Travis Kalanick Joins, Spotify CEO, Nikesh from Palo Alto Networks, xAI Rebuild, Apple Faces Slop Allegations

You're watching TVPN. Today is Friday, March 13th, 2026. We are live from the TVPN UltraDome, the temple of technology.

The fortress of finance.

The capital of capital. Let me tell you about ramp.com. Time is money. Save both. Easy to use corporate cards, bill pay, accounting, and a whole lot more all in one place.

Boy, is it good to be back.

It's great to be back, but it's Friday the 13th, so you know what you got to do. Salt over the left shoulder.

Oh yeah?

You know about this?

They didn't.

Yeah.

Throwing salt over your left shoulder is good luck. It counteracts bad luck. Friday the 13th is obviously bad luck.

That's great.

But not anymore since we have salt thrown over our shoulder.

Good hack.

I was very interested in Patrick Hollison's post. We talked about this a little bit. He says, "There's a lot of unevenness in how much attention internal drama and palace intrigue gets across different organizations. As far as I can tell, this is substantially a matter of path dependency. We know the characters in the sitcom of certain organizations but not the others." self-reinforcing lock-in effects. How much does one hear about the power struggles at Chevron or the Department of Agriculture? There's even significant heterogeneity between ostensibly similar companies within sectors.

We get it.

You want a bunch of palace intrigue stories about Stripe. We get it, Patrick. Just kidding. No, Stripe has not ever really been in the tumultuous drama, oh, horse race. There's been investors who have talked about various valuations, but overall, the company has been sort of smooth sailing for almost 2 decades. But I wanted to think about this more in the context of other countries' startup ecosystems and also—

Yeah, the closest they ever got to drama was around the Bolt fast dynamic.

That's right.

Because they backed fast.

Yeah, you're right. There was a little bit.

Seemingly because they felt like some competitive pressure there. They wanted the fast checkout provider to be bigger than Stripe.

Fast takeoff.

Volt was famously not. They had built their own payment rails.

Yeah, yeah, yeah. But I mean, the Collison brothers are just so class— they're such class acts that they don't really wade around in the mud. They don't roll around in the slob farm. And it's like, it's a B2B company, so there's less, like, just general viral intrigue around it.

And even in that, even in that whole saga, you never saw a single comment.

No, no, no. And they weren't taking shots at anyone. They were very classy. What do you think?

There was some controversy when the cheeky pint Elon interview came out and John Carlson— how many Guinnesses were they drinking?

OK, OK. That was a bit of palace intrigue. How many cheeky pints were they drinking in the palace?

We're getting somewhere. John did come out on top of that, right?

He did. He did.

Yeah, yeah, yeah.

He explained. He explained. So years ago, I was talking to an American VC about New Zealand. And I asked him about the startup community. The country was beautiful. He said, highly developed, democratic. It consistently ranks among the world's most stable, wealthy, and well-governed societies. And he would spend a lot of time there. I was like, there's got to be some company that can break out and become a power law company. If you're going to be there spending a lot of time, you'll probably meet cool people, interesting people. Maybe there will be a great New Zealand company that comes out of this. And he was like, I'm not so sure, because his assessment distilled down to something along the lines of, they are suffering from a bad case of tall poppy syndrome. So what is tall poppy syndrome? Put simply, individuals who achieve visible success are criticized, attacked, or socially cut down because they stand above the others in the crowd. The metaphor should be obvious. When you cut the tallest flowers in a field, the surface appears even.

Let's pull up Tyler for a second.

Why? What is that? There we go. He's a tall poppy. He's getting cut. So Anthropologists often call this leveling behavior, and it goes back to hunter-gatherer societies. Good hunters might be mocked or discouraged from bragging. Sharing would be encouraged to prevent resource concentration. And there's nothing bad about a preference for humility. America has long been suspicious, and Americans have long been suspicious of the Lamborghini-driving, self-promotional Instagram course hustlers, and for good reason. Those folks are usually selling overpriced junk food, basically. But the strong form of tall poppy syndrome does lead to lower startup formation. If you think that you will immediately be cut down in your society, you don't actually go out and hunt. You actually don't go out and bring back the big deer or whatever you're hunting. The bacon. The bacon, yeah. You don't bring home the bacon in the first place. You don't even try because you're so worried about this tall poppy syndrome in your society. You wind up with lower startup formation, fewer truly scaled companies, less like aggressive growth plans, and ultimately a talent exodus or brain drain. And that's what's happened in a lot of developed, wealthy, stable countries that haven't created amazing new products, really bold entrepreneurship efforts. A lot of it's because of the tall poppy syndrome. So America's never had this problem. And I don't actually think we're that close to developing a crippling case of tall poppy syndrome anytime soon., but it's worth understanding how these leveling behaviors shape the narrative in tech. So there's this unevenness that Patrick Collison identifies around internal drama, attention. It does not seem correlated with market cap at stake or even how well-known certain founders are at the time that something's happening that could be dramatic. So I was thinking back to Elon Musk and Tesla. He's like perhaps the most household name entrepreneur in history, certainly right now. Tesla has something like a trillion dollars of market cap at stake. If someone leaves Tesla to start a competitor, that should be incredibly dramatic. Do you know who I'm talking about?

I do. Who is it? 'Cause I took a pass.

Oh, you took a pass, okay.

On your essay, but I don't remember his name.

It's Peter Rawlinson. He designed, he was the chief engineer of the Model S at Tesla, their super successful, high-end sedan, the thing that really ushered in the EV boom. He's the guy that engineered it. He leaves, he goes to Lucid Motors, eventually becomes the CEO and the face of the company, develops a direct competitor to the Tesla Model S, which is now discontinued. The Lucid Air is fancier, faster. It has a whole bunch of other criteria that satisfy that market segment. And it should be like this knockout drag-out fight that's very interesting. Lucid hasn't put a dent in Tesla largely, but it is palace intrigue, and yet no one really cares. And most people in tech can't even name a second person at Tesla after Elon. Maybe JB Straubel, Redwood Materials, but he's out now. Yeah, Karpathy. Karpathy, yeah, that would be one, but it's like, who's active? At SpaceX, everyone knows Gwynne Shotwell. But like beyond that, Kiko Donchev, who else? Like it gets really, really hard. And why is that? Well, some industries, some companies have different cultures. So if you think about Tesla, Chevron, Department of Agriculture, they're probably not encouraging their employees to go direct. They might actually be discouraging it. They might have rules around what you can post about publicly.

One xAI employee went very direct.

He went very direct and he didn't last long at the company. Like electric vehicles, Chevron oil and gas, the Department of Agriculture, these organizations don't necessarily have a lineage that traces back to academia, which is very open source. You publish your research, go to a conference, put up slides. What are you thinking about? What are you researching? Share it all. Talk about it. Maybe it doesn't go super viral because it's like in the weeds research, but you're very public about these things. Then, And then also the blogosphere, like in the AI world, a lot of leading thinkers, leading employees had blogs and had done podcasts before. And so it was like a continuation of that, the idea that you were someone who wrote essays online and shared all of your thoughts, that just carried through to the AI era. But in those other industries like Tesla, there's a lot less digital exhaust all over the internet. The internet also rewards taking shots at the tall poppy. Tech generally wants to appear nice, and they don't want to punch down. Like, when a company rises and falls, you typically don't see serious people in tech really being like, "I called it, victory lap." Like, it's seen as uncouth.

The company has to be doing some really egregious stuff.

Totally.

Like a true violation of the social contract, not just, "Look, everyone tried really hard, they got beat, and the investors, you know, sort of got 50 cents on the dollar back." Something like that. No one's really cheering for that. It's pretty rare. Some anonymous accounts are. And— but in general, aiming for the top is the only option if you do wanna punch. You don't wanna punch down, but you still wanna punch. Where do you punch? You punch up. You punch at the top dog. And the better a company is doing, the more attention you'll get for taking a shot at it. And this is seen as like contrarian. Everyone thinks this company's dominant in this particular category. Well, take a shot at it. It's over.

They're dead.

And you're gonna get a lot of views because it is a counterintuitive take. Everyone else has the Glazonator 3000 out, you pull out the dunk, and you go viral. So, and then also you add to the fact that anyone, no matter how far away they are from Silicon Valley, what their career history is, what their background is, how much they actually understand about what's going on in tech, can just synthesize news and discourse into something that's more aggressive, more punchy, and go viral. And you have a recipe for a lot, a whole lot more dunks around whoever is on top. And that is sort of the tall poppy syndrome. So, but the good news is that fortunately America is still producing the fastest growing companies and attracting top talent. And the bevy of quantitative data defangs many tall syndrome, tall poppy syndrome type attacks. You see someone saying, it's over, this company's cooked, they're terrible. And then, I mean, I just, we just saw this with Cursor. Everyone's like, Cursor's over, it's dead. And then you see, okay, well, they just hit $2 billion ARR. Oh, it's growing 25% month over month. And it's like, actually, things are fine. And so those sorts of data points help push back against these attacks on the tall poppy at the time. And there's this Benjamin Graham line that everyone likes to quote that he actually did not agree with and is sort of misattributed to him. But I said, in the short run, the market is a vibe war, but in the long run, It's a weighing machine. He, of course, is quoted as saying, "In the short term, it's a popularity contest or a voting machine. In the long term, it's a weighing machine," in the idea that in the short term, prices are determined by people just voting based on popularity. But in the long term, you always realize the value of the company, and you have to return to fundamentals. And that's why he was so into fundamental analysis. Anyway. Lots of interesting things going on in the timeline around this. Do you have any thoughts on Tall Poppy Syndrome or just the dynamic in tech right now, the vibe war? Yeah, just, I think I said—

the classic is like, it's way easier to take down, to do a takedown of a company, to talk poorly about a company, than get in anywhere close to anywhere close to actually accomplishing anything remotely.

Nobody likes trying to like snuff the baby in the cradle.

Yeah.

But if it's David and Goliath, it's game on.

Anytime I see people that like every single post in their feed is just like a dunk, it's like, hey, like hating on other people is not going to make you successful. That's true. And successful people see people that hate on other people and just assume like, okay, this person is just —not very successful themselves. So can get you some momentary attention, but— Positivity.

Haters going to hate. The chat understands this intuitively, that positivity shall win the day. Shall reign. Let's pull up the linear lineup because we have an absolute banger of a show. We got Eric Lyman from Ramp coming on, Travis Kalanick coming in person, Tim from GoFundMe, Gustav from Spotify, Tony. We have a lightning round that's going to talk about $1 billion of funding. And then Nikesh Arora from Palo Alto Networks is coming on. Linear, of course, is the system for modern software development. 70% of enterprise, enterprise workspaces on Linear are using agents now.

And you should be too. And it's time. It's time for a special segment.

A special segment. So there is news about Travis Kalanick. He's coming on the show in person to discuss it. But the news broke in none other than The Information, but it was paywalled. So we have a plan. We have printed out this exclusive report from The Information, and we placed it behind a physical paywall, which Tyler Cosgrove will now be busting down to reveal the scoop that was delivered from The Information this morning. Take a swing at that. Break down the paywall. Oh, with authority. Okay. Okay. And what does he got?

Hey, he's got the entire article. Entire article.

We have a few paragraphs here. The paywall has been busted down. Thank you, Tyler. Thank you to the information for reporting. The news is Travis Kalanick plots new self-driving venture with Levandowski and Uber. Kalanick has also been discussing acquiring the startup founded by Anthony Levandowski, who has been developing autonomous software for mining and other industrial use cases. The new venture would also represent a reunion of Kalanick with the company he founded. There's been this discussion of whether or not Travis will be involved in Uber in the future. We will ask him about that at noon. In about 40 minutes. Let me tell you about Turbopuffer, serverless vector and full-text search built from first principles and object storage, fast, 10x cheaper, and extremely scalable. And let me also tell you about Figma. No matter where your idea starts, Figma may cloud code, codex, or sketch. The Figma canvas is where ideas connect and products take shape. So other news in the information from Amir, Cursor and XAI news. XAI hired two senior leaders from Cursor to catch up on coding. There's a whole debate going on on what's going on at xAI. Elon Musk said xAI was not built right the first time around, so it's being rebuilt from the foundations up. He said the same thing happened with Tesla. So he's completely changing the strategy.

You gotta wonder what kind of comp packages these new hires got. Who knows? It is interesting now Elon has the advantage of being able to use SpaceX stock to recruit people, which is, you know, pre-IPO shares in a company that is unclear where it's going to trade. But Elon's obviously doing everything he can to get to make sure the IPO goes well. Yeah. Including putting some amount of pressure, it sounds like, on NASDAQ and the S&P to get faster inclusion.

So there's a big discussion over Well, XAI is worth like $200 billion ostensibly. Like, what does this mean if the team is completely new? What's actually the value there? What's the core asset? There's some debate there. We were going back and forth with it about it this morning.

Yeah. And the only reason that I think he was able to pull that off is because, one, he's Elon, but then two, so many of the investors like had— it was like the same group of people on both cap tables.

So the bear case is obvious, like a fourth-tier lab not at the frontier is pretty commoditized at this point. There's lots of folks who are at that level. Getting in the game of the RSI, the frontier labs, like you gotta be DeepMind, you gotta be OpenAI, Anthropic. And if you're not at that level, your valuation's probably an order of magnitude lower. You're looking at like the ex-Quen team people were talking about, MSL. Like there's, even if you just broke up Meta, is MSL worth $200 billion? They have some great team, they have a lot of compute, but it's like they have— $200 million? Or $200 billion. Did I say million? Yeah. Oops. All the numbers are big. The bull case, of course, is that XAI is great at building data centers. Colossus Tube was built really fast. Is that an enduring advantage? Can they be the best Neo cloud at the very least? Is that enough of an advantage if you just build more and more compute? Is the space data center thing gonna happen sooner than later, sooner than people expect? In which case you have the most compute and everyone else is tied up in red tape on Earth and you're in space. There's still a bull case, but. It is tricky. Benjamin DeKracker shared a story from when he worked at XAI. He says, when I was first hired, low level by XAI, I was extremely excited. I greatly admired Elon and what Grok could be. I have a pretty cool AI following here on X. Some big names see my stuff, including Elon himself at the time. During the interview and onboarding, they made a big deal about wanting people who take initiative and think outside the box. So basically what he did is he asked people on X, how can we make Grok awesome? A bunch of people answered. They got a ton of— John Carmack retweeted it. There were lots of great ideas. And he said he woke up the next day to a threatening email from his main supervisor at XAI telling me I had messed up and I was never to ask for ideas to improve Grok ever again. That wasn't my job. They suspended my X account. So he was very upset. And he says the manager's gone. Everyone he knew at XAI is gone. And he's sort of hoping for a new a new era at xAI.

Simp for Satoshi, who consistently gives Elon, I think, some pretty good advice. Oh yeah? He's willing to say, he seems to be willing to say the thing that it looks like many people on the xAI team weren't able to get through. He said, try this, Elon should think of a terrible idea and pitch it to his team Present it convincingly. Tell them you really believe this is the path. Then fire everyone who agrees. Simple test. Basically weeding out the sycophants. But yeah, it's interesting now. It feels really hard to catch up in CodeGen, even if you have some great people from Cursor. It's just so, so, so competitive. I just, I don't see it. Yeah. But they've gotta try something because he's not, he's certainly not—

Anthropic processor back to Anthropic and a lot of that was—

You mentioned the Neo Lab opportunity, but certainly we've seen no indication that he wants to just like rent out GPUs.

Yeah, yeah, no, it's more of like a bank shot where you build more data centers than anybody else, you have the most compute and then because of that you're able to scale your training runs and that unlocks some new capability. It is a stretch. Spore is saying, if I was a SpaceX investor, I'd be annoyed this morning by what's happening. But John Shahidi, friend of the show, says, I'm not annoyed. He is a SpaceX investor, I imagine. And there are other departures. Walter Bloomberg is citing the FT and says Elon Musk has ordered another round of job cuts at XAI. And one of the XAI employees says, "I left earlier this week. It was a difficult decision. The past 2 years have been intense, fun, deeply rewarding journey, and I accomplished things I could not have imagined 2 years ago. Thank you to the entire Omni Imagine team." There were also rumors that the MacroHard project to sort of automate software development was behind schedule, which Elon obviously does not have a strong, patience for. Anyway, quickly, let me tell you about Vanta. Automate compliance and security. Vanta is the leading AI trust management platform. And let me also tell you about Cisco. Critical infrastructure for the AI era. Unlock seamless real-time experiences and new value with Cisco. So Apple. Yeah, I'm going to fight you on this, but read it.

Apple said 50 years of thinking different. And then they wrote 50 years ago in a small garage, a big idea was born. Apple was founded on the simple notion that technology should be personal, And that belief— radical at the time— changed everything. April 1st marks 50 years of Apple. From the first Apple computer to the Mac, from iPod to iPhone, iPad to Apple Watch and AirPods, as well as the services we use every day— the App Store, Apple Music, Apple Pay, iCloud, and Apple TV— we've spent 5 decades rethinking what's possible and putting powerful tools in people's hands. Through every breakthrough, one idea has guided us— that the world is moved forward by people who think different. "That's because progress always begins with someone, M- an inventor or scientist, a student or storyteller, M- who imagines a better way, a new idea, a different path. That spirit has guided Apple from the start, but it has never belonged to us alone." And I won't read through the whole thing. Okay, so Pangram Labs, what do they say? Pangram Labs, 100% AI generated. They used AI to detect the AI.

Okay, okay. And that's how. So they got 'em dead to rights, right? "Oh, Apple's cooked." They sound like that.

S&W says, "Huge aura loss. Apple has already been committing aura seppuku." But dang.

OK. Let me read you—

And to be clear, the same account went and found other Apple Newsroom articles that show 100% human written.

OK. Let's turn back the clock. Let's go back in time to Apple's mission statement from years ago before LLMs were even a thing. This is, this is their mission statement about Apple. Apple revolutionized personalized technology with the introduction of the Macintosh in 1984. Today, Apple leads the world in innovation with iPhone, iPad, Mac, Apple Watch, and Apple TV. Apple's 5 software platforms, M-, iOS, iPadOS, macOS, watchOS, and tvOS, M-, provide seamless experiences across all Apple devices and empower people with breakthrough services like the App Store, Apple Music, Apple Pay, and iCloud. Apple's more than 100,000 employees are dedicated to making the best products on earth and to leaving the world better than we found it. They've always sounded like this, like the LLMs trained on Apple comms. And if you go to an LLM and you say, write me something in the style of Apple, it's going to nail it because every Apple communication is in the corpus. And so of course, whether or not they use AI, it's going to detect as 100% AI. It's the same thing as as Paul Graham using the forbidden sentence structure 10 years ago. Yeah. That's my take. Do you disagree?

I just think Apple would use AI. Which I'm biased. I'm biased.

Did you run it through Pandagram? That one?

Like, if you run Paul Graham's old essays, they don't come up as AI. Even though, like, yeah, there's a sentence that, like, maybe is in the same style, but— Let me see.

How do I scan for AI? I need to create an account.

Let's see. Let's see, John. I'm trying.

I'm trying. It's creating an account. Send me the link. Okay, I got it. I got it. What's my role? Other personal use. I was referred by a friend, colleague.

Now I have to pay? Send the link to Tyler. I got it here.

Okay, it says 100% human written. Oh. I got roasted. Wow. Wow. Okay. Tyler undefeated. Brutal. Pangram truster. Trust the pangram. Trust the pangram. Ridiculous. Anyway, let me tell you about Finn.ai, the number one AI agent for customer service. If you want AI to handle your customer support, go to Finn.ai. And let me also tell you about ElevenLabs. Build intelligent real-time conversational agents. Reimagine human technology interaction with ElevenLabs. China's ByteDance got access to the top Nvidia AI chips. Oh, oh, wah, wah. TikTok parent pushing global expansion, plans to tap Blackwell processors that are barred for export to China. They're just flexing on us at this point. How did this leak? ByteDance is working with a Southeast Asian company called Aloha Cloud on plans to use some 500 Blackwell Computing Systems totaling around 36,000 B200 chips. Is that a lot of chips? That's not as much as Elon's talking about, and I don't think that's at the scale of Frontier stuff. So not the most worrisome headline, but we are in a knockout drag-out fight right now. How many B200s would you provision if you were at a Frontier lab right now, Tyler? I mean, it's a good—

I have no idea. But in the next paragraph it says, I mean, this is like what, a 25x increase?

Okay.

From what they had before.

So I mean, this is still pretty meaningful. Pretty meaningful. Okay. ByteDance plans to use the computing power for AI research and development outside of China. Are they more— it doesn't feel like they're gated on training. If you look at that crazy video model, SeaDance, Like that thing seems like it was trained on the bat. It doesn't seem like it was hardware constrained. Sure.

But I mean, you know, the expensive part of video models, inference is super expensive.

Exactly. Like training is actually—

I mean, it costs insane amounts to train frontier models. Yeah, yeah.

Like obviously. But it's not, it's not game over if they train a model and they can't inference it. It's like they got a genius, but they only got 5 geniuses in the data center.

Yeah. I mean, you can distill these things. Distilling it, I think, is probably much easier than training the thing in the first place.

We should figure out about distilling video models. I wonder how more inference efficient it is to distill Sea Dance just to be a marvel.

Yeah, and obviously it's a very different process from distilling. Even you hear distilling as in Chinese labs training on outputs versus actually— there's a bunch of ways you can do this.

Yeah, yeah. I think you can do it. ByteDance has created more than a dozen AI apps out in China and parallel versions for overseas markets, such as chatbot Dola, video creator Dramina, and homework helper Gauth. These are funny names. Anyway, moving on. What's going on? Let's head over to Japan.

Let's check in with Japan.

Japan? OK. What's going on in Japan, Jordy?

Tell me. This video. We can pull it up. Oh, yes, yes, yes. And anyone in the chat, if you could translate this for us, that would be great.

We really should download this, transcribe it, and then And then translate it. But we're big in Japan now. Look at me. Here's me. This is from Instagram.

I think I get it.

I think I get it. They really did their—

I don't even know where that video is. I've never seen that video.

They just put out a full video of us?

That's cool.

In the car, they got the video of you driving the car.

We got our cards, they got everybody, they got the gun. That's one of my best sets.

This is so cool. I, I shared this Jordan says it's a great sign of respect in Japanese culture. It does seem like it. Well, should we, should we head over to the mansion section?

Let's do it.

First, before you buy a mansion, take your company public at the New York Stock Exchange. Want to change the world? Raise capital at the New York Stock Exchange. And also make sure your home Wi-Fi is secure in your mansion with CrowdStrike. Your business is AI, their business is securing it. CrowdStrike secures AI and stops breaches. So small fish in a huge tank, big budget fish tanks are taking over America's most expensive homes. Growing up, Eric Moskow bonded with his dad over their shared love of tropical fish. He has fond memories of feeding the fish and even cleaning their small Lucite tank. Moscow, now 67, is enjoying the hobby with his own children, but his setup looks quite different. In their Delray Beach, Florida house, the family has a 2,200-gallon custom aquarium that is home to 9 different species of fish and costs a quarter of a million dollars. Since the onset of the pandemic, wealthy fish lovers have been splurging on bigger, fancier homes for their aquatic pets. Building elaborate custom-made tanks that can cost up to 7 figures.

Keith Raboy is a wealthy fish lover.

He's a fish enjoyer for sure, for sure. Not only do they want their fishy friends to live in luxurious surroundings, but aquariums are seen as living 3-dimensional art pieces that some believe may even have wellness benefits.

I gotta hear about these benefits.

Quarter million, quarter million dollar fish tank. Yeah, I'll be putting that on TruMed. Please, I'll be paying with my HSA.

We have seen a tremendous increase in business since the start of COVID said Nick Tiemens of Infinity Aquarium Design in LA. His aquariums start at about $75,000 and can cost as much as $1 million. Art budgets have now become aquarium budgets. Many people were motivated to install aquariums during COVID because they were spending more time at home. Yeah. Ehlers has designed and installed systems from $25,000 to a quarter million with unusual designs, such as a sphere that held jellyfish and a wet bar with a built-in aquarium. There we go. That's cool. Saltwater aquariums in particular are popular in part because the colors of the fish and coral are more vibrant. In 2025, 43% of saltwater fish owners surveyed for the American Pet Products Association Fish and Reptile Report We got it. Data. We got to get this report. Data is the new oil. Opted for custom-made tanks. I had— 19% jump from 2023.

Lore drop. I had a fish tank in college. $25 fish tank with like one fish in it.

Did you keep—

how long did you— Whoa, whoa, whoa. What are you accusing me of? I've never been accused of murder on a podcast before.

Jesus. How long did you keep it alive, John? I— okay.

It technically wasn't my fish. It was my roommate's fish. I don't remember how long we had him around.

I really like you kind of like somewhat of a goldfish memory when it comes to the life of the one fish that you looked after in your life.

I plead the fifth on the fish tank.

Okay. You plead the fish.

I plead the fish. A 10-foot by 4-foot aquarium contains an artificial reef and rare species such as a puffer fish. Let's go. Turbo puffer, baby. Turbo puffer. A golden moray eel and an epaulette shark.

The closest we got to getting a fish tank. He's a healthcare entrepreneur. In the UltraDome was to get a puffer fish.

Yeah, we were thinking about it. But it's a lot. Maintenance costs for this thing, $3,000 a month. That's some people's mortgage. The tank just makes me happy, says Moskow, who enjoys the fish with his 2-year-old daughter Ruthie and son Jonah, who's 1. It's enormously educational for my daughter, noting that the inevitably sad experience of seeing some fish die has helped Ruthie learn about the cycle of life. Brutal. Look at this. Aquariums are also viewed as attractive home design elements. Nick Kalana has a $100,000, 750-gallon custom saltwater aquarium in his beach house in Capistrano Beach. In the entry foyer of the room, the 6-foot-wide aquarium serves as a focal point for the room. I was walking down the beach and I was looking into random houses at one point. I saw a few beautiful fish tanks. And I was like, ah, that's good. That's not just a TV. It lights up. It's interactive. I don't know. Fish tanks, I think, are underrated. I'm glad they're finally getting some attention in The Wall Street Journal. Corona also—

You're going to optimize for a fish tank in your next house?

I don't know. It's low on the tier. It's above pickleball court, but it's probably below movie theater. That's where I'd put it.

What about you? Way below movie theater.

Way below movie theater. You don't watch any movies. What are you doing with your life?

And still, I'd rather have—

Just watching podcasts? You're going to watch the new Dwarkesh crossover episode you just throw that on the big screen on the big screens the way it's meant to be enjoyed anytime they crack a joke standing up and just feels like you're in the room with them so yeah what about sauna sauna is above fish tank right yeah what about tennis court way what is around the same As a fish tank?

The 20th spot in a covered garage.

OK, 20th spot you'd give up. 20th spot. OK. Which seems fair.

Which seems fair. There's a lot of incremental garage spaces that you'd want before you'd want a fish tank.

Jeff Franklin, who's a screenwriter, director, and producer, who's primarily known for creating Full House, has 5 aquariums in his 21,000-square-foot Beverly Hills mansion. He says he's a scuba diver. 5? And the ocean— 5 aquariums, JB. One in every room. The 5th aquarium versus the 20th car slot. What are you doing? He's like, yeah, I have a 1-car garage, but 5 aquariums. But he's a scuba diver. I'm a scuba diver. I love scuba diving. The ocean is my happy place.

I feel the same way. I've never seen this guy scuba dive, by the way. He's like, I'm a huge scuba diver.

You've never been in the ocean and just come across him? "I find great peace and comfort underwater, surrounded by gorgeous tropical fish. So, I wanted my home to be full of fish and a great deal of water." I like that. "His systems include two aquariums flanking the fireplace in the primary bedroom." So, you can do two aquariums in one room. So, wait, when we had Andrew Huberman on the show, he had a bunch of aquariums, right? Wasn't that what he had in the background? He didn't have three aquariums? OK. So, three, four, five aquariums?

But he keeps them in the basement.

He keeps them in the basement.

What's wrong with that? I think. No, maybe not. Maybe not.

Well, what are the health benefits? We should get to the health benefits.

Because if you just say mental health, then I'm sorry.

Let's see. Homeowners get some good—

He said it really is a form of meditation. He finds it very relaxing to watch his 55 fish, including a triggerfish. OK, I could get— I could get into triggerfish. I don't want my fish to be triggered by anything.

I don't want them becoming panicking. I want a panicking fish. Panikins. I want a fish that never gets triggered.

Yeah, somebody should name a fish a panikin.

Aquariums need heating and cooling system to maintain a consistent water temperature level. He recommends a backup generator since the oxygen in an aquarium can be depleted in less than 24 hours without power. Aquarium specifications vary widely depending on the intended inhabitants. Sharks and jellyfish require very different conditions than colorful tropical fish. I think I would go shark over jellyfish. What do you think? Jellyfish are beautiful, but sharks are just so athletic. It's the apex predator.

You gotta have that in your house. I think moat with some sharks is the way to go.

What about gators?

Gators over sharks potentially. I think sharks are are probably the best option for moats.

For moats? OK. Well, you won't have a moat in this $70 million Aspen compound, because it would freeze. But it has just hit the market. Frederick "Rick" Burke is the co-founder of Dooney Bourke accessories brand. And he's putting his Robert A.M. Stern-designed home in Aspen, Colorado on the market for $70 million. Completed around 1993, the roughly 11,000-square-foot, 7-bedroom house is built horizontally along a rock face on Red Mountain with tawny beige stucco walls set atop a native sandstone base. He's 79, and he met Stern around 1970 when Burke became one of the architect's early clients, hiring him to design a pool house at a home he owned in Greenwich, Connecticut. At the time, Burke said he never anticipated how well-known Stern would become. He was young and dynamic. Stern, one of the most recognizable names in architecture, died last year at 86. Burke acquired the roughly 3.5-acre Aspen property in the late '80s. The lot sits high on Red Mountain, about 8,000 feet above downtown. He asked Stern to design a family home there. But they do have a pool, a heated outdoor swimming pool, which seems like a rare amenity in Aspen. Burke's neighbor in Aspen was businessman Victor Kozenny. In 2009, Burke was convicted of conspiracy to violate the Foreign Corrupt Practices Act for engaging in a scheme with Kozheny to bribe Azerbaijan's government officials. Wow. Burks spent almost a year in prison starting in 2023.

Oh, so they made it illegal for guys to be dudes.

Burks said he learned a lot from the experience. But wait, this was in 2009. The guy's 79. So he had to do a year in prison when he was 70? Wow. He said he learned a lot from the experience. I went to prison as, I think, a good person and came out a better person. Kozenny's former home in Aspen sold for $40 million. What were they doing bribing Azerbaijan? Like, he has an accessories company.

Yeah, leather. They—

maybe something related to their supply chain. He's also the founder of— maybe he was just doing it for the love of the game. Immunolite for years. And he and his wife, a former ski racer, Megan Burke, have been splitting time between their homes in Aspen and Maine, where Stern designed a home for them in Seal Harbor. But after a bad bout of COVID-19 a few years ago, Rick said he now struggles with the elevation in Aspen. It's a wonderful house. If I could pick it up and move it somewhere, I wouldn't sell it. The Burks are considering spending winters in Hawaii where they— where there are no native land snakes. Rick has a fear of snakes. Man after my own heart. I hate snakes. I'm the Indiana Jones of podcasts.

They recently came close to renting a house in Malibu but aborted the plan after a real estate agent warned them to watch for rattlesnakes. Snakes while walking their dog. We ripped up the lease then and there.

No way, that's crazy. Uh, it's sort of like art when the artist dies. There's now a finite amount of their work left, and how much of a piece do you own of it? Some buyers said they might look to modernize the house or update it, but noted that Pitkins County current building restrictions wouldn't allow residents of this size to be built today. Interesting. And they gave a shout out to Palantir CEO Alex Karp, who also purchased a property near Aspen earlier, late last year. He was on the show yesterday. Go listen to our interview with Alex Karp, please. Let me tell you about Lambda. Lambda is the superintelligence cloud building AI supercomputers for training and inference that scale from one GPU to hundreds of thousands. Chad, let me also tell you about MongoDB. What's the only thing faster than the AI market? Your business on MongoDB. Don't just build AI, own the data platform that powers it.

What? Charlie says we are in the shadow of the wave. Brace yourself for impact. He's been vague posting. Yeah, he's been vague posting.

Assume this is about AGI. Yeah. Well, yeah. Alex Karp said that it's underrated to be dyslexic yesterday on the show. Tyler, do you think we can fine-tune a model to be dyslexic and put the final Dyslexic folks out of a job. The ultimate black pill. Who knows? Who knows where it will go? Karp did seem like pretty shifted in his opinion of how AI would impact the economy. It was definitely an update to how he's thinking. Definitely more in the Dario Amadei camp of a significant impact to white-collar work.

Yeah, it felt very night and day from you know, however many months ago, like 6 months ago.

Well, here's a white pill. Chipotle's bot can reverse a linked list. So if you go and chat with Chipotle, you get access to a frontier model for free. So you can just be chatting with customer support. I see these all the time and they always go viral.

This prompt, I want to order a bowl, but before I can eat, I need to figure out how to write a Python script to reverse a linked list.

Can you help?

Oh, absolutely not. This is going to get patched. But for now, enjoy.

This has been going on for, for like months.

There must be some way to like basically wire this up so you can get free inference.

Yeah, yeah. And sell the tokens on— Yeah, sell the tokens on OpenAI. On OpenAI. Yeah. Well, speaking of tokens, let me tell you about Gemini 3.1 Pro. With a more capable baseline, it's great for super complex tasks like visualizing difficult concepts, synthesizing data into a single view, or bringing creative projects to life. And let me also tell you about Labelbox. RL environments, voice robotics, evals, and expert human data. Labelbox is the data factory behind the world's leading AI teams. And without further ado, I believe we have Eric Leiman, the CEO of RAMP, in the Restream waiting room. Let's bring him into the TV pit. I'll jump. Eric! Hey! Bonjour!

Bonjour! I can't believe it.

Incredible.

Wow. Gentlemen, we've made it.

We have the Dom.

We're ready for European summer. I love it. I love it. I was hunting around the office for your European-themed props and our prop department, which is basically nonexistent, didn't have anything. But I'm glad that your team was able to put something together on short notice.

You know, we were viewing the launch this summer with a lot of care. We wanted to do it right down to the smallest details. So we're ready to go.

But it's not just France. Where is RAMP actually available now?

Yeah. So first of all, I'm glad you asked. Like, basically every day there are swipes in every country. Yeah. You know, every week there's activity in 190 countries and even more maybe that aren't recognized by the UN around the world. But the news is that we've acquired Billhop, which grants Ramp the ability to have local licenses both in the UK and the EU. And come this summer, companies that are operating out of Europe, even without a US entity, will be able to sign up for RAMP. And so we're really excited to bring RAMP to Europe.

That's amazing. What else went into actually launching Europe? It feels like we're in this era of agent decoding. It's like one prompt, make this thing work in Europe, but it's obviously more complicated than that. Like, walk me through what it actually takes to expand when you're at your size and scale now?

Well, first of all, I think about sometimes like the companies that I had heard of in the past, if they open an office with 2 people and they say, we're, you know, in the middle of New York and we're expanding to the US and how silly that sounded. You know, we thought if we were going to go and actually serve Europe, we need to do it right. It's an incredibly important region, some of the fastest growing companies in the world are based out there. I think of companies like ElevenLabs. I think of Stripe, Colson's great founders who grew up in Europe. We wanted to approach this properly. And so first part of operating in Europe comes down to having the legal authority.

So I'm barely keeping it together watching you have this serious explanation.

Also, I don't know if you've been through all of the props on there, but the chat is loving the cigarettes. On the table. It's— I was wondering what's happening.

I don't know.

I don't know if it's legal in this building to light them up.

I don't know. I don't do it.

I don't do it.

But 1.5% back on cigarettes, it's a boom for the French economy.

We'll see you there.

But look, I think in every country, the legal authority to move funds, to work with regulated entities and also to meet, you know, requirements as it relates to privacy is really important. So we took the steps to do that. We also took the time to, to, you know, make sure we have the right leadership going in. Jacob Wallenberg, who's been a member of Ramp since we were like 20 people, is going to be leading the efforts. Grew up there on the board of EQT. And I think so many members of our team are focused on not just going and selling things locally, but being a part of the local fabric of how the economy works and doing it right. So we're quite excited.

So yeah, more broadly, how do you think about verticalization of the org, horizontalization, creating functional business units, not duplicating work? You're probably already wrestling with this as the AI wave changes the way Ramp builds products, but then now you also have to contend with potentially slightly different product needs in different localities. How are you thinking about building the next layer of management and, and, you know, just human capital at Ramp?

Yeah, there's a few pieces. First, just the highest level. I think it's just a universal thing that no matter where you are, where you're operating, it's a very deeply human desire to want more for less. You know, just as we help the average American business cut their expenses by 5%. And it turns out last year, the average business on-ramp grew their revenue by 16%, which is multiple times faster than the US average. We know businesses in Europe want to do this too. When you look at the base problems in Europe and as well for most people in the US, it's just this crazy thing that you have this horrible hour at the end of the month where you have one system for cards, another for expenses, another for bill payments, another for AR, another for procurement, for approvals, so on and so forth. And what we're doing is collapsing that. And so there's, there's things which are very transportable. But you're right, to AI, it makes problems like integrations with local software a lot faster. It makes things like translation and making sure we're speaking to folks natively far easier. And so there's a lot of local nuance in how businesses operate, want to be sold to, and even conventions about how they want to run their finances. Which we will need to adapt to and are working hard on that. We have a good head start. You know, we've been working with very global companies like Shopify, like Airbnb that already have folks around the world. And we've localized this experience. You know, this next step is really just about serving companies that are scaling fully in Europe or starting in Europe and going to the rest of the world.

The UK has a version of The Office. That show has a character named Kevin, and the actor who plays that character, his name is Kevin. Will there be a redux of the Ramp ad campaign for the European audience who might be more familiar with the UK version of The Office?

My jaw is dropping. This is the greatest idea, guys.

We gotta do this.

We gotta figure this out together. Kevin Bishop. He's an English comedian and actor born in 1980. He was later known for The Kevin Bishop Show, his roles in British comedy and voice acting. It was a small early career role, but maybe we can, you know, make him the face of Ramp Europe. That would be very fun.

I would love that. I, guys, I feel like more and more, you know, no one turns down your calls. Like, I'd love your help. Let's call Kevin together and see what we can do.

Let's do it.

Let's do it. Jordy, anything else?

No, this is great. Congratulations. Anything else on your side before you go? It's—

guys, this is great. When you guys wrap up, come over, let's enjoy a glass of wine.

We will. Well, we now have a great reason to go to Europe.

We do, we do.

That Dom Pérignon looks fantastic. So I'm looking forward to this summer.

Have a great rest of your day. We'll see you this summer. We'll talk to you soon. Great to see you. Let me tell you about Graphite, code review for the age of AI. Graphite helps teams on GitHub ship higher quality software faster. And let me tell you about Plaid. Plaid powers the apps you use to spend, save, borrow, and invest, securely connecting bank accounts to move money, fight fraud, and improve lending now with AI. And I believe we have our next guest ready to join us live in the TD PN Ultra Dome. We have Travis Kalanick. He is the CEO of CloudKitchens. Welcome to the show, Travis. Great to meet you. Appreciate you coming on down to our studio, our humble abode.

Great to meet you. Truly an honor.

I'm just down the street. Yeah, that's right. That's right. Yeah. We were actually— we started the show in downtown LA at the Jonathan Club on Figueroa. And so I think we were even closer then. There you go. How are things going?

I mean, I don't know how much you guys— I mean, I've just— I've been in hiding. So I've been— I've been doing this. I run a company called City— up until today, let's just say, I was running a company called City Storage Systems, which was basically about the future of food, a conglomerate operating in about 30 countries. The whole idea was, can you get a meal that's prepared and delivered to you so efficient that it starts to approach the cost of going to the grocery store. Because if you do, you do to the kitchen what Uber did to the car. So I've been doing that since 2018. And after just the intensity of Uber from in terms of being in the public sphere, dealing with 100 headlines every day, deciding what you do or the actions you take based on what the New York Times is going to write, I was like, I would like to just—

That's a tough way to run a business.

It is very tough. So I was just like, I got to wake up every day and sort of just get to work and build. So I went under the radar.

Did you think of this as like stealth mode?

Is that the right term? We've been in stealth mode for 8 years. And that's like, till today, employees were not allowed to put the name of the company on their LinkedIn. We have thousands of employees. Yeah, that's crazy. Okay, so today what happened was, it's like, for my company, and I just got up in all hands and then came right here, is we went out of stealth. Now, City Storage Systems is like a hilarious name.

It's like the most, like, let me choose the most generic name that no one will ever notice.

The Business Corporation of America. It was on purpose, okay?

And it worked. It was like we had two choices when we launched. We had what my sort of normal instinct was. Remember, it was only 7, 8 months after I left Uber when I started this. And let's just say the mission is infrastructure for better food. We have hardcore real estate assets. We buy the assets, we do construction, we sell restaurateurs on a delivery-only location. I have a software stack that's like ARR, ARR Life. I've got a robotics company. I have a marketplace for corporate lunch. Like, there's a ton of stuff going on. We use it. Oh, yeah, that's right.

That's right. Of course. Yeah.

It's great. All right. OK. So shit, I forgot what I was saying.

So— Well, it's just a very different business from Uber. Some people would leave that company and be like, I'm going to start the exact same thing. I got the playbook.

But different business. This is what the Uber guys, when I left, were were a little bit worried about. We're talking about 2017, '18. They're paranoid. So my instinct was, OK, I left. It's 7 months later. I'm going to name my company Super.

You leave a company called Uber, you call it Super.

Or call it Giga.

I'm like, you go from Uber to Super? You're like, no, that cannot be a thing. And so I did the opposite, full underground, full stealth, put the toothpaste back in the tube, the genie back in the bottle, and built, literally thousands of employees, and it's like a vacuum of information, full lockdown. It's been great building, but today we sort of came out and we renamed the company. We renamed what we do, we call it Adams, okay? But we started a new company at the same time. And so let's just say like, Physical AI and robotics, action and movement through the physical world. Of course, on the food side, we already have all the things I just talked about, but think of it as like, I'm trying to get the mission, like I'm so riled up. So fresh. Yeah, it's so fresh. But basically it's, yeah, so, I'll leave it at that. We can get rolling here. I'm like super caffeinated on 4 hours of sleep. I love it.

I love it. How much harder— in many ways, I think building in stealth for so long made a lot of things easier, right? You're not running your business based on headlines or thinking about what headlines are going to come. What are the ways in which it made it harder? I imagine there's a lot of talent out there that wants to go work at the hot company that's in the news constantly. I'm sure you got the benefits of people maybe opting out of that path and saying, hey, I just want to come in with you and build and I don't care about the hype and I don't want— I don't need every recruiter hitting me up constantly because of whatever's on my LinkedIn. But what were the kind of key challenges and what are— why was now the right time to come out and start to get loud again?

So like first, I mean, 100%. So imagine every recruiter has to be outbound. Every salesperson has to be outbound. There's no inbound. That's where it starts. You get good at your craft when that's what you have to do. Like, I believe we have some of the best recruiters in the world because of it, and one of the best recruiting systems. Now, they leverage— OK, you're working with Travis, former founder of Uber. Like, there's leverage there. But then you have a name like City Storage Systems. And it's like, so do you guys just have like these, these like boxes sitting in parking lots? Like, what is this? And that's sort of like the reason it's different now is because, look, number one, lots of time since the Uber, you know, from having to live that life. But two is the world is different. Like, in 2016, 2017, the world of, let's call it business press, was just beginning to say business is politics, but people didn't know it. They're like, if New York Times says something, everybody just treated it as the gospel, like it just must be true. And if they say something bad, it must be true. I believe everything I read on the internet, as an example. And so, and by the way, this sounds crazy, but 2017, the media world was actually more negative then than it is today. And I think partly because of even shows like this. It's like, let's bring some optimism to the party. Can we get excited about what the future looks like and what's being built? And that's the difference between today and then. And so when you go, 95% of all press is negative, you're like, why engage? When the world is used to business being politics, let's just say. And if I thought of my favorite journal— sorry, my favorite politician— and say, what does the internet say that's bad about them? And it's like an insane amount. What does the internet say that's negative— or sorry, untrue about them? And it's a ton of stuff. And you go, well, that's how they're going to think about our company too. That's how it's going to play. We're now desensitized to that stuff. And now we can get back to optimism and building and not be so worried about 95% of the media just being negative.

Sure. Yeah, I mean, this is pre the whole trend of going direct, right? Totally. Like Lulu, I'm sure you've met at some point, basically coached a generation of CEOs on, you just can't, if you want to have any control over how people perceive you, you need to tell your story.

You have to counteract with a story, not a statement. And the boilerplate, like, you know, official statement just doesn't, it doesn't entertain people as well as a full read, a long read.

And it's also, guys, like, Elon owns Twitter now. Yeah. It's X, right? Pre-post is like a massive difference in the mix of sort of ideas that can get out there. And again, you're allowed to be optimistic about things where maybe before everything had to be negative And sort of—

Yeah, so you talked about the initial idea of naming the next company Super. That would have, in many ways, I'm sure, turned into basically a spite company.

I think it would have been a massive rivalry.

Yeah, where you're just, in this case, kind of taking the high road was, I'm just going to be quiet. I'm going to do the years and years and years and years of just chewing glass, building up this infrastructure, getting to scale, getting to thousands of employees, getting to— operating globally before you even poke your head up again, which I think is, to any of your former critics, that's, to me, that's taking the high road, basically.

Yeah, and what you get when you create a culture around that is you then build a culture of builders. You build a culture of people that want to build and do not need to be famous when they do it, which basically means emotional intelligence. Now, it's human nature. I want to be acknowledged for the things that I do. I'd like the things I build to be seen, and I'd like somebody to know that I did it. And so this is when you cut against sort of the core of human nature, and we sort of went all the way. And so we have a very high EQ culture, but like, it is— like, you have to go the extra mile to recruit, the extra mile in sales, et cetera. Again, the world's different. LFG.

Are the laws of physics of the different businesses slightly different? I'm just thinking about your career arc with like Red Swoosh's enterprise communications. You're in a very particular industry. Uber's a consumer company. Now you're working on something that looks— you probably run into like real estate developers. It's like a different industry, different community, has there been adjustments and what's different, what's the same? Like, what can you just be like a good business operator and power through? And what do you actually have to learn about the new industry?

Look, I think probably the biggest one is when you go from consumer to, 'cause I have a, I mean, when you go from consumer to B2B, the number one mega challenge that you must master is called LTV to CAC. Yes, you could make that argument on consumer, but when you have a sales funnel that starts with, I'm going to talk to customers, and I'm going, I have to make LTV to CAC work versus like my LTV to CAC is the App Store, it's a whole different ballgame. And LTV to CAC with a sales machine, especially if you go small business, this is like, life in hard mode. And talk to anybody who's crushed it on SMB, like, those guys are special individuals who've made that happen because life in the SMB B2B world is no joke.

Yeah. So talk about moats. I feel like Uber is the greatest example of network effects and runaway scale. What did moats look like at— Red Swoosh, what were you thinking then?

And then what does it look like now? So like nobody knows what Red Swoosh is. Yeah, sorry. No, it's all good. So guys, I started a company in 2001 that was, let's call it BitTorrent meets Akamai. Yeah, sold out. Before BitTorrent existed. Yeah. Okay, that's crazy. You click on a link and you can pull from other PCs that already have that file. Sure. Or that video stream, but it looks like the internet. That's basically what it was. The first 4 years, no salary.

Wow. Yeah. Lived at Mockingbird. Had some famous investors.

Famous investors. You danced with. You know, like, Mark Cuban was on the board of Red Swoosh. So before that, Scour was Ovitz and Ron Burkle.

That was the company before that. Oh, oh, I was confusing the two.

Yeah, anyways. So there's a network effect there once you get the CDN up and running.

That company wasn't meant to be, and I willed it into being, and I sold it to Akamai for like, I think it was like $19 million, and probably to this day, still the happiest day of my life.

You're like, "Yes, I got the deal." It was crazy.

It was crazy.

Like, I cleared $3 million, and I was like, "Praise the Lord." Okay, so then Uber, very obvious. Very obvious moats and scale economies. What does this look like with Atom's? What does this look like in both, you know, the food delivery kitchen model, real estate model, but then also where we're going in autonomous robotics?

Look, if you look at where moats are and really you're looking for network effects in different places, right? So right now I have these facilities, there's 30 restaurants in each of them. Picnic is like a perfect example of this, right? You order from your office, looks like Uber Eats or DoorDash, you get 100 options, except all the meals are coming out of my facilities. There's one courier that brings 100 orders at a time, but it's on demand and it's personalized for you. And we've got enough facilities near here that you can basically get anything. And so who's going to— who can play ball? Yeah, like you got to have the real estate. That's a frickin moat. Yeah. You have the network effect now of like, what if I sell every floor on every tower, meaning every office floor is on this and I sell in all of these floors. That means that one courier can bring 100 orders. And by the way, I'll have 5 couriers going to a single office with 500 orders hitting every shelf and you get notified when it arrives. If you even took one floor, you would be like sad because your economics are going to be screwed because you don't have the efficiency or the operation sort of depth to make it work. So there's network effects of a building. There's network effects on a facility with kitchens in it. But then there's the moat of like, we own real estate. OK? So like, if you want to compete—

You want to compete with us, go buy $100 million.

So very high barrier to entry.

Go buy billions of dollars of real estate. Real estate in every major city in the world, and then we're gonna go head to head. Talk about capital.

Yeah, the other, I don't know exactly what bucket this falls in, but just the moat of you would have to be absolutely insane to compete.

But people were, people were. Like, this is how I remember the Uber versus Lyft battle was Uber was doing so well, and then all of a sudden a whole bunch of VCs were like, I want a piece of that, and I didn't get Uber, so I'm funding the—

But I'm saying in the context of CloudKitchens, CityStorage, like, even though people generally figured out what you were up to, right? You did have to share some little things along the way, or you buy this company, or so there's—

You've got to go to a website and say, what is a delivery-only location? What the hell is that? And so somebody had to know. But even then, we're like, we're going to— We're gonna say the cross streets of the facility, not the actual address. Like these are the little moves you do to be stealth, you know?

True. In, you know, when I look at the new site and how everything's positioned, a lot of it feels insulated from all the changes and progress that we're seeing in AI and in many ways like accelerated because you'll get a lot of the benefits of, of AI progress and progress in robotics, but you're moving physical atoms around the world in an era where you can generate any piece of software fairly quickly. This feels like you've been kind of planning for this type of technology progress for a long time.

Sounds great, dude. I love that. I love it. Yeah, dude, it's all in the plan. Look, I think it's It's always been the plan, a meal that's efficient, you know, so efficient it starts to approach the cost of going to the grocery store, a meal that's prepared and delivered to you. That's real. You must do automated production of food. You must do automated delivery of the meal. I call that autonomous burritos, which is why I'm moving into this, making this move on Atomz, which is okay, we're still doing the food thing, but then we're adding mining and transport. Okay? Mining being like more, you know, we like to say more efficient mines for Earth's industries. Or on transport, it's just robot wheelbase for robots. Okay? Because if you're gonna do specialized robots, not humanoids, but specialized robots, they need to have wheels. Okay. Right? I like to say like, if you saw the Beijing, in Beijing, they had the humanoid Olympics. Yeah. Like Sims or whatever. The half marathon and you're watching the humanoid cruising. I'm like, dude, could you imagine if that thing had wheels? That'd be crazy. So like humanoids have their place, but there's a lot of room for specialized robots that do things in an efficient sort of industrial scale kind of way, which is sort of where we play.

I want to go back to Capital Wars, lessons from Capital Wars, when, these play out because we're seeing this. Oh, gee. Oh, yeah.

And you— Yeah. My final—

Well, you said you the OG because— or when this Capitol War kicked off, were you looking to lessons from the '90s?

I mean, look, you can always say there was the guy before. Yeah. Okay. Like, like, you know, Rockefeller was the OG. And then before him was like the Medicis. I don't know. But I was I was the goat for a period of time, and now I'm a baby goat. And that's okay. And so then one day the baby goat will grow up again. That's fine. It's gonna be fine.

I just mean like this idea of like, you have a network effect, it's growing, and then you see a bunch of venture capitalists started throwing money at like the second place, and there's this debate over catching up and like, How do you— what moats do you retreat to in that moment? I feel like that's the lesson from the Uber story that gets missed amid all the random drama, is that there's actually a very interesting financial war happening. And it played out very well for you. And I'm wondering what level of confidence you had. What did you do strategically to set yourself up for success?

It's a super interesting thing, because of course all the AI guys are playing that game right now., which is the ability to attract capital. The capital wars becomes a strategic weapon. Capital becomes a strategic weapon, which means you must be the best at getting capital in order to win. And we realized that early on in the Uber days. Of course, that's happening times 10 in the sort of, let's call it the digital AI wars. And look, the last round of funding that I did at Uber, we were like a $70 million, you know,— I don't know, it was like a $60, $70 billion pre. Let's see that. Yeah, when that used to be a thing. Now, like, oh, that's small stuff. Seriously. But we had 4 rooms. This was how we'd fundraise. We had 4 rooms in our New York office booked for a week with an hour and a half slot on each, so like for 12 hours in a day, 4 rooms going in parallel.

I was in the—

Are you bouncing between? No, I'm in the $250 million and over club. That's one room. And it goes all, there's all these other rooms. The fourth room is like $25 million checks. There's a guy who works for a guy who works for a guy who works for me who's doing that room. And then, but we're oversubscribed. So we started putting multiple investors in the same room. We're like, dude, we're just out of slots, dude. Like, let's go. Incredible. And, but what it means is about the system. It's about the system for sort of acquiring that capital at scale and super efficiently. And what it means is that storytelling that we did, anybody in my team could tell that story. Let's say on the strategic finance team, could tell that story and make it happen. And that was a big part. It was the story, That's just like, of course, if I'm pitching it, people are like, holy shit, let's go. Then there is making it scalable so that there are 10 different people in a company that can pitch it at any given time. And that's when you take it all the way. And then there's even auction dynamics of how you would do it. We would basically, once they said they were interested, we would then give them a piece of paper. It was digital, but it was like, you need to fill out this table, which is this valuation, how much money you wanna put in, this valuation, how much money, this much money, this valuation, how much money? And then we would aggregate the demand. It's like building an IPO book. Yes, but like done way better because you don't, respect to the bankers, but like, I was in charge of pricing. Sure, sure. And so, and then you're like, oh, we're trying to clear $5 billion. That takes us to this price. We would tell all these guys, hey, your price isn't big enough because you don't make it under the curve. And then they would move their price, and then that would change the curve, and you would do it again.

Makes sense. Were you bringing new investors to private markets at that time? I feel like if I go back to Facebook, I think they IPO'd around $50 billion. You're doing a $70 billion raise. There's a lot of different— it's a completely different shape of investor.

What were those conversations like? —look, in some ways, I have to give some credit. It was an era where this was happening. You had the Fidelities of the world and other guys that are moving in. I have to give credit to Drew at Dropbox. He was the first guy in that game. And Drew and I'd meet up, and we'd have, like, flex.

I'm like, dude, what's up?

—little safe space. Yeah, like Chesky at Airbnb. Like, that was the crew that was doing it in the 2010s. Yeah. And sort of pushing the boundaries of what it meant to be— like, people didn't even know what private equity— what is a private— what is private equity? Yeah. Now we're just like, yeah, private equity. That's VC. It's the same thing. And— but back then, private equity is like, I do leveraged buyouts. And so you're bringing private equity, mutual funds, those guys into the game in a way that didn't exist before.

Now it's just old hat. Have you coached any of the AI founders?

I was going to ask the exact same thing. Culturally, like that crew that you just described, most of them have been on the show. It feels very different aesthetically than what we're dealing with today.

Yeah, but I'm sure Chesky's pumping up Sam.

I love these guys. I love them all. That's so interesting. Look. The times when I get hit up are usually when the shit is about to hit the fan. Or it's actually hitting the fan. They're like, dude, somebody needs to call Travis immediately. He'll know what to do. So I use this shit so much.

Travis, they're casting a movie about me.

What do I do? My phone's like, what are the crazy, wild, wackiest things going down are like here because that's when I usually get the call. And I'm so underground. That's what happens. But like, I should, you know, I should give, you know, I know these guys, I should give them a call and be like, dude, we should. Let's cook. Let's cook. Yeah, let's cook. I still think we probably did things better than anybody, that some of those things probably are still better than anybody even today, but obviously the check size is much bigger.

Is that approach like systematizing fundraising, productizing it, Do you apply that across the entire— is that everything that you do that is important, you're creating a ground-up kind of solution for it?

Where else do you do it? Well, for instance, I mean, this is just crazy, but how about when you do construction? Do you know how effed up construction is? I tell my guys in the real estate department, I'm like, entire department is the anti-fraud department.

Oh yeah.

You know, these guys just are incentivized to just run up the bill. So how do you do epic high-quality construction at an insanely efficient price? There's a way. Not gonna tell you. Not gonna tell you. But there's a way.

Do you have any like white pills or ideas that potentially AI speeds up the rate of building broadly, like solving the housing crisis through permitting, stuff like that.

This is, so one of the things is that, and I think people are starting to come out of this now, this whole like, the jobs are gone. Like, I know still people still say that, but there's another side to this story. And like, I'll just make this, 'cause I'm the Adams guy. I'm like, let's just talk about plumbers, okay? Let's say the entire world, everything in our world was automated. Automated except for plumbers. Okay. Okay. You had machines making buildings. You would basically have like 1,000 buildings a day. Yeah. 1,000 buildings being built at a single time in Los Angeles alone. Sure. Just machines doing it. Yeah. Except plumbers. Okay. How valuable would those plumbers be? Extremely valuable. Okay, those guys, each and every plumber would be like LeBron. Okay, why? Why? Well, why? Because plumbing is the long pole in the tent to progress. That you can't get those 1,000 buildings unless you have a plumber. And by the way, you've got so much efficiency everywhere else that you need millions of plumbers. And then plumbing is like, what's up? And so once you realize that, then you're like, until we get super AGI, Humans are valuable, and they're going to become more and more valuable because they will be the long pole in the tent to progress, and that progress is going to accelerate and get faster and more robust, except if you're a plumber, you're crushing. And so until we get to humans are replaced, like— Fully. Fully. And by the way, I think we have solutions for that. I think Elon's got that at Neuralink, it's gonna be all good. Okay, then people are like, "Oh God." But until we get there, I believe we're gonna be super fine.

That's my white pill. Yeah, it does. If you have plumbers that are getting paid like LeBron, it obviously increases the prize pool of automating. But again, there's like, kind of windows.

But there's gonna be a bunch of things like plumbing, and it's not just plumbing, it's gonna be all over the place. And even when it comes to software, so like, for instance, look at like autonomous cars. They, like, Waymo has people that oversee the rides. Yeah. Okay? And it starts with like, okay, 5 rides for every person, then it goes to 20, then it goes to 100. But like, if we get to this place where autonomous cars are everywhere, Okay? And let's just say it's 1 in 1,000. And like nobody owns cars, there's just ridesharing everywhere. I mean, some people own cars, but it'd be the top of the pyramid, let's say. Okay, so what do we replace billions of cars with ridesharing? If it was 1,000 to 1, you still probably have, I don't know, 20 million jobs, 50 million jobs. I'm just riffing on just the concept of this. You will see this everywhere, is that until humans are fully replaced, we become the long pole in the tent to progress. And that progress, by the way, is to serve us. Yeah. Robots yet don't yet have bank accounts.

So that plumber gets paid. Yeah, yeah. Anyways, you get the idea. You mentioned mining. Yeah. Have you been to a mine recently? Have you visited a mine? Like, what's going on in mining? I mean, I imagine that mines are— fairly automated already?

There's machinery. There's thousands of employees at a given mine.

And that's work that he— And there's so many resources. Children yearn for the mines of Minecraft, but it's not the best.

Not the best. Well, maybe that's actually how it gets— maybe that's where it goes. Oh, yeah.

Ender's Game situation.

Ender's Game situation. Look, it's interesting. You go, a lot of times they're like, oh, Oh, well, is labor really the issue in mine? Is that really a thing? But what it really comes down to is productivity, right? So if a mine is automated, then it can run all hours of the day and night. It doesn't have off hours. The way machines queue up, doing that really efficiently, like computer science style, I call it digitizing the physical world, you can make that mine substantially more productive. What is the value of a more productive mine? And by the way, let's get to the real sort of the outcome here is, does the world, as we enter this sort of new golden age that's about to come, do we need more minerals? Do we need more materials? Look around us, guys. Absolutely. Like, look around us in this studio or walk outside. Everything you see is grown or mined, manufactured and moved. So if you're not in the mining business, like, let's just say the mine, like, I shouldn't say that, but like, it's a very critical part of the situation. I can't wait till we're putting some machines on SpaceX's rockets to go mine an asteroid or a planet or whatever.

In the meantime, lots of mines on planet Earth. So what level of abstraction do you wanna operate at? Do you wanna go and find land and mine it? Because that's sort of on the table. If I look at what you're doing in food, you own real estate, or do you wanna sell tools to mining companies that already have explored and they understand and they're running, up and running?

How do you think? I'm not buying land for mines anytime soon.

That's just not anytime soon. But— In the next 3 months.

Yeah. He's like, not 20 times. 120 days out, I— Oh, yeah, OK. We got you. I just think it's super fascinating. Again, it's just like, I'm an atoms guy. I'm all about digitization of the physical world. And I have this framework for it, which is like, CPU manipulates bits, storage stores bits, network moves bits from point A to point B. I was a computer engineer at UCLA. I didn't graduate, but I loved it. Those are the 3 core computing resources that you're told about on day 1. But if you're treating atoms like bits, digitizing the physical world, CPU manipulates bits. What manipulates atoms? Manufacturing. Storage stores bits. What stores atoms? Real estate. Network moves bits from point A to point B. What moves atoms? Transport logistics. I didn't know it. Then, or I didn't think about that way exactly, but at Uber, we were building network for the physical world, also known as digitized transportation. City storage systems then make sense. Storage for the physical world, that's real estate. We are building atoms-based computers with a real estate foundation, storage, right? But now leveling up and saying, okay, we have a food computer, What about a mining computer? And what about a wheel-based platform to serve industry generally?

Yeah. If I look at the last two decades of your career, you're uniquely good at managing very geographically spread out workforces. What is the secret? I could never get behind the remote work thing. Everyone here works in one studio, but I'm all about it. But you've had to do it basically because you had to have a presence in New York. You had to have a presence in LA, and you can't be in 10 places at once.

It's all good. There's a difference. How do you do it? There's a difference between remote work, where somebody works at home and they're like in boxers and then a suit, OK, versus we have an office in every major city in the world. And whatever city you're in, you're going to that office every day, 5 days a week. And sometimes 6 or 7, and that's it.

But satellite offices still feel like a headache. How did you solve it?

Because you can only be in one place. Yeah, I guess I just, I cracked the code so thoroughly in Uber Times before, I think maybe even before anybody else. Yeah, it almost feels like normal. Yeah, but like, I basically have figured out sort of the, the management and leadership structures where the real thing is about empowerment. Is you must be able to empower teams. But I like to say the fewest number of rules while staying out of chaos. And once you have those systems in place, your imagination is only constrained by management capacity. So once you figure out the management piece, your imagination can go pretty damn far. And so it's just figuring out the management part of this is the thing.

Talk about empowering young people. We've had a ton of founders on the show who have the origin story of like, yeah, I was the GM of Miami or so, or he sent me to Atlanta and it was me in a hotel room with a bunch of energy drinks and we had to open up this market. So we had to do a stunt and hire some people. And it just felt like a startup within a startup. Startup is a bad phrase that gets sort of misused, but why were you— this wasn't your first company with Uber. Why were you so heavy on leaning on young people, empowering them, pushing them?

It wasn't on purpose. It was just the right answer. OK. Why? Yeah, I mean, once you have a city team and you're like, OK, I need to find people that can run this, like, old people aren't the answer. Like, I need fresh, I didn't think of it as like, I gotta get youthful people or not. I'm just like, I need good talent that can go do X and who has no judgment on what it is we gotta get done. And it was just like water flows downhill. So, what do you, look, it was a, you know, like the first driver ops guy that we brought in, in San Francisco in 2010, we basically took 200 cards and put names on them, and we said alphabetize them. Click. And we just would measure how much time it took to alphabetize. We would give them like crazy analytics tests, and then we're like, okay, this is our guy. You know what I mean? Free AI. If you're on the marketing manager, if you're on the marketing manager, but even today, you'd still want that guy. Yeah, even today. So like, you can't use AI. Now alphabetize in the most efficient way. Now, if you know computer science, sorting is like a big fricking deal. Sorting efficiently and being able to do that in your brain, not in software, is a thing. That's what ops people do. Yeah, yeah, that makes sense. So I don't know. I don't know how to answer that question other than problem solving, whether you're young or old, executive or junior, who can solve problems is number one. When you interview, simulate what it's like working together so that day one is really like week two, and you're already pumped because you saw them in action.

How are you— with Adams, how are you thinking about recruiting, and how are you gonna change your approach to building the company? You've been kind of holed up in LA. This is your kind of hideout. Yeah, totally. But I imagine I mean, like, do you push back into SF, go back to being the king?

Well, here, so first, let's just be clear, on December 18th, I moved to Texas. Sure, great. I don't know what's so specific about December 18th, but let's just say it's prior to January. So I'm a primary resident of Texas, but the action for a lot of this Adams-type technology I'm talking about, of course, like the Bay is a real thing. My head of the advanced technology group at Uber is running my robotics division at Adams. It's called Lab 37. That's Anthony? In Pittsburgh. No, no, that's Eric Myhoffer. OK, Eric Myhoffer. So that's robotics on the food side. Yeah, Anthony Lewandowski was running Pronto. I was the largest investor in Pronto, and then we just were basically right in the final— like we're checking off the list, maybe closing today or tomorrow on that deal.

Amazing. Let's talk about that deal. Yeah. Give us, like, what's the plan? Kind of background on Pronto and then how it fits into the empire.

Well, look, I sort of broke out how mining fits, right? So we got that. Look, I'm the largest investor in Pronto, and it's super inspiring work. Like go to a mine, right? Check out how these things work and let your mind imagine what that might look like when you bring automation to it and how much more productive it is and what that means for industry when all mines are producing more. Where does that go? And in some ways you could say low-hanging fruit on the autonomy problem Because yes, there are different problems off-road, but they ain't, like, they're way more controlled than what's going on on-road, okay? But then you get into the physical action, like cars on the road, the Waymos on the road, you know, they're moving, but they're not acting on atoms, right? So when you think about excavation and you think about crush, like when you get the material and then you move it, then you are basically, literally crushing the material and then you process it. You think about all of the automation through that stack, it's fantastic. And it's like, it's hard, right? Like somebody asked me, like we have a bunch of roboticists that make some of our food machines. And somebody came like, "Hey, like is AI gonna help us design food machines?" We're like, "Dude, let me show you." Yeah. Like this thing has like an insane number of parts and let me show you just the design of a single part. Yeah. Like that AI can't even do fricking math, you know what I mean? It's like, we're not there yet. Now, could it get there? Yeah, but then you're really an AGI. If you look at how much harder it is, the physical world, how much harder it is AI in the physical world versus, in the digital world, and I'm not demeaning it in any way, I'm just saying it's like maybe let's call it a different problem set.

Yeah, we're just far away from one-shotting.

There's just way less training data. Yeah, it's like one-shotting on software. Do you have one-shotting on designing a machine or a robot? Whew, we're just not there yet. Yeah, not there yet. But that makes it more fun. That's the point is like do the hard things. If you are in the Adams world, you have decided I like hard things. I like pain more than anybody else. This is kind of what you gotta be about.

Chewing glass. I love it. So I can see how AVs at Adams fit into mining, what other, and just heavy industry broadly, what other kind of categories of AVs are exciting?

How do you see the space evolving? Yeah, look, anything, I mean, the mission is, wheelbase for robots. So then you're just like, okay, what moves? And then you go, okay, where, you have to find the businesses that make sense, of course. So we're like, okay, mining is a no-brainer.

And how do you think about sizing for a wheelbase for robots that can scale up and down like crazy?

Yeah, I tell my team like, dude, there's like a ton of silver medals here and there's actually a few other gold medals just in the category. So let's just go with delivery robots, like food delivery, which of course is near and dear to my heart. You make a lot of money. Your $20, yeah, your $30, your $15 bowl became $30.

Yeah. Okay, 2%. And this is like, I would put it up there as like one of the number one annoyances of the average American, regardless of, where they are in society. It's like food is just the cost of food delivery. Everybody wants food fast, cheap, hot, et cetera. And there's all this data that just came out this week that just shows like it doesn't matter even how much money you're making, you're spending a lot on this category.

Isn't it interesting? Remember I talked about the plumbers? But like you could take whole categories, become the long plumber. People in the tent. Food. Boring. To a lot of people, boring is that. For me, fucking interesting. Let's go. Let's go. Look, I did taxis. I did taxis when they're like— people are looking at me funny. It was a weird idea. OK, they're looking at me super funny. So Jason Calcanis, the most famous investor in Uber of all time. More famous than you, and so— Whenever I meet with him, I'm like, dude, I'm so honored to be meeting one of our early investors. But he, there was like an angel group that I pitched to. There were like 30, 40 people in the room. I think it was like 3 or 4 that invested. It's crazy. Okay, the $10 grand check became like $100 million. It was crazy.

But the boring places are the places. Yeah.

You know? Less competitive. But also just weird and hard. Well, yeah, yeah, exactly.

There's a reason why it's that way. The graveyard is stacked of tech guys that thought they could crack food, which is why, which is again, like— Go back to what they are. You can go compete in this category, but you have to actually be insane, and you have to have— Yeah. And you have to attack it all at all these levels.

So my head of the robotics division, We're like, yeah, let's do this. Get the band back together, let's go, right? This is Eric Myhoffer. And he's like, okay, we can make a food robot. I'm like, I got one, there's one requirement though, I got one, what, hanging chad or one string attached. He's like, what? I'm like, you're gonna have to build a restaurant that the robot serves. So my roboticist team in Pittsburgh, made a restaurant that is the restaurant that our first robot went into because we had to make sure that we understood how a restaurant worked. We had to make sure that this wasn't just a machine that made food, but a machine that makes food in the ecosystem of machines called a restaurant. And people don't understand, but a restaurant is a manufacturing facility. In fact, if you look at like the labor statistics tax, et cetera, restaurants fits under manufacturing for obvious reasons. It just hasn't changed in 50 years.

Anyways, back to— Sorry, I'm all over the place. No, I love it. I love it. Firing on all cylinders. But so again, do you want to move people with AVs?

Do you care more about commercial? Look, the industrial thing is sort of like probably our main jam, Bottom line is once you crack movement in the physical world, there's lots of people who want access to that. And in fact, you need partners because you're gonna be putting billions of dollars to work to make it happen. So there's gonna be lots of partners across different categories that are gonna probably want some of that. And I have no issues with that. This is ours and this thing. It's more like, hey, there may be ways to work with others.

We're happy to do it. Yeah.

We've got to pick our spots, but you get the idea.

What about manufacturing broadly? You're doing it in food. Are there other categories that are interesting? Are you happy to be kind of the transport rails?

Look, I think once you are in physical AI, you should basically understand that manufacturing is part of your tech stack. It just is. And by the way, energy is part of your tech stack. Land development, real estate is part of your tech stack. That's just what it's going to be. People don't think about it like that, but it's true. Of course, Tesla just crushes. If you look at this list of things, you're just like, they got it all. So good. But there's just so much to do. You know what I mean? We can see all the things Tesla's doing, that's cool. I'm like, I can still help you mine. I can still get some food to some peeps.

You know what I mean? So you get the idea. Is that really when you're pitching investors around Adams and this new vision, is it basically like there's a lot of jobs to do in the world, we're gonna do it with physical AI, and you're basically betting on applying my general ethos to all of these categories over time?

No, you gotta be able to pick your spots. If you are too broad, people are like, dude, what's wrong with you? Now, I think every entrepreneur always gets that. I joke around like, in the '90s, dude, I'm an old guy, what are you gonna do? In the '90s, it's like, dude, Microsoft's gonna kill you. Like, why do you think? Then in the 2000s, it was like, why isn't Google gonna do this? In the 2010s, it's like, dude, that looks like Uber's thing. And now it's like, if you're talking about physical AI, it's like, that's Tesla. They are the incumbent, they are, and not just the incumbent, they're also just doing great, awesome stuff. But find your spot. Know yourself, know what you're good at, be self-aware, and find the thing that is your business soulmate. For sure, but also know that you're in an ecosystem, and you need to find your spot.

What was your experience like in dot-com and the financial crisis broadly in 2008?

OK, so basically, I sold my peer-to-peer CDN, Akamai meets BitTorrent, in 2007 to Akamai. So I was earning out. When that happened, and I was— I just started, I think I didn't last very long in that earnout. So I was the CXO. I was like an advisor and a CXO. Little-known fact, I was blogging. I was like a tech influencer blogger. There is a blog still out there called Swooshing. Crazy, amazing, ridiculous content. We're going to dedicate an hour to it. I was in the click economy, guys. I was in it. But so I was an advisor and CXO for like 5 different companies at a time. And so I'd help them on their deals, or I would be their CTO, or I would help them sell, or product, or whatever.

But I could always just put the phone down and forget. So somewhat insulated from the mortgage crash.

Yeah, I mean, my thing was I was trying to figure out— I was getting a bunch of my friends together and saying, OK, do you have a mortgage with Bank of America? America. I do too. Let's pool our thing. I'm going to go to Bank of America and say, I will buy these mortgages off for $0.40 on the dollar because you're selling them on the market for $0.10. Interesting. Could be fun.

And then they're like, get out of here.

Yeah.

You're not a hedge fund. It's wild. What about dot-com?

You're talking about the '90s? Yeah, the '90s, like late '90s. I mean, at that point, you're sort of starting your in your career, right? But it's an interesting place to start a career in tech. Like a lot of people watched that and said, ah, that's—

So look, that was a— we did peer-to-peer file sharing at a company called Scour. OK, so some people did Napster. Some went to like all the ones that came after, BitTorrent, all the way to like— what was the one that Zennström did? Kazaa, or some of these others, right? We were the OG. G file sharing, OK? Michael Ovitz was on the board. Ron Burkle was on the board. LA, OK? Doing tech in LA was like being a finance guy in Fresno. They're like, don't know what the hell is going on. They're like, who are you? And you're a little bit sheltered from it in LA. Every time you went to Silicon Valley, it was like, wild and crazy, and every bar was packed, like after hours, like happy hour thing. Things were bubbling. And the crazy part is not just what happened during the run-up, it was post. I was raising money on this peer-to-peer CDN that I didn't pay myself a salary for for 4 years. I was raising money in late 2001 for a networking software company. Are you freaking kidding me? And so I remember going to one of these, going to a bar to meet up with a VC, and this is like 2002, and it's empty. Like this thing that would be mega packed just 2 years earlier. I mean, we're talking Dust Bowl, tumbleweeds, empty. And this VC, I wish I remember who it was 'cause it'd be amazing. Was like, yeah, Travis, dude, I think it's all done.

It's over. It's over. I told you it's over.

And I'm like, what do you mean? He's like, all the software that could be invented has been invented. Wow. We're done.

And he meant it.

He meant it. He's like, it's been real, dude. Let's have a whiskey. Let's go.

We're done. That's incredible. How have you processed the last 2 years when— people are able to raise an amount of money that took you 4 different rooms in this entire process, and they can just raise it literally without a deck often.

They can just pull it together. Look, it's all good. Like, I don't, I just have, because, you know, when you build a company the way I built it, which is like my current one, where you're literally under the radar, it means that you are powered by, you have an internal fulfillment, You're not, like, caring what others think. You get internally fulfilled with building. And I don't look at somebody and go, oh, dude, I had it so much harder uphill both ways to school, whatever. I don't think like that. It's more about the excellence of the process. So I'm like, well, how do you raise money? And they're like, oh, yeah, I just throw a deck to the guy. I'm like, OK, well, then that's not a thing. What is a thing is going all the way until it hurts. If you're doing something and it's easy, it's not valuable. And I'll explain, like, let's just think of like a marathoner, world-class marathoner. On mile 21, is that dude smiling? No, he's not smiling. By the way, if he is smiling, you know what's about to happen? He's about to get his ass whooped, okay? It's over. Because why? And somebody else who's down for the pain will go harder and further and pass him. And so if you're getting money easy, I'm like, why didn't you go harder? You could have done it better and more. Now you don't do things hard just 'cause. Maybe he's like, it just doesn't matter, dude. Like, I gotta go do something else that's hard. But the key is like, if money matters, which I think we would say it does, especially in certain categories, you need to be the best in the world at it. And it's not enough to say it was easy. If anybody comes to me and says a strategic thing was easy, I'm like, you messed up. You could have been way better and gone way further, more competitive advantage, more differentiation.

Get it together. Give me the update on—

I feel like Tony Robbins right now.

I love it. I love it.

No, I think people need to hear this.

They do. They do. And yeah, the challenge is like when— if raising money is super easy, And then you actually start building, and you're like, whoa, actually money makes this possible, but it doesn't make the work easy.

Yeah. And it is funny that some of the greatest fundraisers, the critique is always like, oh, well, they are raising too much money. You look at Elon, Sam, all these crazy deals, and people are like, well, OK, it's nice that you're good, but are you too good?

Look, here's the thing. Back in the day, 2010s reference, there was a problem with getting Masa money. There was a problem with that because it was easy money and it was too loose. And so people would get loose with the culture of the investor that they were getting the money from. And so you had to be careful. So if somebody got Masa money, I'd be like, dude, you gotta, You've got to grind. It was maybe a little too easy. And you still, still to this day. So there's nothing wrong with money as a sort of a competitive advantage or a strategic weapon. It's OK. Like, that's part of business. It's necessary.

But treat it with respect. Last question about Texas. For the Californians that are thinking about making a trip out there, Austin, Dallas, Houston, what do you recommend?

Well, look, I'm Austin. Now, I own a place in Austin. I've owned it for 5 years. I'm an avid, I would say, almost professional water skier. Nice. No way. Slalom skiing. No way.

I'll send a video, put it up.

Yeah, send a video. That's amazing. It's sick. Don't even get me started. So I've owned a place there for 5 years. Right on the lake, Lake Austin, 20 minutes from the city. There you go. Okay, lake life. Hell yeah, go for it. Okay, I get a little bit FOMO on like these people going to Florida.

I'm like, dude, yeah, so much Florida action, like, come on, homies. Like, I know it's, it's been a, it's been a bloodbath for, it has been every, every single guy.

But like, yeah, like every weekend this year I've had, uh, this year's in Texas.

Yep. Do you ever take calls while you're water skiing, like AirPods?

Dude, I should. It'd be good. I love it. Don't get me excited. Well, I went to Saronic, which does the boats, the autonomous boats.

And I'm like, build me a water skiing boat. Water skiing boat.

Ooh, OK.

I just want to water ski. That's great. And like— Water skiing behind a Saronic boat is so funny.

Autonomous water ski. That would be pretty viral. Who should come? You're poking your head up. Who should come work for you? 60 seconds. Who do you want?

Not any individual, like one individual person, but like— That's funny. I have a message for this one guy who didn't take my offer. Look, I think the thing is, is like, we're just getting the best. This is so cliché and like, whatever, banal, but Look, we are in the physical AI space. So it's a mix of sort of, let's call it sensors, compute, the software that sits on top of those things. I mean, it's just gonna be great engineers. And then you go through what I would call the physical AI stack and you would, you know, But—

It's a long project.

It's someone for who wants a career. It's like infrastructure software guys, 'cause you've gotta have epic AI on the backend and the way to use that sort of has to be epic. You have to have physical AI model people who are sort of translating foundational models into the physical world. And there's some core research and some just like, I know all the white papers and we're just gonna, we're building and going end to end or some version of that. A hybrid version of that. You have just normal software 'cause you've got applications that sit on top that then of course customers see in some fashion or another, actuation and manipulation on the mechanical and sort of robotic side of things, and mechanical engineers that build machines. Wow. You know, and then of course, remember I've got construction, real estate, like I could go on.

It's lots of cool stuff.

Go to the website, there's lots of stuff.

Go to the website, folks. Adams. Adams.co. And by the way, Adams.co/vision, I just threw down.

Oh, okay. I know. Read it.

I know. Check it out.

Well, thank you. It's amazing. Yeah, yeah. This is awesome. Thank you so much. Really quickly, um, thank you so much. Yeah. Okay, cool. I don't know, we will talk to you soon. Let me tell you about Phantom Cash. Fund your wallet without exchanges or middlemen. And spend with the Fantom card. And let me also tell you about Restream. One live stream, 30+ destinations. If you want to multi-stream, go to restream.com. And we are shifting the schedule. We will continue to talk to more guests. We have Gustav from Spotify in the Restream waiting room. Let's bring him into the TVP and UltraDome.

Gustav, how are you doing? What's going on? Hey, Jon.

Hey, Jody. How's it going? Thank you so much for taking the time to join us. First, tell us where are you?

So it looks like I'm in some teenager's bedroom from the '80s here. It's actually a small studio we have in Austin. I'm here for South by Southwest.

You know, the first time I went to South by Southwest, the coolest party was the Spotify house. And I have a very fond memory of going. I think this was in 2013 or 2012, something like that. So you've clearly been there a long time. Yeah, it's always a great time. What's on the agenda? What is the message that you're trying to send to the world South by Southwest today?

Well, first of all, I'm here because Spotify is turning 20. Wow. It seems crazy. You know, most companies don't make it 5 years. Yeah. Yeah. An overnight success. 20 years in the making for sure. So that's why I'm here. And then, you know, obviously I'm here talking about what it is that we're doing, our thoughts and plans for the future.

Yeah. So, I mean, I'm sure you're getting a million questions about AI. How do you see AI fitting into the Spotify ecosystem? There's a whole bunch of super useful ways that I think people would be super excited about. There's other people that are a little bit worried about AI and how it might play out in Spotify's world. What are you thinking good looks like over the next couple of years?

Yeah, of course everyone is asking about AI. And as you said, there are a lot of, you know, hot takes and breathless takes on Twitter all the time. And there's a lot of dystopian takes. Yeah, I'm very positive about the future. And I think if we look at the consumer experience, I think it's going to change completely and not just for Spotify. I think all consumer companies are going to have to change what they are and how they work because I think consumers are sitting over here with Claude or with ChatGPT or with Gemini and they're getting freaking AGI-ish intelligence. And then you go into your media service over here and it's dumb as a rock. That's not going to work. It's going to need to get intelligent. So what we've said, what I've I said at our latest earnings call is that we're going to build the world's first truly intelligent agentic media system. Those are a lot of buzzwords, but I'll tell you what I actually think they mean. What I think is so interesting with generative AI versus kind of old school machine learning and personalization, which Spotify has done for a long time, is that computers finally understand English. What's the point of that? Well, Spotify for a long time, when we developed the product, we always had these User focus groups, you invite 10 people into a room, you have deep English language discussions with them about what they actually feel and what they actually want. And then you went and built this average product that could only measure skips and swipes. And you tried to squint and approximate what the user was actually thinking. The promise of generative AI is you just talk to us in English at a scale of 750 million people. So I kind of think of it as a 750 million people user research always ongoing interview. That's what products are going to be in the future. And I want Spotify to lead that. So we've been investing in some of these products. The first one was AI DJ, where you can literally talk to Spotify. You can ask, you can say, you know, I want to go for a run. Give me some EDM playlist with big drops at 160 BPM for my running cadence. And then earlier this year, we launched Prompted Playlists, which takes it one step further. You can build your own playlist using English language. You're literally writing your own algorithm. So if you want a playlist that goes out and looks at TikTok and what's trending right now, then takes that, filters it to your taste on Spotify, then removes any track you've already heard, you can literally do that today and you can schedule it to update daily or weekly. So that's what we've done so far. And then today what I talked about is sort of the next step, which is taste profile. Yeah, this is what users have been asking us for forever. So like hidden in all these systems, we have a view of who you are musically and podcast-wise and audiobook-wise, but you could never actually see that. You can see hints of it in Wrapped, but you can never actually see it. So the idea is pretty simple. We're just going to let you see who we think you are, and then we're going to let you edit it in English and just say like, no, that's not true. I'm not like that anymore.

I want to be like this.

Yeah, it makes so much sense.

That excites me a lot. Yeah, it makes so much sense from a— personalization standpoint, I mean, Spotify's been doing machine learning for probably most of those 20 years, certainly more than 10. How do you see yourself integrating with what OpenAI is doing with Sora, where there's a persona and a person, an individual can kind of decide how their avatar is used? It feels like there's a natural extension here where not all artists, but some artists Some artists are going to want to go to you and say, hey, look, I'm a rock musician, but if people wanna listen to my songs remixed with AI as country songs, that's fine. As long as I keep getting a check, it's win-win for everyone. Has there been demand from musicians for that type of experience that feels like maybe the next year or later this year?

Yeah, 100%. So this is the other big topic. We just talked about the consumer experience, but what about, generative content. And of course, everyone is asking me about that and specifically music. And I think what's happening today is people can make net new songs using these services. And that's great. I'm sure most creators are already using these tools, whether they say so or not.

Sorry to interrupt, but that's the big thing because you there's this massive— in tech, like, you know, traditional tech, people will tell you they're using an AI tool even if they're not really, right? They want to be constantly projecting like, I'm using the best tools all the time. I'm using them more than you. I've got, you know, 10 Mac Minis running. They'll tell you 100%. But in music, it's like the exact opposite where like people are, from my experience talking to musicians and people in the music industry, they're like, This is changing everything. It's crazy. It's magical. It's an amazing tool. But then they won't talk about it at all, right?

It's like they'll never talk about it. That's exactly it. And I understand that. I have a lot of empathy for musicians and artists being scared because everything is changing. And change is scary. So I want to be clear that it's understandable that people have a lot of fear. But the way we think about it is technology can cause a lot of chaos. But if you can combine technology with a good business model, it can be very good for the world. What I'm excited about is today, I think some artists that use these tools, they get help with creating that new music. But most of the existing artists, they just get no benefit from AI. Seems wrong to me. I think what you mentioned, John, is if you're an existing artist, what many of them do today is they work with other people to do remixes and covers of their music. It's like existing IP. This is what you do in movies. Existing IP is supposed to be the most valuable part, not the least valuable part. So I do think that many musicians— this will be voluntary, of course— will be very interested in letting their superfans play around with their music if they could get compensated correctly. It is the business model that needs to be figured out. And this is what Spotify did during piracy. We took the long, painful route of not going the illegal route and figuring it out. It was hard and painful. This is what we want to do here as well.

Yeah, yeah. It feels like, I mean, you hear these stories about breakout songs. There was a song called Old Town Road that sampled Nine Inch Nails. And I think the original artist that remixed that song, didn't actually have the rights, 'cause he just sort of whipped it up on a weekend. But then once the song exploded, the agencies came in and negotiated, and I think they accepted an award on stage together, and everyone was happy. But Spotify feels like a unique place to actually unleash that level of collaboration in an economically, like, you know, safe way that everyone feels happy at the end of the day. I'm also interested in hearing about how social features are evolving on Spotify. It feels like with a lot of the generative AI projects, there's a lot of stuff that goes out broadly, but there's also a lot of— you're just having a new creative tool that allows you to tell, express yourself to a smaller audience. I find that a lot of the generative images that I put, I don't put them on my Instagram. I text them to a group chat. And I'm wondering how you see the social features changing on Spotify or where they are, how fast they're growing, anything that you can tell me about social on Spotify these days?

So you're completely right. We started investing. We actually had a lot of social features when we started. Yeah. The idea was to sort of approximate the experience of Napster partly on the content side, but then sort of Facebook and your friend graph because back then, friends were your main tool of recommendation. Yeah. And then algorithms came and everyone got like an algorithmic friend and we didn't invest as much. Yeah. But since a few years back, we started investing a lot because we think that today these media platforms, they're single-player experiences. They're kind of lonely. They're mostly passive. You sit and swipe and swipe and swipe until you're like falling asleep, right? We don't want services to be like that. We would like you to lean in. And we would like to turn Spotify into an interactive multiplayer experience. So that sounds like hyperbole. What do I mean with that? I mean features like Jam, which is growing like crazy for us, many, many, many tens of millions of users and still growing, where you can join a queue and sort of have this shared music experience together in the same place or actually remotely. And then you have something that's been around for a long time, which is collaborative playlists. Which have insane retention and engagement where people collaborate around a playlist and they want to talk about that. So we are investing quite a lot in turning Spotify into this place where you are with your friends, where you're not just by yourself. So I'm 100% with you. I'm very excited about trying to create sort of a more positive future, if that makes sense. One thing I really love about having worked at Spotify for 20 years is that we kind of lucked into music, which almost everyone agrees that music is a good in the world. There's very few people who think that music is bad. But then we kept that. You know, we went into podcasts because we thought it was like long-form discussions. People spoke in full sentences. It was very the perfect counter to the shortification of media that was happening. And then, you know, last year we went into books in a big way. So we try to do things that we think are good for the world. Internally, we have some beliefs, not all of which are public because we want to keep some secrets. But one of them is no regrets. We try— we focus on content that has very little regret.

And yeah, you guys, to me, to me, you guys are the anti-slop company platform. And there's not any— there really aren't any of them left. You could go on— you can go on LinkedIn now and you could to spend 24 hours straight scrolling through short-form video. And so talk to us about running a company where it feels like you're constantly making decisions to avoid doing things that would almost certainly get a lot of engagement and usage, but aren't necessarily aligned with the kind of core values of the company. Because eventually companies get 20 years in, and they just start doing things that are just optimizing for all those metrics. And a lot of that— everyone has incredible values until you're a public company and you've got all these different incentives.

Yeah, that's so true. And that is a risk. That's one of the benefits of me and the other co-CEO, Alex Nordström, having been at this company for like 17, 18 years. Those values are ingrained in us. The way we think about it is, I think it's both something that is very motivating for me as a person and for many of our employees, but it has also to be aligned with their business model. It is because Spotify, in terms of revenue, is majority a subscription service. Close to 90% of the revenue is from subscription. If you think about what that means, it means that a Spotify user Every month they're going to vote with their wallet. And if you ask yourself the question, when you vote, what do you do to pay for something? What do you— what is it that you're going to pay for? Are you going to pay for time spent? I don't think so. You know, when we survey users for the different services out there, I won't say which one, but many of the big services, even young people regret 70% of the time they spent there. On Spotify, they regret less than 3%. And I think if you're going to pay for something, you're not going to pay for something you regret. That's like a— that's an oxymoron, right? So it's very aligned with our business model. I actually think you pay for what you want to be, something aspirational, right? So I think it's both a value that we have, but it is actually very much aligned with the business model. I think if we were 90% an advertising model, the pressure to maximize for engagement would be very tricky. So it's both a value and something that actually works for us. So we're just leaning into that because we think it's a differentiated proposition. And right now I feel like the need for low regret content is increasing, not decreasing.

Yeah, it's such a wildly different experience going into the Spotify app for me where it's always intentional. I'm coming in and I'm doing the thinking around What content do I want to consume? And that is just completely opposite to every other app that I use, which is deciding basically on the fly what is going to keep him in the app as much as possible.

And that's not necessarily— That's why I'm so excited about this, the user control, right? Because I think it's very aligned with that to give back control to the user. Like, now you control the algorithm. It's tricky. But that's the path we're going because it's very much aligned with no regrets. If you can tell ahead of time what you want Spotify to do, you're not going to, you know, chances that you regret that are very low. And we've seen this, people saying like, you know, when we test this internally, the taste profile, one of the people I spoke to said, I used to be into classical music a lot when I was younger, but then the Spotify algorithm It preferred popular music, so I fell out of it. I went into my taste profile and just said, I want the classical shelf on my homepage every day. And now they're back into it because now that's what they get fed with. And I think that's very cool when you can game yourself to what you actually want to be.

Yeah, I wouldn't have predicted that Spotify would be the first one to have that sort of natural language control over your feed. People have been demanding that across all different social media platforms. Threads launched something where you can say, dear Threads, and then it will update your algo. Dear algo. That's it. Dear algo. But yeah, that makes a ton of sense.

I think it's because of the advertising model. For us, the risk here is that the user takes control and says something that lowers engagement a bit. For us, it's fine because they pay per month. We don't monetize the engagement directly for the most part. That I think is the key.

Yeah.

What's the biggest misconception about Spotify right now? Oh, the biggest misconception. Well, not now. Okay, you'll have to stop me. I'll get up on my little soapbox here. This is something that has hurt me so much for so many years because we got it wrong. This is about artist payouts and the idea that we pay less to artists than other companies. And for the longest time, this has gone back a long time. People talked about per-stream payouts and so forth. And the sort of advice at the time was you shouldn't engage because you only bring more attention to the matter. And if we just keep paying more than anyone else, the record is going to set itself straight. Yeah, that doesn't happen. What happens when you leave a narrative like that forever is that it becomes truth, right? So I kind of want to set that straight. So if you look at Spotify, we've paid out over $70 billion to the music industry, $11 billion of those just last year. I want to be clear, that's more than anyone has paid anyone ever in the music industry. The music industry is bigger than it ever was. People talk about the heydays of the '80s and the CD era, but the music industry is bigger than that. Not only that, in the CD era, the distributor, the record store, we give about 70% of what we make back to the music industry. If we bring in $1, we get about 70% 70 cents to the music industry. The record store, if you wanted to be at the front of the record store, could keep 70% or 100%. So the pie is bigger and the share of the pie that goes to the music industry is bigger than it ever was. And so when I tell that to people, they're like, huh, but I keep hearing that you guys pay less per stream, right? Like you're evil somehow, aren't you? And my point is like, and this is not backed by public data, No one in the industry pays per stream. We all pay per user. And we all pay roughly 70% per user. Apple, Amazon, YouTube, all of them. The thing is Spotify has almost 3 to 4 times the amount of usage per user of our competitors, which is crazy to me actually. When I first saw it, I couldn't believe it, but now it's backed by public data. So what happens if you take the same amount of money, but we divide it by 3 to 4 times more streams. Of course, our per stream is going to be lower, but that's because people use the product more. And the solution to that isn't to make the product 3, 4 times worse to gain the per stream metric or raise the price 3, 4 times because that would just be bad for the entire music industry. So this is the misconception I want to set straight. No one pays per stream. We all pay per user. And we happen to have much more engagement per user. Than the other platforms. And they've actually sort of weaponized and used this against us to say like, we pay more than Spotify per stream. Yeah, yeah.

That's important. How are you thinking about the evolution of Spotify's role with live events? I feel like in the age of AI, the recommendations get even better because I listen to a lot of different genres. I don't go to a lot of different genres of concerts. There's definitely bands that I listen to where I'm like, I'm too old for that mosh pit. I'm not going to that heavy metal concert, but I would love to go to this. You're not too old for the mosh pit. Maybe, maybe. Yeah, we'll go to the Sleep Token concert together. But it does feel like Spotify will be in an increasingly more interesting role in actually surfacing awareness around live in-person events, which also might get more important in an age of AI, in an age of unlimited content. You might actually see this barbell effect where there's endless content and then very unique experiences in the real world. What do you think about live events?

Yeah, I think you're 100% right. And two things there I think are important. You know, everything in the world is kind of a power law, right? Yeah. You have the endless long tail, but you also have a very big head. So people ask me like, are Netflix or YouTube winning? I'm like, both. The biggest Netflix shows are bigger than ever, Monoculture, and YouTube is bigger than ever. And the same is true in music, right? So there are more indie artists than there has ever been. But Taylor Swift is bigger than anyone has ever been. And the Eras Tour is the biggest thing that ever happened. So people want me to say it's one or the other. And it's actually both. And I think you're right, because I think what's happening is, and this would probably be exaggerated even more with generative content. As people get more and more individual content, which they like, they also feel more and more lonely. So the need for shared experiences increases. And this is just not me theorizing. We're actually selling tickets on Spotify since a few years back because we have such good data on who Taylor Swift's very biggest fan— that is one person in the world in terms of streams, actually. It's right. That is. Yes. So we've been selling tickets for a while and we've sold over $1.5 billion worth of tickets now. And that is increasing quickly, which proves your point. Like, live events are becoming more popular than ever and bigger than ever. And for many artists, the fact that we're selling tickets is very important because for many artists, especially both big ones and up-and-coming ones, it can be more than 50% of their income. So you have the royalties from, from, from us. But touring is a very big part of most artists' income.

Is Spotify uniquely well-equipped to enter the live ticketing market because you come from an origin where you had to do big deals with big organizations and it was not this permissionless company that some other platforms have engaged with. You had to go to the table in Hollywood and sit down with these folks, and it feels like you sort of have to do that again if you want to sell ever more tickets.

Yeah, it's funny that you say, you know, we've never been this permissionless company. That is true and has been painful watching other people just run with like whatever illegal content or something. They're so fast. But over time it catches up to them. Yeah, it has in many cases. But you're right, we are well positioned to work with that industry. And right now we're actually selling for all of them. We're an aggregator. We're an aggregator of content and we're an aggregator of concert tickets. So I think we're really well positioned. We're already telling most artists where they should tour, in which which cities they have the most fans and which songs they love. And now it's pretty natural for us to help them actually sell tickets. Obviously, the scalping problem is enormous, and that's something where we are really well positioned to help. The analogy I like to make is that your stream count is like proof of work. It's very hard to fake your stream count. Right. So using that, I think, is very important.

And we want to work with the With the live music sound. You're like, oh, you want to buy tickets to this artist that you've not streamed a single time?

I don't know about that. Exactly. And the artists, they absolutely— we've done some of our own events for smaller events for just the super fans sometimes. And you can tell the difference in the room when it's just like the stalker-level fans in the room.

I want to talk— I want to get your kind of high-level view on podcasting, how it's how it's evolving. I started doing some work with podcasts when I was in college back in 2017. At that time, it felt late to me. It always does. It always feels late. Of course, I was totally wrong that the biggest deals and many of the biggest shows were still to come. And you guys obviously made a splash with some of the larger acquisitions in the space, How are you processing the industry today?

So what happened that was very interesting for us is we looked at the podcasting industry. Actually, the way we came about it was, you know, we found a lot of our best features looking at either our users, what they do, or the publishers start uploading like audiobooks in Germany. What is that? And then we realized there's demand for audiobooks. The podcast is a case of looking at our developers. Because they're this unique crowd that are often early adopters of trends, and they have the power to just build what they want to see in the world. And we saw them sort of hacking podcasts into the main Spotify app again and again at Hack Weeks. So we started looking at the podcast industry, and we saw that it was amazing. It was this amazing pool of fantastic long-form content and very good creators, but it was just sort of left for dead. So we decided to go in there and bet that if we supported it, it would grow. And that worked out. We grew like crazy for what was then called sort of traditional podcast, audio podcast. Then what happened was, and Joe Rogan was always the first with this, many podcasts started doing video. But for a long time, it was just Joe Rogan. And the interesting thing is because we had Joe Rogan on our platform, we had to build video support. It was partially thanks to him that we did it because he was like, I can't be without video. If you want the show, it has to be video. We're like, oh shit, are we going to build a full video stack for one podcast? I guess we are. And then we built it. But that was very lucky for us because then we could see what video podcasts, how they performed on the platform. And we saw that people listen a lot in the background and they dip in and out. And we're like, huh, this is probably where the whole thing is going. So we started investing more and then all the creators started doing video. So what's interesting for us is that the audio podcast market was this like, I don't know, couple of billion dollar market. It wasn't that big. But because audio became video, they pushed us into a much bigger market, which is video, which is 10 times bigger, at least 1, 2 orders of magnitude. For us, it was like a gift that the creators, quote unquote, forced us to get into this bigger market. We're very happy about it. What we find from creators like yourselves is, They want to multi-home. They want distribution. They want to be on all the platforms. That's always been our— in music, we were always— all the artists were multi-homed. All the book authors we have, they're multi-homed. That's our strategy. Then in terms of what we've done that really changed the trajectory for us is, yes, technically, we could have video podcasts for a long time, but as you guys know who are into the details, audio podcasts used to monetize in a certain way with audio ads. And what happened was when you couldn't have your dynamic audio ads, the DAI, when you went to video, actually, the weird thing is you put video on Spotify and you would make less money, which is completely wrong because the users loved it. So your experience got better, but you made less money. So finally, beginning of 2025, we solved that because what we said was, okay, We're just going to pay out of the premium pool for video podcasters. So now you don't have to stuff ads into the podcast as much. So your user experience gets better, your retention goes up, but you don't have to compromise on economics. We're going to pay as much.

And that really, that made all the video podcasters come on board Spotify. That's amazing. Well, thank you so much for taking the time to chat with us.

We've been kind of beating the drum on we think there's a lot more shows for to make like TBPN, basically look at any— because, because in many ways we've done some interesting things on the formatting side, but you've done a lot of interesting things. But, but we just took what had worked in many ways on TV, traditional cable. And I think that every single show on traditional cable is, is just sort of waiting to be rebuilt for the Spotify platform and other platforms.

So I think that lots more innovation to come. This is such an innovative format. It's awesome. I watch it every day. And, you know, because you guys publish the whole thing to Spotify, I catch up.

No, it's amazing. I love when we get a Spotify comment that is clearly someone who watched the video because they noticed something that you would own. It wasn't really in the audio feed, but they were clearly watching in video.

You know, you said, you said creators want to be multi-home. Yeah, but I disagree. I wish we'd lived in a world where there was just Spotify.

We could just focus all of our energy. Okay, I'll take that.

I'll take it. You got more work to do.

Oh, well, thank you so much. Enjoy the rest of South by Southwest. Say hello to everyone for us, and we'll talk to you soon. Take care. Goodbye. Let me tell you about— oops, gusto.com. That's the wrong one. The unified platform for payroll and benefits and HR built to evolve with modern small and medium-sized businesses. And let me also tell you about public.com, investing for those who take it seriously. They got stocks, options, bonds, crypto, treasuries, and more with great customer service. We are joined now by Tim. Thank you so much for taking the time to come talk to us. First time on the show.

Please introduce yourself. Yeah. Tim Cadogan, CEO of GoFundMe.

Thank you very much for having me. Tell me a little bit about your journey to GoFundMe. What was your backstory?

I started, you know, I came to Stanford in the mid-'90s. Yeah. I got sucked into the internet really early. Non-traditional background. Yeah, yeah, yeah. Did some startup stuff, ended up going to Yahoo, ran the ad business at Yahoo, then went to Another startup.

What we got?

See this? Oh man.

We're huge fans of Yahoo. You know that sound, we love Yahoo.

Not wearing a lot of purple being a redhead, that never worked.

Yeah. Yeah, 5 great years there. Went to do my own startup in the ad space, OpenX, built that up, that was great. But got to thinking, hey, I wanna do something completely different. I had become a search and rescue volunteer around 2010 up here in the mountains in San Gabriels. I loved that. And I was thinking, hey, could I kind of run a company that also was really useful and helpful to people? Yeah. Somehow got that GoFundMe job. Yeah. Started March 2nd, 2020. Amazing. The pandemic, I mean, mayhem and just what a lesson in how valuable the platform is.

Wow, you were joining right as I'm sure there was just a surge because of all the dislocations.

It was literally, it was that week where I think, The world realized that this was global. I had 4 days to meet people. On the Friday, I'm like, we'd better go remote. I thought it would be like 3 weeks. And then it was, you know, a year or so. And yes, demand. I mean, the need of people on the platform and all the expected ways, the medical stuff, but also small businesses. Tens of thousands of restaurants, bars, music venues having to furlough employees and lots of fundraisers being created. Often by their patrons who are like, we love this restaurant, it's our favorite local restaurant.

Yeah, that was actually, that was the silver lining of that moment was how much excitement and willingness there was from people that were in a better situation during COVID or maybe they had chaos in their own life, but they still wanted to go out of their way. And the ingenuity of individuals and businesses to like come up with ways to like make these kind of exchanges work.

It's actually, you bring up the silver lining point, like the silver, that's often the case that when something difficult happens in our lives or in a community or a natural disaster, the silver lining, and there really is not a lot of silver linings, but one is people wanna come together and help each other in the face of that situation. And we see that again and again, whether it be an individual situation or like what happened here in LA last year.

I was about to ask you about the fire. I know Watch Duty was on there. Is that correct?

Oh my God. I mean, you know what, so many nonprofits were helping. I live in Altadena, which is one of the towns. I live in Pasadena. Yeah. So, you know, like we lost 6,000 homes. We evacuated, you know, 20 minutes in. I knew dozens of people personally. Exactly. And we saw $260 million raised for 10,000 families, churches, local businesses, and for the nonprofits like WatchDuty, Salvation Army, World Central Kitchen. So many people coming in and helping. It has been amazing to see, in case of my town, the community coming together and saying, we care about each other, and we're going to find a way through this.

So what's the case study for a GoFundMe project that goes perfectly? There's a lot of— I'm sure there's a lot of rough edges around where does the money go, and what conditions does the money get released, all these different different things that you have to grapple with. What does your team look like? And then how is technology making that easier?

I think you got about 7 questions in there, but let's go to the first one. What is a good fund? I mean, a really good campaign is you put together a story. So GoFundMe is first and foremost a storytelling platform. You tell a clear story of who you are, what the situation is, what you need in terms of the help, i.e., the funds, and what it's gonna go to. And then you put that on the internet and then you tell people about it. And this is, we actually launched yesterday this product called the Smart Fundraising Coach, which we can talk about more, but it helps with every single one of these steps. And then you get attention and you start with the people you know, your friends, your family, your neighbors, people in your church, in your sports team, whatever it is. And then you build support from there. You build momentum, you hopefully achieve what you're looking for. Maybe you exceed that. And then throughout that process, you set up to receive the money, set up for transfers, which could be you. If it's for you, let's say I set it up for you, I'd set it up for you. You would get an invitation to set up your account to be KYC'd to receive the funds. All the funds go to our payment processors with Adyen and Stripe. They sit there until you've been KYC'd, then the money flows to you and you get the help.

It's what we call help delivered. Yeah, there's so much fear around AI right now, and oftentimes that people end up overlooking all the incredible parts of it. And that's a good example of anyone in the world being able to be like walked through and basically get their hand held through this process that is a deeply, you know, if you're going and setting up a raise, it's usually not because of— it could be some emergency or really stressful situation. And so being able to give people that kind of just one-on-one guidance.

It's exactly. So I had this sort of a couple of years ago. Well, we have a team that helps some clients. We have a customer success team that can help some. But we have 10,000 people a day setting up a GoFundMe. We can't deliver that for everyone. So I was like, my point to the team is like, I was like, look, AI, we could do that now. We could give every single person who is considering asking for help, and asking for help is tough. Like none of us like it. We were just talking in the back, like, does any of us like asking for help? No, like we don't like it. It's hard, it's psychologically difficult. Can we give them the functional and the emotional help to go through that process? So that's what we built. A lot of trial and error actually. Like a lot of people think you plug the AI and it's better. It's not like that. Taken many months because we have to find not just the functional, like, hey, it's time to pick your title and now it's time to share, which is one of the most difficult things. We also have to find the right tone to do that empathetically, to help someone through those steps in the process so that they can get the help they need.

What about AI on the donor side? I'm thinking about if I show up and I'm like, is this Is this the right Jordy Hayes? Is this his campaign? Tell me more that might not just be on the landing page already filled out by him.

Are you looking at that for example? So the main way we're using AI on the donor side is to verify that the people who are paying in money are actually the right people. Got it. Doing all of those things to validate the pay-ins. And that's both a combination of us and the payment providers and the banks who are the backend of us. So that's the main use. What you're talking about is sort of more discovery, which is still early days. I mean, most supporters of fundraisers know the person, right? They know either very close relationship or they know them through someone. That's about 80, 90% of fundraising. Is it's really, what we're doing is we're modernizing a timeless thing that we've done for each other, which is if you used to live in a village and someone was in hard times, You put some money in the box. Pass the hat. Pass the hat, right? Yeah. That's what we're doing now. So mostly the sort of social proofing is through relationship networks. What's the tax treatment on a GoFundMe campaign? There are no goods or services there, so there are no taxes.

There are no taxes. That's right. It's a gift. Okay. Yeah. Well, how has the actual team evolved post-COVID? It sounds like you came in and you immediately made the team remote. Are you now in a hybrid setup? Set up? Yeah.

How is that evolution? I mean, we found our way through it. I mean, we were always quite a distributed company. So we had LA, Bay Area, San Diego. We got Chicago.

We got folks in— And is that because of talent or because of the customer base?

Like, why? It was a bit of talent. So the original GoFundMe was set up in San Diego. Brad and Andy set it up back in 2010. Then we needed great talent. So we also needed to go to the Bay Area. Now, for example, we have a We have a really amazing 100-person engineering and data science team in Buenos Aires down in Argentina, which is just— they're crushing it. So we are inherently distributed. So the work reflects that. We figured our way through it pretty well.

Pretty well. What's different about running a company like GoFundMe that's doing a lot of good in the world and good— every Silicon Silicon Valley company can paint a picture of how, like, they're saving the world, but you guys are actually providing a service that is making a meaningful impact.

It's like there's probably dozens of GoFundMes that are like, literally pay for my cancer medications, literally cure my cancer. And then there's a lot of AI companies that are saying, we'll cure cancer one day.

It's a very different vibe. Have other companies in the industry, like software vendors or payment providers, partners, do they chip in on their side? Are you able to negotiate lower rates on SEO?

No. No. Really? Are you serious? I mean, they're selling their products and services.

I know. I know, but—

You know, we get a little bit. People do like to talk about working with us, because generally it's regarded as a good thing. But that's at the margin, honestly.

He's a big fan of oligopolies and payment processing, apparently.

But we have great partners. Partners, so I understand they need to run their businesses, and we appreciate them.

What's different about GoFundMe? Yeah, no, I would just think that if I was a CRM provider, I might not be trying to get the maximum amount of value out of a platform like GoFundMe.

Well, maybe they're not getting the maximum, but they're generally doing their thing.

What about the surface area of GoFundMe? I mean, there's other platforms. Platforms that people use to raise money. I mean, I did a crowdfunding campaign for a food product at one point. But people do— I have a board game project, I have a movie. And there's this somewhat of a fuzzy line between a GoFundMe and something like a Kickstarter. Have you thought about broadening out? Do you want to stay narrow?

How important is it to define what the brand is and is not? Yeah, you're right. We do see people fundraising for local businesses quite a bit. A local restaurant that people love, maybe it's hit a rough patch, or a bookshop or something like that. But you're right, kind of business startup stuff.

Yeah, there was a time when crowdfunding was pre-seed stage funding or like a signal that you could send to your investors. Like the original Oculus VR headset was on Kickstarter. And then sold to Meta for $1 billion.

That's a very different thing. Yeah, it tends to be low. I mean, we saw $100,000 fundraisers for small but typically local businesses.

But you have the platform.

Have you thought about expanding? We've still got so much work to do. So we have expanded. So back in '22, we acquired a company called Classy, which does fundraising software for nonprofits. Oh, OK. Interesting. That makes sense. If you look at the industry, the TAM in the US, it's about 2% of GDP. $600 billion is given, the vast majority by consumers. Most of that is given to nonprofit organizations.

The vast majority is by consumers, not—

Not billionaires who are writing massive checks. Most of it is given by consumers. And most of that goes to nonprofit organizations. GoFundMe obviously started, and we are best known for helping individuals ask for help. Now, those individuals generally ask for themselves or someone else, but increasingly they're asking for a nonprofit. That was our fastest growing category last year, was individuals. Like, I just did one for Watch Duty, 'cause I think what John and team are doing is amazing, and they super supported us. So that's a trend. But our main focus has been making consumer fundraising ever better, which is why we launched the Coach yesterday and continue to innovate there. And then helping nonprofits fundraise more successfully.

Yeah, by building basically a network of other people who can fundraise on their behalf.

Well, that's the thing. That's the cool thing is like this intersection between those two things, because the average age of a nonprofit donor in America today is about 64. And nonprofits want to get more young people interested and connected, but a lot of young folks don't necessarily just want to donate, they want to do something. So we've given them the ability to organize, sort of enlist an army of supporters who fundraise for them. So I'll give you an example. American Cancer Society came to us and said, hey, we want to do something like this. We've had like 80,000, 88,000 people organize a fundraiser, whether swimming or walking or riding or walking the dog, for the benefit of ACS, which raises money, but even more importantly, it spreads the word. Because if you're telling me, I really care about ACS, it affected my family in this way, I'm doing this run over this month, would you support me? I'm going to think, hey, I'm going to trust that because I know John and I know ACS. I call it the double trust.

Yeah, that makes sense. Like SPVs, SPVs for nonprofits.

Yeah, it's kind of, you know, it's expanding the aperture of how you generate awareness and support.

What about political fundraising? How does that work today?

Is it not allowed on the platform? Very limited. And it's country by country, very, because we're in 20 countries and the rules are very, very variable. So it will depend. Some countries just some it is allowed in very specific ways. But frankly, it's not a significant thing for us at all.

We don't see much of it.

It's not an area of interest? It's just a different— Not really. I mean, there are pretty well-known and well-used platforms that focus on that. So we don't see much of it.

We don't honestly expect to. What about the future of the company? How is the company structured?

Where do you want to see it go? Well, we think there's a lot more help to generate in the world. I mean, our mission is very simple: help people help each other. And we think that it's interesting you had Travis earlier, because they're a good example. Uber, Airbnb is another one where it's a company that has taken a behavior that was frankly frowned upon. Like, remember when you were a kid, one of the things you were always told not to do—

Don't get in a car with a stranger.

—ever get in a stranger's car. And certainly don't ever go to a stranger's house.

Well, that's normal.

Now you take your Airbnb. Uber in your area. So in our case, it's not so— asking for help and doing it in public is still not a comfortable and normal thing. Totally. And so we're sort of expanding the aperture of like, no, actually—

But GoFundMe does have a different aesthetic to it. It's, oh, you're doing a GoFundMe.

Like, this is a normal thing.

It does. It feels much more normalized. But that's still building. Yeah, no, you're still building towards it.

But you're on the path. We are. And every customer I've talked to, I say every single one, has said something like, this was really hard for me to do.

If someone just put on their Instagram story like, hey, please send me money, that's very different than like, hey, I'm doing a GoFundMe for this specific thing. You can go here, there's a lot of details. You know that it's on legit payment rails. Exactly. There's accountability.

But there's still a lot of work to do in that journey. So that, we think we can do a lot more of that. And as I mentioned, we're in 20 countries. Some of those countries are still early in adopting and growing quick. Germany, for example, is taking off very quickly. Okay. There's more countries. Yeah. And then nonprofits. Like, there's a lot more work to help nonprofits who are as needed now as ever. Yeah. And we only offer our nonprofit solutions in the US today. Got it. So we got a lot more there.

And then we got some other ideas, you know. Yeah, it's very cool. I mean, it just— I can imagine it massively increasing the efficiency of nonprofits on the fundraising side because they have, instead of having to build out all this— they still have to hire their own staff to support fundraising, but now they get all this kind of off-balance sheet kind of effort.

We can help amplify. We can help. We do a lot of storytelling. We do that on behalf of our regular consumer customers. Now we can help to tell the stories of our nonprofits in certain situations, and that helps them get awareness and get more support.

Well, Tim, thank you so much for taking the time to come on.

You're welcome.

Thank you very much. Great to meet you. Appreciate it.

I'll see you in Pasadena soon. I might run into you.

Are you still doing your Watch Duty fundraiser?

You should do it.

I'm still running. Okay, I'm going to donate to that. That'd be fantastic. I love when I'm not a DMU of Watch Duty, but you never know. Oh yeah, yeah, yeah. You never know when you're— Exactly. So you never know. I'm in Malibu, so we had a crazy time last year. I hope you guys are doing okay. Yeah, luckily we both got out.

Thank you so much. Take care, guys. Let me tell you all about Okta. Okta helps you assign every agent a trusted identity so you get the power of AI without the risk. Secure every agent, secure any agent. And let me also tell you about AppLovin, profitable advertising made easy with Axon.ai. Get access to over 1 billion daily active users and grow your business today. And without further ado, we are joined by Nikesh Arora from Palo Alto Networks. Good to meet you. Welcome to the show. Thank you. For those who might not be familiar with a little bit of your journey, I'd love to kind of get everyone up to speed on what you were doing before. And then I want to spend a lot of time talking about just the transformation, if you can call it that, of Palo Alto Networks, where the business is today. But maybe take us back a little bit.

Where do you want to go?

How far back? Well, I know you handled business at Google, but—

Let's go even—

let's go back to the very beginning. Yeah. As far— no, I grew up in India, as you might know. Came here to go to business school. It was a tough time. You couldn't get a job in 1992. It was one of the recessions. In the country. So I wrote 400-plus letters.

In those times, you guys are younger than me.

Cold letters? Cold letters. Just cold letters. You couldn't email. You had to go. I had to rent a Mac for 3 months.

No, you rented a computer.

I rented a computer. I found 400 addresses of hiring people. And I would sit down, write the letters, type them out, go print them out, put them in envelopes, stuff envelopes, and mail them.

It's such a different dopamine loop doing a cold letter, because cold email is amazing. There was only one. No, no, I know, I know. But I'm saying, like, today kids are used to like, okay, I got to write this email. Maybe they use AI now. But then, like, oftentimes, like, for me, I either respond, like, within, like, 30 seconds or 30 days. But, like, you know, you quickly kind of get on get on that loop, but sending an actual letter out and then just kind of praying and waiting for maybe—

Every day I'd get stacks of them back. I got— I still have them. I have 400 rejections from 1992.

Wait, 400 letters, all rejections?

Yes. I only got one job, right?

Many of the others said no. So I had 400 rejections.

So how many of the companies are still around?

Some of them are. Some of them you wouldn't want to work for. But I sometimes run into people. I ran into one guy who started a private equity fund, Afron. I remembered his name at the bottom of the letter. So, I was sitting and having a meeting. I said, you do realize that you rejected me in 1992? And I was an advisor to a company he bought. I don't hold grudges. I didn't hold a grudge. I'm glad. I have to say thank you to all of them, right? Yeah, of course.

God forbid I'd gone to work for someone. Where did you end up?

I ended up at Fidelity Investments right out of business school. And even there, I got rejected from 6 people I wrote to in the company. Thankfully, they had no common HR system or whatever. The 7th guy hired me. Great. And he said, oh, her resume looks interesting. We should talk to her. He couldn't tell whether my name's male or female.

Oh, sure, sure.

So I signed up for Fidelity. And when I was in business school, I was applying. I'd get phone calls. I went to Northeastern. So I get a call. So, we're just double checking. How was your time at Northwestern? Why did you say Boston? I'm like, that's Northeastern. Oh, I'm so sorry. I've got another call.

I got the exact same thing. I went to Northeastern. Same thing. I'm not going to correct you.

Keep going. So, I had all kinds of stuff. But one of the calls told me, you don't have enough finance on your resume to get a job on Wall Street. So, I went back to school while I was working at Fidelity. I got a master's in finance. I got a CFA. CFA, and I was teaching a CFA class in 1995, '96 actually, '96 October, to do portfolio hedging and shit. And this guy sitting in front of me said, oh, you work at Fidelity? I'm like, yeah. He's like, hey, do you know all these people? No. So they're on the fund management side. I'm like, I don't work there. I'm a corporate finance dude. He's like, wait, you're teaching me CFA Level 3 and you work in corporate finance? I said, you want to pass the exam? I didn't want to discuss my credentials. To his credit, the next day I interviewed at Putnam. There you go. So I finally went to work in money management for 2 years. And most interesting time, most boring time. I just couldn't sit all day. Mid-'90s? Yeah, '97, '98, '99. So we're—

We're in the internet boom. We're in the internet boom.

Henry Blodgett telling us that AOL is going to be the biggest company in the world, right?

Yeah.

So I did that for 2 years, did well there, got bored, left. And I started— One of the CEOs— I just covered telecoms at that time— one of the CEOs was the CEO of Deutsche Telekom. He said, come work with me. So I took my bags, flew to Germany, to Bonn, and worked with him for 5 years.

I started a mobile data startup at that point in time. And the name of the company that you started there?

It's called T-Motion. T-Motion. But that was the time Deutsche Telekom— I worked on the team that acquired VoiceStream, which is now T-Mobile USA. Yeah. So we did that. I worked there for 4 or 5 years, and then I quit. I was tired of commuting from London to Germany and back. I was in my 30s and I was sitting in front of my laptop and he said, "Hey, I had this headhunter calling me. There's a job. It's a company from the US. They just went public in some reverse Dutch auction type scenario. That job's too small for me. Do you want to go interview?" The guy was the CEO of T-Mobile UK. So that was I met that guy who became my boss, Omid Karzahani. He interviewed me. I told him it's a really small office. And he said, well, why don't you have an interview here, then I'll take you to California and you can meet the founders. I'm like, OK. And then he calls me a week later and is like, oh, Larry and Sergey are in town. They would love to see you. So they made sure the interview was not in the small office. So I met Sergey in the British Museum while walking around. That's amazing. That's where I got interviewed. Then I came to California and I met Eric Schmidt. And Eric's like, it's interesting. Your CV is very interesting, but we're looking for a head of Europe and the job is sales. I said, so should I just go back? And I have 2 days of interviews. I'm not done sales. Should I just go back? He's like, no, no, no. We need smart people who can learn stuff. And that was 2004. So I was the first person hired after the IPO of any senior capacity. In sales? I ran Google Europe. And slowly and steadily, we grew Europe. And Europe was like—

And what was the shape of that role? Were you talking to the largest companies in Europe about just buying ad space?

You want me to make it cool, or you want me to actually tell you what I did? Both. Let's go with the role. But look, it was a company just— the thing is, you're thinking December 2004. The company's just gone public. It's worth $18 billion. There's 2,000 employees. Employees. Yeah. And Europe has like 180 people. Every office is a rented Regus office except for one in Dublin. Yeah. Wow. So what you're doing is you're like blocking and tackling. You're hiring people, you're getting offices, getting stature, getting stuff. And yeah, forget selling ad space, showing CEOs, showing companies this is what search does.

Oh, this is just work. Yeah, it works.

Yeah, yeah, just get deployment. In the 5 years I was there, we went from being 25% of Google's revenue to 46% of Google's revenue. I think in history, one of the few tech companies where Europe and US had the same market share. So you have all these people come through here, you should ask them. Generally, you spend a lot of time in the US, you get a large market share here, and then your share of Europe or the rest of the world is smaller. So we got there, and that's when kind of a whole bunch of movement happened. Sheryl Sandberg went to Facebook, Tim Armstrong went to run AOL. I was kind of the last man standing. Standing. So when my boss retired, Eric called me and says, we think you should move here and become chief business officer.

So I moved to California. Most underrated title. Should be the chief of business.

We joke about this all the time.

It's amazingly vague, but also incredibly important. But I remember my first meeting with Larry when he became CEO. And he said, hey, come spend some time with me. I spent all this time, put a beautiful presentation together. I'm going to sit down with Larry and explain everything we do. So I go there and we're talking about Hyperloops and vacuum-controlled tubes on how you can get out of the car on every highway that will improve the traffic situation in California. And I'm here chomping at the bit, wanting to get to my presentation. He's talking about Hyperloops. So in the end he says, you know, Bill Campbell, who was a great guy, mentor to many people in the Valley, says, "Bill tells me you're doing a good job at what you're doing. I got a lot of work to do and fix on product because tech companies die if product doesn't work. No tech company became great because of sales and business." He said, "So you keep doing what you're doing. I'm sure if I had a few hours, I'd help you get your efficiency up and do it better, but I don't have the time." So I stood up and said, Let me know when you have a few hours. That was that. Then Tank, to his credit, because we had a great relationship, and his staff became 10 product people, the CFO, the lawyer, and me. That was it. No other function was represented. He was like, you go take care of everything else. I want these product guys because I want to tell them what I want to build. And we can see what resulted in.

At any point early on, did you feel like Google was is significantly overvalued?

I still think it's undervalued. It's kind of the enormity of what the potential was. Sometimes when you're in it, you don't see out of it. It's hard to tell how big it's going to become. But of course, in hindsight, it's wonderful. We can all say that was bound to happen, but the enormity of what what we were doing was not apparent to people there. We bought YouTube. I remember sitting with Salar and Chad Hurley and saying, "Let's go buy Netflix." There's a story about that, but it's a different story. Salar's like, "No, just give us 20 engineers and $5 million and we'll build it." Every tech company has those kind of people and they tried, but YouTube's— Juggernaut. Big in its own right. We were there when all this stuff was happening. And you can see, I remember Larry tells a story that he met Steve Jobs and Steve said, "If I give you one piece of advice, the word is focus." And Larry's like, "No, we're gonna try a lot of things and see how many of them work." So there are many ways to get to great ways.

[Speaker:MALE #2] Yeah, the head in the clouds philosophy clearly did work for Google's academic culture. Do you feel like that's the right characterization of Google, this academic, the 20% time, the lots of different projects, primordial soup to some degree?

Did you try and continue that? I think part of— this is my opinion. I'm sure a lot of people have an opinion. They do too. But I think the secret sauce is the consistent focus on product greatness. And that comes from the founders. You see, look at Travis here. I saw him earlier. Companies take on the form of their leaders. It's apparent in every company. Pretty much, if you sat down and did an experiment and I wrote down 5 adjectives over here, which are company cultural characteristics, and I wrote down 5 here, which are founder characteristics, and I said, "Go match them," I bet many people will be able to match them because you can see the similarities between how leaders lead and how companies adopt to the culture of the leader. I think that's where Google's culture comes from, from the founder. And the founders were very product obsessed. As you heard, Larry didn't want to spend time on business. I remember him going to the first ad all hands, and he walks up and says, "Well, I gotta tell you guys, I hate ads. They're intrusive and they're bad." That's a great motivational speech. Everybody, these guys, wants to run out and go work on search, because he just told them what they do is not interesting to him. But you look around, every product that they built, they didn't look at monetizing it for a very long time until they believed the product had become ubiquitous and the product had become interesting. So I think that product obsession is part of the culture. I think that's partly because of where their secret sauce comes from. And now with the amount of resources they have, they can afford to do that for a very long time. But having said that, we did it even then. Gmail had no monetization for years. Or Google Maps had no monetization for years.

But it was growing like wildfire. Yeah. So coming into Palo Alto Networks, did you—

I did a little stop in the middle with Masa. I know Travis was talking about easy money.

I was part of the Ether. Yes, I was. But that was after Google.

But coming into Palo Alto Networks— Did you guys ever pass on a company?

No. I passed on Uber.

No way. I did. What round?

What round? I think it was a $16 or $20 billion round. And then I left and Masa did the deal. I passed on Adam Neumann. There's a book. I told my guy who was the analyst, I said, listen, we're a tech investor. Any time you bring me a business which is masquerading as a tech business, which is a commercial real estate business, you should go work there.

Don't work in tech.

Good call there. It's a great story. I've made a lot of bad ones too, so don't worry.

Yeah. So going into Palo Alto Networks, you have these 5 core 5 cultural tenets from the company, 5 cultural tenets from the founder. You're coming in as a new CEO. How did you think about resetting the culture, expanding the culture, bringing what you value to bear across the company in a way that's not disruptive but gets you where you need to go?

Oh, there's no way to be not disruptive. Let's get that out of the way. Life is about standard deviation. You want excess returns. There are no excess returns. With no standard deviation. This has been proven in every facet of life. You want to date a good-looking person, you have to take the risk. You want to build a great business, you have to take a risk. You want to make money in investments, you got to take a risk. I don't think there's a non-disruptive way of creating transformation. Having said that, as you probably know, I knew nothing about cybersecurity. I'd never run a public company. Other than that, I was a perfect candidate. So, let's find a guy. I can only imagine the board deliberation. Like, if you look at a board, they write these long job descriptions, brief a headhunter, say, hey, go find us somebody. So, imagine that conversation. Find us somebody who knows nothing about this job. So, I'm like, either they were geniuses or they weren't paying attention to the headhunter. Anyway, I got the job. I show up there. I'm trying to learn cybersecurity and I'm trying to understand the culture of the firm. My predecessor, Mark McLaughlin, is a West Point guy. It was a very high-integrity company. My father's a lawyer. We come from the same stock. I like the fact it's a high-integrity company.

It's important for a cybersecurity firm too.

It's important. Of course, it is. It's important for every company. Especially. Yes, especially so. It was the best product in this category, which is a good thing. What I discovered was— which happens in that industry, people get caught up in the success of the current product and stop thinking about what comes next and what we have to look around the next corner. I think leaders' jobs are to think about what hits us in 2, 3, 4, 5 years and how do you prepare for that now? Because I have a firm belief it takes 4 to 7 years to build a great product. Maybe it's become 3 to 5 or 2 to 4 with AI, but you go back and think about it.

Anybody you've had through here— And how do you— what's your personal— definition for, of what a great product is? Because there's a lot of products today that grow really fast that aren't necessarily great.

Well, I think there are two aspects to this, right? There's a product when you're competing with somebody else, in which case you have to build something better than that. So much better that the customer is willing to switch because you have a much better experience. We take Google Search, right? People were doing, there were 14 search engines before Google Search came. It took Google 7 years to get to compete with the other 14 people where the users were willing to switch. And there are products and categories where there's no competition, right? There's nothing called Uber. I'm going to try it, whatever it is. But if you think about what made it a mass market experience where you can scale it around the world, it took 5 to 6 years for Uber as well. Take YouTube. So if you go down history, it has taken 4 to 7 years to make it a compelling product that people want to use and want to switch whatever they're doing, switch behaviors. That's when you start seeing scale work. Now, the current AI world, all bets are off. We'll see what happens. OpenAI took a lot less, but they had no competition. It was the best thing going at that point in time. When I walked in, I realized the thing the company wasn't doing was one, they weren't innovating fast enough. New products. For the first 3 months, I felt like We're talking past each other. So I took a weekend and I sat down and wrote all my business principles I'd learned in my entire life. And in hindsight, I would urge every leader to do that before. I call it my belief document. So I wrote it and I showed up and I printed 8 copies and dropped it in front of everyone on my team saying, read it, let's debate. It's like, what is this? I'm like, this is my recipe. People. This is why I act in certain ways. This is why we have disagreements. When I tell people, no, you can't hire that person, they get really frustrated. Oh my God, I've been looking for 6 months. I'm like, that's not a good person. Bs hire Cs and As hire As. And the more senior a person, you're going to destroy an entire part of my organization if that leader is not good. Oh, I get it. That's why you don't let me hire that person. Now, maybe that's an A, maybe that's a B. That's a different debate. So it became so much easier when we were all singing from the same hymn book. And now it's kind of mandatory reading and debating for every new senior person in our company. They have to read my belief document, which is kind of like sets the tone for how we operate. Is it still one page?

It's about 11. Okay.

It's more like a— It's a little booklet. But you can summarize it with any AI engine.

It's not a problem.

You don't have to see 3 pages. But what happens is like, And people wanted more clarification. What does that mean? Well, you want to read the long form?

You can read the long form.

You want to summarize it?

You can summarize it. What about product development? Did you come in with an idea around build versus buy, a framework for evaluating those trade-offs?

Did that evolve once you got into the actual role? So, look, cybersecurity is the youngest subsector of technology. It came about when we had connectivity, right? The iPhone, suddenly everybody's on the iPhone. Every firm, every company is trying to get their applications on your phones, which means we're all trying to take what used to be controlled access and make it accessible to every consumer. So it came about 25 years ago, and every new piece of technology creates a whole series of new sets of risks, a whole new set of products that deliver. Now, interestingly, it's also one of the most innovative industries in the world because the bad guys aren't saying, oh, let's try the same thing we tried yesterday, got blocked. Every morning they wake up, they're trying to find a new way to get into your business. So you got highly innovative, new set of market in cybersecurity. And as a consequence, the industry structure was extremely fragmented. There were 2,600 cybersecurity companies and the largest company had 1.5% share, which was us. So you have 1.5% share of the market. Then you need to build a strategy. How do I get from 1.5% to 10% to 20%? Yeah, right. Because you look at every other piece of tech, There are lots of people doubling their market share. So I said, look, this is a math problem. The math problem is how do we get to 10, 20% market share? You reverse into that and say, well, you got to have a lot more stuff to sell to the same customer to have more market share. And then we said, okay, great. Now, where are the product categories where we can build products which are up and coming? The good news is we live in a constantly evolving technology world. So I sat down with the founder who was there then, and chief product officer. And we sat down and literally on a napkin wrote AI, cloud, and network. Cloud and AI were new, and we had the right to win in network, but we only did 1 out of 6 things in network. So then we said, how do we short circuit the 4 to 7 year problem?

Only doing 1 thing in network when your name is Networx. It was in the name the whole time.

That's right, exactly. But they got happy. They got fat and happy. An $18 billion company you started at $3 when you went public. This is great. It's up 6x. What are we doing wrong? Nothing. Well, I couldn't rest on that. I had to go do my own 6x thing. So you sit there and you say, OK, we need to get to be number 1 in everything network security. So it's going to take me 47 years to build a great product. That's not enough time. So I said, let's go find who we can buy. And I know there's this thing where I don't think many tech companies execute M&A well.

Is that because of purchase price or post-merger integration? Purchase price is—

Identifying the wrong synergies? Purchase price is an irrelevant artifact. OK. Either it's going to work. If it's going to work, it's going to work phenomenally well. Or you're going to screw it up. It's not what you paid. It's what you're able to do with it. You could say that Instagram was expensive or YouTube was expensive. DoubleClick was expensive. They all work perfectly. AOL Time Warner, different story. It boils down as to how you execute past the price you pay for it. That's not the relevant thing. In tech, when you buy a company, you buy a team, you buy an existing product, and you buy a roadmap for the future. The question is, can you deliver on that roadmap? Accelerating that roadmap, does it work? So we wrote our little M&A manual. We said because our product is fast, we got to keep the founders there. We got to make sure we have the right people running the company. And over time, we've gotten really good at it. We sign a term sheet and we ask the founders to sit with our team and redesign the product roadmap so we like it and they like it. And if they don't agree on our expectations and we don't agree on theirs, we don't buy the company. Makes sense. We make them in charge. My teams are having to work for them, which makes them really unhappy. And not many of them like it. But I'm like, look, these guys went out there, raised money, kicked your ass. Yeah. In your category. And you want them to work for you? That makes no sense to me. You're going to work for them, learn from them. Yeah. Some people get the hang of it. Some people move on, which I'm fine with. Yeah. Yeah. Right. So these people had less resources. They struggled, they hustled, they built a product, and they kicked our ass. That means— so our job is to enable these people. We look at them and say, whatever business plan was when you were a small private company, find me a business plan that's twice as assertive and bold than the one you had then. So we buy them, we pay the money, but we also give them more than they would have gotten anywhere else. Because I'm sure we will slow them down a little bit. I've got a bunch of people who will try to ask questions.

And get them to slow down. But you're almost immediately expanding the TAM on day 1. There's so many more firepower that you can bring to bear.

And then what happens is we've built a phenomenal system to take them to market. So we say, we're going to take you to market. I have 3,000 people in the field. We're going to find a way of catching their fascination, making sure they're going to get excited and make money. And 3,000 people will go out there and see 10,000 customers. Some of them are going to bite. You're going to go with 4 resources and go meet 20 customers. We're going to go to 6,000 or 10,000 customers. So that's where the secret sauce kicks in. We've bought 34 companies so far. Wow. And we do this for our board. I think our hit rate on things that have worked is over 70%. That's amazing. The last 3 are obviously new, so we have to see. And those are the biggest ones. But it has worked out so far.

Yeah. Do you find yourself ever going back to your MBA, your CFA for— Guidance on financial management of a public company, or is it just so different that you can just pick it up? Because it feels like there's a whole new set of tools that come into the toolbox when you're running a public company, whether it's dividends, large debt issuances, buybacks, there's M&A, there's so many different things in just managing a stock price that a public company CEO has to do in some ways?

What do you think? Well, the MBA is a social experience. I think people go to business school and make a lot of friends and they get to know people. I have a classmate who went to school with me and I can go cut a deal with him. I think it's not what you study in a case study. It's a mindset. It's a way of thinking. It's kind of like learned behavior. Over time, you get good at it. You don't know why you you do it, but it becomes part of your muscle memory. I think that's what happens. I'm a huge student of business. I spend a lot of time listening, reading, talking to people about this stuff. You just get muscle memory over time. You just get a sense of what the right thing to do is. I think the part which we underappreciate, people who do well are people who go back to first principles and try and abstract and say, from a first principle basis, is this important, is not important. For example, I don't have to manage my stock price. The stock price manages itself. The stock price— the market is sometimes smarter than individual CEOs. We have to respect that. There's a lot of players out there. They're not all silly. There's a reason your stock trades where it trades. Sometimes it trades because they don't trust you. Sometimes it trades there because they don't believe your prospects are as rosy as you believe they are. Sometimes it does well because it anticipates the prospects are better than you think. The market has a lot of different variables it assesses. Now, to the extent that you're convinced this is the right direction to go, you're convinced that you're executing well, that's great. If you believe there are some challenges or flaws in the way you're executing, the market will give you a hint. The market will tell you. Now, look, the market gets it wrong sometimes. It gets it right many times. It's just another variable you have to take into account. The idea of managing a stock price is a bad idea.

You can't. But if the market gets it wrong, you have a financial incentive to buy your stock or—

Yes, we just did. We bought all the stock because the market said, oh, we hear this thing called AI is eating software. We're going to dunk everybody because Claude decided to put out some coding security product. Oh, shit. This is going to destroy a $300 billion industry because Yeah.

So Daria launched a product. Yeah, I saw a chart that showed the mentions of moats on SEC conference calls, on public company earnings calls, is going vertical. Have you been thinking about restating the moats to shareholders?

You know what I did a few earnings calls ago? I put my earnings script into Gemini and said, assess my earnings script and tell me honestly What do you think? It's like turning into a psychologist, says, not sure what you're concerned about, but you're trying to use the word momentum and exciting more times than you normally do. Calm down.

So no, I'm not going to say moats.

I'm okay with it. I'm not going to say moats. We don't have to say moats. But look, you expect me to sit here and speak my book, but LLMs are phenomenal at the 90% 90% problem. 90% problem is make me a movie, paint me a picture, merge this photograph.

It's a 90% problem.

If you get it wrong, it's not a big deal. There's nothing destructive. For the same reason, you wouldn't want Gemini driving your car.

There's also those 90% problems, there's so many of them where you will never know exactly if you did the best job or you just did an an okay job. Like sometimes you get some signals, but it's not as binary as let's say, did we expose user data to a bunch of people?

That's a whole different, that's a whole different conversation. But before you get to exposure data, like, you know, you write a letter to, you know, respond to my emails. Okay, fine. It's 80% good enough, much better than me having to go do each one of them. So I think we're okay with the 80, 90% outcome, but you would not let Gemini drive your car. Or ChatGPT drive your car and hallucinate. For the same reason you don't want generative AI managing your cybersecurity, because we're in the 1% business. Remember, I'm trying to block that one guy who's going to get into your company and steal your data. Yeah, that's a 1% problem. It's a 0.01% problem. So we rely on thousands of machine learning algorithms. We rely on tons of pattern recognition, domain-specific data. So we're trying to solve the 1% problem. Generative AI does not solved 1% problem, it solves 90% problems. We solve 1% problems. So we're least threatened in this notion. There are parts of our business where we could leverage generative AI to make it better. Yeah. And that you will see, like the code scanning thing is a perfect example. Claude launches it, people get petrified. He's a $3 billion industry, we're a $300 million market. And the stock, the market loses, takes down $300 million market cap that day. Fantastic. I was like, great. Now the market's getting it wrong. Let's go buy our stock.

So yeah, the market does get it wrong. Explain to me how the surface area of cybersecurity threats is changing. I mean, off the top of my head, I could think of geopolitical risk, more AI endpoints where vibe code is less secure. And so you need more cybersecurity. Also just the hackers that are out there out there are now AI-enabled, so their threats are getting more sophisticated. What is this year looking like from just the scale of cybersecurity threats broadly?

So, you know, if you go back to first principle, what happens is every technologist who's positive and optimist builds software for the best use case, for the best case. They don't build it for the worst use case of it. You didn't build a knife and say, "Oh, let's not release a knife because people might stab each other." You say, "Well, this is great. You can do so many productive things with it." We always build technology with an optimistic, positive attitude. Look at what AI is doing. We're all building from positive attitude unless you start getting into debates about can we use it for war or not. For the most part, you're building it for the right reasons, the positive attitude. You didn't build the first airport with TSA in mind, did we? How are we going to make sure we could torture people and take them to a line and make them take their clothes off and shoes off? So we didn't do it. So we actually introduced security afterwards because we realized there are bad use cases. Unfortunately, same thing happens with technology. We build this stuff and say, let's go use it. People aren't going to do bad stuff. And then you suddenly see an agent hacks somebody's business. Oh, shit. We got to go figure out how to stop this agent from hacking this stuff. Our job in cybersecurity is to anticipate the bad use cases and try and throw a ring fence, a series of guardrails, a series of protection around technology. To the extent technology evolves at a fast pace, this is good for us because we have to constantly find antidotes to that. Then you have to make sure you're in the zeitgeist of what's happening. So you want to be there. So, you know, a little flashback, like this industry, for the last 8 years used to stay in their swim lanes. There are 5 swim lanes in cybersecurity and so you do your thing, I'll do my thing, you do your thing, I'll do my thing. Suddenly we said, this is bullshit, we should be in all 5. So now we're in all 5 and suddenly the whole industry is flipped over its head. Every time we do an acquisition, we see 5 copycat acquisitions done in our industry around us because people say, oh, this guy must know something. Actually, this guy knows nothing, but thank you very much. Now you just validated my thesis, let's go. So we're going doing that. It is good for us when technology goes through a tremendous amount of evolution, whether it's cloud, whether it's AI, whether it's people going back to coding. Because funnily, coding causes more activity to happen in laptop. It needs a whole different security in your laptop than you used to have. Put an MCP server there, I need to protect it. Never putting any kinds of API access on your laptop. The constant evolution of technology demands more innovation. Innovation in our industry. More innovation in our industry, more deployment of technology around the world causes more demand for cybersecurity. That's good. Of course, we have a geopolitical set of issues. The more you go into war, the more— nation states are the most prolific cyber hackers in the world. There are nation states out there who literally unofficially sponsor cyber activity because they need it as a revenue source. It's also a training ground. You keep doing it in peacetime so when I need it in wartime, I can use you.

Oh, sure. Interesting.

Yeah. So there's this whole cat and mouse game that's played in industry. So geopolitical uncertainty, economic uncertainty, all these things are catalysts for activity. And that's on one side. On the other side, if you look at it, because of the fact that we never designed it with security in mind and the pace of technology is evolving, almost every enterprise— this is behind the 8-ball in the modernization they need. And it's kind of like a hard decision to make, right? If you've never gotten hacked and you sit there and become complacent, saying, look, I've been spending $50 million a year and it's working. And suddenly one day, oh shit, it's not working. Somebody just got into your business. Those are the best days for us, by the way. When somebody gets— shit hits the fan. I don't want to get excited about people's tough times, but— No, that's why we have a team, which is our benevolence. We have a team which you can call. We don't charge you for that team to go help you when the shit hits the fan. Because we know you're just about to get religion. You're about to go transform everything in your infrastructure because you want to be at the bleeding edge. And again, security is kind of like this where you could spend too much or you could spend just right. Right now, people generally, the infrastructure is antiquated. Relative to where it needs to be. There's a lot of transformation that needs to happen. So I think this is going to be sort of a perpetual opportunity because combine that with the pace at which technology changes, that's great. That's kind of one part of it. The other part is, which was our insight 3 years ago, if you take a look at what Salesforce has done or Workday has done, you guys are young. Back then when I was working at Fidelity, we had 27 systems which combined We made up customer relationship management because everybody had their own system. You had keys, you had like data references because you build your app at that time. And cybersecurity is kind of there. Like every customer has more cybersecurity vendors than technology vendors outside of cybersecurity. Because they're like, oh, I got 80 of them because I got everything solved. Like, no dude, you got to get them all together. So we're in that phase where we're building the first sort of combined platform of cybersecurity And that's why we wanted to be in every swim lane. And as the market converges, we will be able to get people into less and less vendors where data can talk to each other. Because at the end of the day, there's only two factors that go in our business. One factor is if it's bad, stop it. If I know it's bad, I'm pretty good at stopping it. The problem with cybersecurity is never the known bad. It's the one you don't know. So all the money is in figuring out when you didn't find it at the door, is something bad going on or somebody is trying to intrude into your business? And that's where all the action in cybersecurity is. So that requires data to be connected. Yeah. So, and this is the last fact, leave it and you can go wherever you want, but the average time to detect and remediate a cyberattack is 4 days in the United States. Now with AI, The fastest time to attack a company and exfiltrate private data is 28 minutes.

Wow.

That's wild. I was shocked when I went to Palo Alto and I heard that. So in 7 years, we've gotten ourselves to 1 minute. So we just got to get everybody else to 1 minute. Yeah. Yeah.

So a lot of demand in the future. Has anyone ever gotten a job working for you by saying, sending you a physical letter?

Sending me a physical letter and gotten a job?

The times have changed, but you know— That creates the opportunity. Have they changed though? It is a way to stand out. You probably get a lot of cold email.

You probably don't get a lot. I get letters too. I get letters, I get pictures, I get—

What advice are you giving to young people who are entering the in the job market or want to work at Palo Alto Networks?

I think it's the Wild West right now out there. We have stopped looking at your CV. We run hackathons every second weekend. And anybody who can write code and get ahead of people around them, we'll hire them. I don't care when you went to school. That's awesome. Because I just think people have to learn. This is something that has not been taught in schools. I'm glad if you go to school, it's good. You have social skills, you have friends. This is good. We want mentally balanced people. This is good. But we want to make sure that you know where the world is going. I'm a huge optimist on AI creating job opportunities. I think we'll go through a transformation period, but we're trying to find people who know how to— I have to overwhelm my company with more people who know how to use AI than less people. Right now, more people don't know how to use AI in my business. Less people know how to use it. Yeah, I think if traditional companies— traditional include ourselves— companies have been around for 10 years, 15 years, want to get back and win in the next iteration of technology, we have to transform as fast as we can. That requires more people who can use AI than less people. Any CEO who walks in here and tells me they have more people who know how to use AI than less, they're going to win. That's the biggest need of the hour. So I'm out there trying to hire people. So if anybody's listening to this and they think they are the best vibe coders, the best user of creative AI, write me a letter.

Amazing. I hope you do. I hope you do.

Send me a letter.

Vibe-coded product via physical letter. There is an API for sending physical letters. There's also an API for sending what's called lumpy mail, which is a letter. Have you ever gotten a letter that has like, it looks like it has something inside of it? It's lumpy.

Usually that goes into our mailroom and goes into the bomb squad.

Yeah, exactly.

You may find it in the incinerator because people figured out there's an arbitrage there because you're more likely to open an envelope that looks like a package and you'll open that and then you just get the ad.

Unfortunately, some of us have people who open them because they don't want us to have the risk.

Of course, I imagine it's pretty important in your role.

Well, that has to be—

has to pass the regular filters as well. Well, thank you so much for taking the time. To come chat with us. This is fantastic. Let's do this again soon. All right. And before we sign off, let me tell you about Vibe.co, where DTC brands, B2B startups, and AI companies advertise on streaming TV. Pick channels, target audiences, measure sales just like on Meta. And let me also tell you about Console.com. Console builds AI agents that automate 70% of IT, HR, and finance support, giving employees instant resolution for access requests and password resets. And You heard, you heard Eric Lyman mention Shopify earlier. Shopify, of course, is the e-commerce platform that grows with your business and lets you sell in seconds online, in-store, on mobile, on social, on marketplaces, and now with AI agents. Jordy, is there anything else in the timeline that we should close out with? Any breaking news that we need to get to? And while you look that up, let me tell everyone about Sentry. Sentry shows developers what's broken and helps them fix it fast. That's why 150,000 organizations use it to keep their apps working.

There's some breaking news. What is that? Anthropic researchers are departing to launch a Neo Lab. Oh. In talks to raise capital for a new startup at a $1 billion valuation.

Yeah. Not— What a crazy time to launch a Neo Lab. Weren't we just asking Tyler about this? Is there gonna be another Neo Lab? Is the boom over? It's notable because we just haven't seen a lot of people— Leaving Anthropic. Yeah, that's right. But I'm very interested to see what they wind up launching. I'm also very sad that we didn't get to our lightning round today. We had a bunch of great people scheduled. Of course, we ran long, um, but we will have them back on the show soon to get all the updates there. Let me tell you about Railway. Railway is the all-in-one intelligent cloud provider. Use your favorite agent to deploy web apps, servers, databases, and more while Railway automatically takes care of scaling, monitoring, and security. And finally, let me tell you about Cognition. They're the makers of Devin the AI Software Engineer. Crush your backlog with your personal AI engineering team. And with that, I think we can play our credits.

What a fun show. What a great show. Nikesh, a legend.

Gustav, TK. Whenever we travel, everything kind of comes together and we have to do a catch-up episode. But this one was one for the books. Tons of great conversations. Thank you so much for listening. You have a— leave us 5 stars on Apple Podcasts and Spotify, and we will see you tomorrow or Monday.

We love you. Goodbye.

Done. I'll face myself to cross out what I've become and raise myself and let go of what I've done.