than you and I was covering the Uber IPO at the time. And I remember these investment banks just throwing these outrageous valuations at them, 100, 120, 140 billion dollars. And you know, it took Uber, I think, until 2024 to actually hit those. It’s always good if you’ve been doing this long enough, and Dara certainly, has to know where the pipe runs out.

Rohan Goswami:
And to Dara’s credit, he really did manage to win the street over and financially right size the company. Not something we talked about that much, but he’s done a remarkable job transforming the financial profile of the company, transforming how the street looks at the company, taking from where we were at IPO to now to say nothing of obviously Uber’s brand is pretty resilient.

Liz Hoffman:
It is hard to remember just how toxic that brand was when he came in. And I remember covering the succession raise Travis had left and I think the names were, it was Meg Whitman and Jeff Immelt, and everyone was chasing and chasing and chasing. And I don’t remember ever hearing Dara’s name until he got the job. And so it’s sort of like a reminder that he is a very smooth operator. And you look back now, I won’t speak for Uber, but I think either Meg Whitman or Jeff Immelt would have been a very different choice. And so Dara is the approximate reason that we all got Quibi for five minutes. So it’s all full circle.

Rohan Goswami:
God, it really was five minutes. It is. Everything is a circle. Time’s a flat circle. Well, that’s it for us this week. Thanks for listening to Compound Interest from Semafor Business. Our show is produced by the incredible Josh Billinson.

Liz Hoffman:
A special thanks to Anna Pizzino, Jules Zern, Chad Lewis, Rachel Oppenheim, Tori Kuhr, Garett Wiley, and Daniel Hoeft.

Rohan Goswami:
Our engineer is Bob Mallory. Our theme music is by Steve Bone. If you like Compound Interest, please follow us wherever you get your podcasts and feel free to review us positively.

Liz Hoffman:
And if you want more, you can always sign up to get Semafor business in your inbox.