Brian Chesky on Airbnb’s emerging reinvention

BRIAN CHESKY: I think we can take the Airbnb model and bring it to all these different categories of traveling and living.

We’ve had to basically rebuild the app from the ground up to do this. We’ve had to build out entire new functions we didn’t really have before. We had to reorganize the company.

It’s a real challenge, but it’s the best kind of challenge. It keeps us young.

You know, one of my favorite musicians is Bob Dylan. He used to say, an artist has to be in a constant state of becoming. And as long as you don’t ever become something, if you arrive there, you’re gonna be okay. That’s the same thing as a company. You have to constantly reinvent yourself to stay young. I think we’re at that moment.

BOB SAFIAN: That’s Brian Chesky, co-founder and CEO of Airbnb. Right now, Brian is taking one of the biggest risks of his career, revamping one of the most successful start-ups ever. He calls it “the great reinvention,” promising a series of big shifts starting in the months ahead. In today’s episode, Brian explains why he’s taking this step now, where the company is headed, and how it connects to the impact of AI, including insights about China’s DeepSeek.

He also explains his unconventional employee philosophy — what he calls “the fallacy of autonomy” — and takes us inside his relationship with OpenAI CEO Sam Altman, revealing previously unheard details about Altman’s brief, yet tumultuous ouster from OpenAI a year ago November. The stories are memorable, and the lessons are rich. So let’s get to it. I’m Bob Safian, and this is Rapid Response.

[THEME MUSIC]

SAFIAN: I’m Bob Safian. I’m here with Brian Chesky, the co-founder and CEO of Airbnb. Brian, great to have you back on the show.

CHESKY: Oh, thank you for having me. I always love our conversations.

Copy LinkThe optimism driving Brian Chesky

SAFIAN: When we last talked back in May, you’d recently announced your Icons initiative, which was quite a creative, upbeat expansion of Airbnb. While there was uncertainty in the world at that moment, including conflict in Ukraine and Gaza, the playfulness of Icons kind of reflected what felt like a broader moment of optimism. Today, the mood in the world feels a little bit more uneasy, a little bit more uncertain with a new U.S. administration. How do you feel about where we are right now?

CHESKY: I think that I am ever the optimist, and I try not to be too affected by the circumstances happening in the world. I think Airbnb as a brand is a pretty optimistic brand. You kind of have to be an optimistic company to think that bringing strangers together to live in each other’s homes is a good idea.

It was a bit of a Rorschach test. Do you think it’s a good idea or a bad idea? Are you optimistic or pessimistic about the human spirit? I think I’ve always believed that humans are fundamentally good. We have a tendency to gravitate towards light, to want to gravitate to love in the long run.

But I also think there is a huge amount of polarization right now. I think the world is better than it’s reported to be.

I’m kind of impervious to some of the negativity. I think the negativity is a little overblown if compared to daily life. It’s not to say things aren’t challenging, but we also have more technology than we’ve ever had at our disposal to fix things. If we can only just cooperate, and I think cooperation is going to require an optimism. It’s going to require us believing other people in optimistic circumstances about them so we can work together to actually fix problems. I worry that a negative mindset becomes a self-reinforcing kind of prophecy.

Copy LinkAirbnb’s position on speaking out politically

SAFIAN: A bunch of your tech peers chose to attend the inauguration, Elon Musk, Jeff Bezos, Mark Zuckerberg, Tim Cook. You weren’t part of that. Was that a complicated decision for you? I know you have an informal network of CEOs that you sometimes chat with. Was this a topic of discussion among the group?

CHESKY: I think Airbnb is a little different. I think these really large tech companies are very much regulated at the national level. What happens at the national level from a policy standpoint has potentially very large ramifications on these platforms. With Airbnb, we have to deal with a lot of laws and regulations.

Most of our laws and regulations are at the city level and the state level. Most of what we need to do is focus on working with stakeholders in 100,000 cities. So we work really closely with policymakers. There really wasn’t a reason for me to attend the inauguration. So I wasn’t looking just to join like a big party or big trend.

I haven’t really spoken out that much about politics recently. It doesn’t mean I won’t speak out in the future, but there need to be some countervailing forces in society, some things bringing people together.

And that’s, I think, where the Airbnb brand best comes into play. I think ultimately, like we have 4 million people staying together. We have Trump voters hosting Kamala Harris voters and vice versa. And that’s just in the United States. We have people from all different cultures, all different religions, all different political groups coming together.

I really feel like if we were put on earth to do one thing, it’s to really help bring people together in communities. That’s probably the secret sauce of Airbnb. But we will speak out when there’s an issue, like when there was a travel ban, we spoke out, because we know a little bit about travel and what the brand implications of banning travel would be.

I think as Silicon Valley is going more political, I think that the best thing I can do is let Airbnb chart our own course, and sometimes we will look like one side or the other. We’re very much about building global community and bringing people together.

We’re also very much about making sure we deliver value to shareholders. So is that a threading of the needle? I mean, I think we have our own views. And I think the best characteristics are when people have their own views.

So I think we just try to chart our own course, and we do have a pretty global mindset. It’s because we have a global view. We run a global business. Airbnb has been used more than 2 billion times. We’re in more countries than Coca-Cola. We’re truly distributed in every geography in the world, and yet we’re still an American corporation. So I do think we have a foot in each world.

Copy LinkAirbnb’s post-pandemic transformation

SAFIAN: You said recently that Airbnb would undergo the greatest reinvention in the company’s history in the months ahead.

CHESKY: Yeah.

SAFIAN: Why is that necessary? Can you share any details about it?

CHESKY: I’ll try to share just a little bit of detail without giving away everything, because you’ll have to have me back on. I always need to leave you wanting more.

Just to give you a little bit of the arc of the company, we had this crazy idea. People said the idea wouldn’t work. It was like the craziest idea that ever worked, and it did work.

And I went on one of the biggest rocket ships in the history of Silicon Valley. Then, of course, the pandemic happened. We lost 80 percent of our business in eight weeks. People were asking, is this the end of Airbnb? It really wasn’t. It was just the beginning.

We had this unbelievable turnaround and went public. Our market cap hit 100 billion. Expectations are really high. So the last four years, what we’ve been doing is really trying to perfect our core business.

We basically asked everyone who didn’t like Airbnb, why don’t you like Airbnb? What are your issues with the service? We were really methodical. We blueprinted the end-to-end experience for guests and hosts. And over the last four years, we’ve made 535 features and upgrades.

I think our core service is better than ever. It’s not ever going to be finished. It’s kind of like if you want to build a skyscraper and you’re a four-story building, you can’t just keep building new floors. At some point, you need to rebuild the foundation so you can hold more floors.

That’s kind of what we’ve been doing. We’ve been improving our service, rebuilding the foundation, waiting for this moment. This is the moment where we want to expand the Airbnb model beyond a place to stay.

This is something I’ve been thinking about for 10 years. The reason we haven’t done it successfully is there’s this old quote in Silicon Valley. I think Marc Andreessen said it. He said, there are no bad ideas, just ideas that are too early, which is kind of counterintuitive. Most people think ideas are too late. Often Silicon Valley ideas are too early, and we probably weren’t ready to expand beyond our core business until this moment. I think we can take the Airbnb model and bring it to all these different categories of traveling and living.

We’ve had to basically rebuild the app from the ground up to do this. We’ve had to build out entire new functions we didn’t really have before. We had to reorganize the company. We had to become much more collaborative. It’s very rare to get out of your groove at this age. It’s kind of like learning a new language later in life.

It’s a real challenge, but it’s the best kind of challenge. It keeps us young and keeps us from becoming a big old company by having to fundamentally change and reinvent who we are.

Copy LinkAI’s role in Airbnb’s future

SAFIAN: And is this reinvention an extension of what you’re doing now? Like, I know you talked to me previously about exploring an AI native interface that you’ve been doing with Jony Ive, the legendary Apple designer. Is that the kind of thing you’re talking about? Or is this something that’s going to blow my mind the same way the original Airbnb made people think, oh, that’s never going to work?

CHESKY: I don’t want to overpromise with the second example, like it’s going to blow your mind. I hope some people say it doesn’t work. Somebody once told me in the early days of Airbnb, don’t worry about people stealing your idea. If it’s any good, everyone will dismiss it. So if no one thinks it won’t work, then I probably didn’t swing hard enough for the fences. I really do want to keep launching things where the audience is a little divided. People are like, “yeah, this seems like a really good idea, but it seems a little bit crazy, and, oh, that’s too crazy to work.”

And that kind of knows maybe you’re on the frontier. And if no one reacts that way, then maybe we weren’t quite bold enough. I think it’s so much bigger than just AI. I say just AI, but I think everyone’s doing that. I think AI is kind of more of an inevitability.

Here’s the simple way I describe AI: you have the compute, the chips, and there’s a lot of development around the chips, and GPUs seem to be powering this whole movement. The new DeepSeek model may or may not even need to use those GPUs; maybe they can use cheaper GPUs, and then you’ve got a lot of development around the models.

Obviously, you have OpenAI’s model, Google’s model, Meta’s model. Now there are some really powerful open-source models, including DeepSeek. And it seems like at those two layers, the chip and the model, there’s been a lot of innovation. But if you think of the models as infrastructure, like a highway, the cars on the highway are like the applications. Well, what do I do with these highways? Can I drive on them? What kind of car can I get? You can think of those as the applications.

Well, here’s a simple test: take your phone and just look at your home screen. Look at the apps on your home screen, and then ask yourself a simple question. How many of those apps have fundamentally changed because of this generative AI? Now, you might have ChatGPT on your home screen. That’s a native AI app. You might have other native AI apps, but most apps, including Airbnb, are not fundamentally different. So, to bring AI, generative AI, to applications to actually improve our day-to-day lives, we are still not decades away, obviously, but we’re more than months away.

I don’t think Airbnb is behind other companies, and this will still play out over multiple years for us. I think we’re definitely one of the companies that is at the forefront of deciding on the interface paradigm for AI after multi-touch or how to get the most out of multi-touch.

And I also think about how you tune the models for very specific use cases. I do not think there’s going to be one AI chatbot or chat agent to rule them all.

I think some people think it’s like Google Search, where there’s one thing that rules them all. I don’t think that. I think it’ll be more like the App Store, with a different thing for each, like a different tool for each job.

Copy LinkExpanding Airbnb beyond short-term lodging

SAFIAN: But timing-wise, it’s a little early even to know what kind of car to build.

CHESKY: I think that if I were to predict it, over the next couple of years, you’re going to see a lot of interesting innovation at the application level. But I don’t think we’ve seen much yet.

Unless somebody is about to launch something any month now that is going to change the world — that’s possible. But, notwithstanding that, probably not. So, to answer your question, I think what we’re going to be doing is something even more fundamental than using the latest and greatest AI technology, which I think it is. By the way, this could still be huge for us because, like everyone’s saying, the future of AI is agentic. Agentic means agent. Customer service agents, concierge, travel agents are kind of the iconic types of agents people think of.

So Airbnb, of course, would be an inevitably interesting place for AI, but I think there’s something even more fundamental, which is in the late ’90s, Amazon was just a bookseller. Eventually, they started selling DVDs and CDs, and people used to buy those. Then they started selling electronics, and then they started selling outdoor equipment, kitchen utensils, appliances, and then clothing.

Eventually, Amazon sold everything. They went from books to all retail. I think we are about to go through a similar transformation where we go from short-term stays — typically, an Airbnb house that you book for under a week averages about five to six nights — to long-term stays, and then a bunch of services and a bunch of experiences.

Everything you need to travel and live anywhere around the world. I think there are literally hundreds of different verticals that one could go into. We probably won’t go into all of them, but that’s really what we’re talking about. It’s really about abstracting the Airbnb platform to be able to go into more things than just housing.

To me, that would be the most revolutionary thing we did.

SAFIAN: But you must have to make tough choices in this, like a one-stop shop. It could sound like sort of a less focused, more catch-all business where you’re in all kinds of things, right? Because you have this community and customer base that you can leverage and introduce to different kinds of experiences and tools. You can’t go into everything.

CHESKY: Yeah, so the governor is three things. Number one, you can’t do everything well at once. You have to focus. You have to pick. Number two, you can only do things you can do well. We can’t do everything well. We should only do things where there’s a reason we do it. We track the number of devices that access Airbnb. That number is 1.6 billion a year. So 1.6 billion devices a year, which could easily, if the average person uses two devices, mean 800 million, close to a billion, people are using Airbnb every year. Think about that. That’s a very big number. Now, we have enough traffic. We could just slap our logo on something that’s not any better than anyone else’s product, and it would sell some, but we don’t want to do that.

If we put our logo on something, it’s because there’s something different, something better, something more authentic, higher quality, easier to book. So, the first thing is we can’t do everything at once well. Number two, we should only do things we can do well and do differently. And the third is, I think Airbnb is known for user interface, great design, and creating a really simple experience. So everything has to be coherent and cohesive.

People use Airbnb to travel, and eventually, they’ll probably use Airbnb in their own city, but they’ll probably use it to explore their community. So if we were to think about that as a workflow, like going on a trip, moving somewhere for a month, that is a coherent way to keep the interface really simple and really cross-selling different things in a way that would make sense to a customer.

So those are, I guess, our big principles.

Copy LinkFrom open marketplace to a focus on quality

SAFIAN: No, that makes sense. I mean, you and I have talked before about physical experiences and Airbnb wanting to be more connected to those, but I know there have been rumors for years about an Airbnb airline.

CHESKY: I can rule out Airbnb rail/airline in this decade.

SAFIAN: Right. Like it’s still going to be primarily for you, a virtual engagement, but allowing people to engage more in the physical world as well.

CHESKY: Yeah, and I think the thing that Icons did for us is we built those spaces, and we’re probably not going to build a lot of things. Like the things we built are probably going to be mostly marketing promotions, just because if we have 4 million guests a night, and each stay is like a half a million-dollar bedroom per bedroom, you’re looking at like 2 trillion per night per real estate. The numbers get really big very, very quickly.

But I do think there’s something between us having no responsibility for the stay, an open marketplace or with limited responsibility, and building a house. I think there’s something in between. I think the in-between where we’re going to take Airbnb is managing the quality. We’re going to verify things, create a quality seal of approval that’s above and beyond our rating system. When you see this, you’ll know we vetted this, we stand behind this, and we have a guarantee of quality. I think Airbnb going from an open marketplace to a marketplace that’s known for excellence and quality is the other vector.

Copy LinkReinventing Airbnb’s core business

SAFIAN: And at the same time, I mean, you mentioned this, over 500 product changes in the core product, you’re still going to keep tweaking and moving that product forward.

CHESKY: Yep. The core is still — eventually we’re going to reinvent our core business too. We’re going to eventually reinvent that. One of the things that’s happening is, as we go into these new verticals, we’re kind of looking at our core business with fresh eyes. The biggest challenge is the biggest challenge of core business.

All these decisions that Joe and I made in our 20s — some of them seem kind of arbitrary — they become kind of doctrine, and there’s this dogma that it’s so successful, you can’t possibly change it. Sometimes the only way to convince people you can change your business is to get them to do something new at your company with fresh eyes, with no dogma, and you realize, okay, we can blow up the model. Once you do, you’re like, wait, let’s bring some of that to the core business. I think you’re going to see over the next couple of years a major step change in our core business. We’re going to keep making metronomic improvements, and then just like we do a big release this year, we’re going to do one again next year and the year after, you’re going to see some pretty big improvements to the core business.

Copy LinkMatching hosts with property managers

SAFIAN: You recently launched a co-host network to pair hosts with property managers.

CHESKY: There were a number of people that were very interested in hosting, but the number one issue people have is they say, “I don’t have time to host.” But we found these other people who are really good hosts on Airbnb, and they told us they wanted to expand, but they don’t have the real estate.

So we thought, what if we built a marketplace where we found people who want to put their homes on Airbnb and matched them with people who want to host those homes?

We now have 15,000 co-hosts. If you go on Airbnb, you can find somebody who can manage your property for you and welcome the guests. Now, these are the best hosts on Airbnb. The average rating is significantly higher than the average.

We really want to help people understand that Airbnb is going to become an ecosystem.

I don’t blame somebody for saying, “You’re kind of the same thing you were 17 years ago.” It’s kind of the same idea. It’s expanded. It’s much more refined. I think the thing that keeps you young is to continue to change. One of my favorite musicians is Bob Dylan.

He used to say, an artist has to be in a constant state of becoming. As long as you don’t ever become something, you’ll be okay. That’s the same thing as a company. You have to constantly reinvent yourself to stay young. I think we’re at that moment.

SAFIAN: It’s tricky for any company to balance expansion and reinvention while keeping their core business perfectly intact. I’ll be curious to see how Brian handles it this year. After the break, Brian challenges business leaders to rethink their organizational structure and strategy for success, and reveals previously unheard details from Sam Altman’s tumultuous chapter away from OpenAI. So stay with us.

[AD BREAK]

SAFIAN: Before the break, Airbnb’s Brian Chesky shared why now is the perfect moment for the company’s so-called “reinvention” and why it’s so important for companies, however old, to stay young. Now he explains his theory behind the “fallacy of employee autonomy” and shares the inside story behind Sam Altman’s mysterious ousting from OpenAI. Let’s dive back in.

Copy LinkLessons from Sam Altman’s OpenAI ousting

You mentioned AI earlier. Sam Altman recently said that you helped him during that brief period in 2023 when he was ousted from OpenAI. You saved him from making some mistakes. Can you share anything about that and the sort of advice you gave him?

CHESKY: Yeah, sure. Maybe just to back up, I’ve known Sam for like 16 years because he was with Y Combinator. He was the first big Y Combinator founder, I think, in 2006, and then he became president of Y Combinator. We kept in touch, and I basically reconnected with him in, I think, the summer of 2022. ChatGPT hadn’t launched yet, but DALL-E, the image generator, had launched, and I realized he had just built a really powerful tool for the creative community.

And I said, “Hey, listen, I’d love to help you, just give some advice about how to build tools to create a community.” My general feedback was, involve the creative community. Don’t create a tool that makes the community think they’re going to be replaced when that community could actually use the tool to make even better art.

And so we started talking about it. I started talking about how I do launches, like this thing that’s now called “founder mode.” I kind of went through all this, little did I know that he was working on this other thing called ChatGPT, which is a large language model.

He launched it November 30, 2022, and the world literally changed in about three to four days. On November 1, 2022, no one was talking about AI other than people interested in AI. By December 5, the whole world was talking about AI.

So it completely ushered in the AI era. At that moment, I told Sam, “You’re basically going to go through everything I’ve gone through in the last 10 years, but you’re going to go through it in like a year.” I said, “You’re going to have a crazy rocket ship, and I’m here if you need help.” At that point, we started meeting every, I don’t know, every couple of weeks or something. I was just giving him advice.

And I gave him advice on hiring, how you scale, building out your brand, your product, and your design — kind of everything that wasn’t the core research and technology. So I’m giving him advice. Everything seems fine.

Then it’s the next November, I think it was, and I got a text on my phone. I’m in a WhatsApp group, and everyone said, “Oh, did you hear the news? Sam was fired from OpenAI.” And I’m like, “Fired?” I texted Sam, “What the hell is going on?”

And I was one of the first people who responded to it. And he said, “So brutal. I was fired. I don’t really know why I was fired.” And I was like, “You don’t know why you were fired, but you were fired, and it was made public immediately.” Then the circumstances got even fishier.

I was told that Microsoft was only given a 15 or 30-minute heads up. There was no actual investigation, and Greg Brockman, the other co-founder, was removed from the board immediately. I said, “Okay, I don’t know what happened, but you deserve a fair process.”

I kind of just helped him with what to do over the next three to four days. I gave him quite a bit of advice on PR. I basically told him, “You have to do two things. Tell them what you know and don’t know. Don’t point fingers at anyone; don’t blame anyone. Just tell them what you know and don’t know. And the fact that you don’t really know why you were fired, say that. Say this is what I know. That’s what I don’t know. Then say, the most important thing is to turn your attention to the employees and make sure they’re taken care of, the company’s taken care of.”

And ultimately, I think he was going to do all this anyway, but the way he handled it, I think the employees really rallied to his side. The board eventually realized an AI company without AI talent isn’t really a company. There’s nothing other than the talent at that company, and they were all going to vote together and go somewhere together.

We thought the board was going to reinstate Sam, but instead, they announced an interim CEO named Emmett Shear. Emmett Shear actually turned out to be somebody I knew, another mentor of mine from the early days. So, Emmett, I, and some others were able to work in the background to get the board talking to Sam and just have a healthy dialogue.

Essentially, what we did was help bring everyone to the table and realize this is an internet treasure. This is one of the most important companies in the world ever in Silicon Valley, ushering in this AI gold rush. But also, we have to set a good example of Silicon Valley.

We want to show that founders are protected to some extent to ensure fair processes. We don’t want to invite more board activism. We really want to make sure that this really important company is held together because the world really needs this innovation and advancement. People way beyond me were able to find a resolution. I was just trying to be a helpful person. I have no official role in OpenAI. I just tried to keep a company together. At the end of the whole thing, Sam asked me, “How can I repay you?” I told him what Ron Conway and some other people told me when I wanted to repay them for helping me. They said, “Pass it forward,” and that’s what I told him. I said, “I’m not asking for anything other than to pass it on to the next generation.”

SAFIAN: Yeah, it really was a wild sequence. Nobody really understood what was going on.

CHESKY: No, it was completely crazy. I think because it was so sudden, people assumed something incredibly nefarious had happened. I asked the questions, and I asked Sam, I asked Greg, and I chose to believe them. I think the board eventually did an independent investigation, and they corroborated what my suspicion was, which was there were not grounds for termination. I do think it’s really important that boards have healthy governing processes. There is a sense of due process. If a board ever thinks there’s an issue with the CEO, it’s really important they have a healthy, open dialogue.

The CEO shouldn’t be blindsided. I think it’s on Sam too, that it’s important that he has good communication with his board, and he’s open and forthcoming. So I think there’s a big lesson in board communication.

SAFIAN: It sounds like you’re saying both the board could listen more to Sam, and Sam could have talked more with the board, and maybe they would have resolved it without having this kind of a problem.

CHESKY: Yeah, and I think it was this esoteric, peculiar situation. It was founded as a nonprofit, and I’m not an expert in this, but it probably, in hindsight, should not have been created as a nonprofit. There’s a reason why most innovation is created by for-profit companies because the best people want equity, they want to get paid really well.

These things are typically very capital-intensive, and I am a capitalist. I think almost all tech founders are. I think a company can do things for the benefit of humanity and still be profitable. So I think the founding charter of it being a nonprofit was conceived before there was any notion that this thing would consume tens of billions of dollars of capital, and I just don’t think a nonprofit is a good vehicle to be able to do that.

So I think there was just this issue where there was a legacy nonprofit organization, and I think that was just not set up to do what they’re trying to do today, even though I think it was a very noble mission, and I think it probably ran its course.

Copy LinkRedefining employee autonomy and empowerment

SAFIAN: I’m going to change gears just a little bit. You talked recently about the “fallacy of autonomy” when it comes to the workplace, that you don’t think most employees want autonomy. But you’ve also stressed the importance of empowering employees. So I just want to ask you, what’s the difference when you say this between autonomy and empowerment?

CHESKY: Here’s what I’ve found. For years, I treated Airbnb employees like customers. I thought I serve employees.

If they’re happy, they’re going to properly serve. And so I would ask them what they wanted. I would try to give them what they wanted, and something happened. People just kind of got unhappy and less happy and less happy. It seemed like the more I asked them what to do and the more I did what they wanted me to do, the worse things got. Because what people say is they’d say, “Oh, it’s kind of hard to work here.” So I tried to divide the company up further. I would give them more space, more autonomy.

Somebody would say, for example, “Hey, it’s really hard for me to do my job because I have to coordinate with all these groups. I need to get the engineering department to prioritize something on the roadmap. I need to get the payments team to add it to the roadmap. I need to get these marketing and creative resources. And there’s like 10 other people asking them for work. So why don’t you let me build my own thing? Why don’t you let me build my own group?”

This is kind of where the divisionalization of companies comes in. They start to divide the company up. Initially, it seems faster. You create these general managers. You start to divide the company. The CEO is told to stop checking the work.

You need to trust us. And if you trust us, you shouldn’t be looking over our shoulder. Just trust us to do our job. So the CEO stops being in the details of the work. Then the leaders model the CEO, and they stop being in the details of their work, and they start to push decision-making down.

We end up having leaders who aren’t in the details. Companies rowing in very different directions, and they just are going in a thousand directions.

I think there’s a different way of running a company that’s worth trying. It’s the way a start-up is run. Number one, stay as small as possible. Stay as small as possible means have as few people as possible, do as few things as possible, have few details, perfect every detail. Stay small like a start-up.

The second thing is be in the details. I think the CEO should be in the details. CEOs are mostly told not to be in the details because that’s micromanaging. If you micromanage people, you disempower them. I don’t think that’s accurate. I think the best way to empower somebody is to be in the details. You don’t tell them how to do their job, but you know everything they’re doing so you can actually help them.

The third and maybe most important principle is everyone rows in the same direction. If I had a thousand people, I don’t want to put them in a thousand boats. I want to put a thousand people in one boat, and I want everyone rowing in one direction.

There are downsides to putting everyone in one boat. You can’t go in different directions at once. You’re kind of making a single bet. But if the bet is right, you get so much more velocity and momentum when everyone rows in the same direction.

So the reason I’m involved is I want to model behavior to be in the details. I want to set the tempo of the company every day. I want to set the standards. I want to make the company feel as small as possible, as integrated as possible. I want people to collaborate. I want them to row in one direction. I don’t want to take big, giant leaps. This is only possible if we’re all rowing in the same direction.

But that’s not the same as empowerment. Empowerment is like, “I have the power to do my job the way I think I should do it. I can actually be really effective.” I actually think people are more empowered today than they used to be. When you give autonomy to employees and you let them row in different directions, it initially feels empowering until you hit up against, “I don’t have enough resources. I have to make friends with everyone in the organization to be effective. I have to manage my influence now. I can’t speak up and disagree with people because then they’re not going to work with me anymore. I have to cut through the bureaucracy and red tape.”

I’m not saying everyone should do everything I’m doing. I do think, though, that people should try to be as small as possible, in as many details as possible, and row in the same direction as long as they possibly can because I think you’ll get a lot more out of your company.

SAFIAN: It sounds like, as a designer, the constraints around you as an employee are helpful too, right? Like you’re empowered within the context of constraints that your manager or your leader is saying this is the direction, this is what we’re trying to get done.

CHESKY: I’ll just say something like, I want you to have any power, and you don’t need to have any friends to get work done at Airbnb. Not to say that no one has power or friends, but you don’t need it. At most companies, you either need people to do whatever you say, because they report to you, and you have budget resources, or you manage by influence, and everyone’s got to like you and be your friend.

You need neither of those characteristics to be effective at Airbnb. The other thing is, we don’t greenlight everything, but if we greenlight something, if you have an idea and we greenlight it, we put 100 percent of the weight of the company behind it. Somewhere, probably more than 95 percent of things that we work on ship and successfully ship. At a huge number of companies, projects get killed, a huge percentage of projects.

So we have the same priority: everyone in the company works on the same thing together. It’s like the top 10 priorities. There’s not one person in the company with different priorities if they’re touching a consumer-facing product.

This idea of everyone rowing together means there are some narrow constraints. But within those constraints, you can be incredibly creative, and almost everything you work on ships and sees the light of day. If you work on it, a thousand people are rowing behind you to thrust you forward.

Copy LinkWhy Airbnb embraces a hybrid work model

SAFIAN: And this idea of empowering versus autonomy, does it play at all into the debate over return-to-office mandates? What’s your position on working from home versus in-person?

CHESKY: I think there’s a huge value to people being in the office sometimes. I have not found a huge value in people being in the office all the time. So, let’s say you’re a global company. I want everyone to be together, but everyone lives in different cities, right? So unless all of your employees live within a 30-minute commuting radius, you either have to limit your hiring to a very narrow talent pool, or you’re hiring all over the world, and you have all these offices, and you’re telling people to go into an office.

But they’re going into an office. Who cares? What I want is, for the most part, people coming to the San Francisco office, but I can’t get everyone to move here to San Francisco, and I can’t get them to fly here every week. So I have a simple rule: we basically ask people to come to San Francisco one week a month.

Some people come for just two or three days. Some people come for the full week. Now all these people in different offices come to the San Francisco office. The reason they’re coming to the office — I don’t worry about how much work they’re getting done.

We have really rigorous milestones. We have really rigorous performance management. If you want a team to work harder, don’t make them come to the office, give them a crazy deadline and check on their progress every week. That’s how you get them to work harder, not by being in the office. I don’t care where you are.

We treat the return to office as gathering weeks. It’s typically one week a month where we coordinate everyone being together, and that’s for collaboration.

But mostly, we have a flexible policy where you come in one week a month, and otherwise, we’re flexible. This allows us to not make a trade-off regarding where employees are, and I think the one week we’re in the office is superior to other companies where people are in different offices, right? All of our leaders are in the same office. They’re not in different offices.

SAFIAN: And flying everybody all over the world to come to San Francisco, is the cost of that worth it?

CHESKY: There’s real estate savings that we have by needing a smaller footprint because people aren’t coming in five days a week, thousands of people. And it is worth it. It might even be more worth it than the old cost. If this new model was more expensive than the old model, it still might be superior because I think the output for us is superior. But my understanding is that this is still, on balance, more affordable.

Remember, we do have a critical mass of people in San Francisco. We’re just really flying those who aren’t in San Francisco.

SAFIAN: Well, Brian, this has been great. I hope we can keep talking about this journey. I can’t wait to see where it takes us.

CHESKY: I can’t wait to take you on this journey.

SAFIAN: Whether it’s Sam Altman versus the OpenAI board or Brian getting a thousand employees to row in the same direction, I’m reminded of the unrivaled importance of clarity and communication. The beauty in narrowing an organization’s efforts to a few goals is the clarity it affords. Whether or not you’re undergoing an ambitious reinvention, today’s overwhelming climate requires more focus and clarity than ever. To unlock success this year might mean being more discerning with our effort and attention. Thanks for listening.