Loading summary
Transcript932 lines
- [00:01]
A
You're watching TVPN.
- [00:02]
B
Today's Friday, May 8, 2026. We are live from the TVPN Ultradome.
- [00:09]
A
The temple of technology, the fortress of finance, the capital of capital.
- [00:14]
B
Jordy has some new buttons.
- [00:15]
A
I got some new buttons. That's why we had to be off for a couple days.
- [00:18]
B
Yes, Deep engineering work. The Great Lock in continues. The Great lock in no. We will go into a review of the last few days, but first we'll give you a review of the next three hours because we have Josh Reeves from Gusto coming on, Jonathan Neiman from Sweet Green, Zol from Refactor, and then Brian Chesky from Airbnb coming in to the TVPN ultradome. We're very excited. Lighter schedule. It's Friday. We're having fun. We're gonna be hanging out here on the stream telling you about the American economic roller coaster.
- [00:54]
C
You've heard this story before,
- [00:58]
B
just non stop. Okay. So the American economic report.
- [01:03]
A
Good to be back. We miss you guys.
- [01:04]
B
It's good to be back. Yeah, we got a bunch of stuff going on. So a whole bunch of economic news over the last couple days. Stock market is absolutely ripping. We should be in white suits. Intel's up 20% today, so almost. And there's this split between the AI economy and the real economy, the American economy. And there's a whole bunch of different things that you need to puzzle together to get picture of what's going on. And then I'm discovering that there are K shapes within the K shapes. Even in the real economy there is divide between different companies. And so I wanted to walk through that. So let's start with the intel news. It's up almost 20% today on the news that they will be making chips for Apple. There's a report in the Wall Street Journal about this. They've released, reached a preliminary chip making agreement. It's not completely locked in, but there's a whole bunch of extra context here that we've been tracking for the last six months, 12 months as some of this was. The revitalization of intel was something that a lot of folks were clamoring for. Ben Thompson, myself, a lot of folks, Leopold Aschenbrenner, were all hoping for something and then plans are starting to come together. So the talks have been described as intensive. Intensive talks between the two companies have been ongoing for more than a year. Apple's been talking to Intel. Interesting. Of course, intel and Apple have been long term partners for the 90s 2000s up until the Apple Silicon program where they started going Apple started going direct to tsmc. Can they get them back? It seems like potentially in some form factors, but we're going to dig into it. So they have hammered out a former deal in recent months and it's still unclear which exact Apple products Intel would make chips for and manufacture them both its own designs. So intel has two main businesses. It's both a design shop and a manufacturer. They have a fab and a and a design unit and in its foundry unit. Both businesses have been underperforming for years before Lip Bhutan came in as the new CEO and was vowing to revitalize them. And so the last summer the Trump administration struck a deal to convert nearly $9 billion in federal grants into intel
- [03:18]
A
stock, giving the US roughly 20$21 a share. It's now at $125 a share. White House has got to be feeling pretty good.
- [03:28]
B
Yeah, huge. I mean I'm pretty sure Jensen has made over a billion dollars on the $5 billion investment. I mean just today if it's up 20% and he must have seen a big growth in that position since then. So just today probably a $1 or $2 billion print. And so the reporting here in the Journal says that the Trump administration was actually key in bringing Apple to the table, putting pressure on sort of both sides to say hey, let's think about the future of American manufacturing resil, reducing Taiwanese dependence on a foreign supply chain and giving Intel a real shot at underwriting the next big fad that they want to build.
- [04:09]
A
Yeah, Nvidia got in at $23 a share with their $5 billion investment. So up around a 5x.
- [04:17]
B
Not bad. That's not bad at all.
- [04:19]
A
The guy really needed it.
- [04:20]
B
He paid for Grok with that. That's pretty sick.
- [04:22]
A
Oh yeah, he just paid for Grok for perfectly. That's why he wired before they actually.
- [04:27]
B
Yeah, he's like, I'm up. I got gains. I need to offset these. So Commerce Secretary Howard Lightnick has met repeatedly over the last year with high ranking Apple officials, including CEO Tim Cook as well as SpaceX chief Elon Musk, Nvidia chief Jensen Huang to try and convince them to get into business with Intel. There was a discussion about should Nvidia dual source the grace CPUs from Intel, SpaceX and Elon's been with Samsung for a lot of Tesla chips. But there's been discussion there. There was that rumor about global foundries and Elon's always been in the in and around the fab world and he's obviously going a lot deeper with the long term Terraform project. So Commerce Secretary Howard Lutnig was trying to convince them to get into business with intel, some of the people familiar with the matter said. With with the Apple deal, Intel is now signed partnerships with all three. So over the last decade, intel fell badly behind rivals from the Wall Street Journal such as TSMC and Samsung Electronics after a series of technical mishaps, leadership changes and failed attempts at consolidation led outside Foundry customers to pull or curb their businesses when intel hired Tan in March of 2025. Wow. Just a year ago, I guess a year and two months to replace ousted chief executive Pal Gelsinger, President Trump raised concerns that Tan's close ties with China would compromise him and called for his ouster. It was a very, very moment, but Tan won Trump over with a charm offensive. I like that. And the government announced its 10% investment in intel shortly after. Following the investment, intel shares price rose sharply on Friday morning, it rose 7.5% to an all time high of nearly $118 per share. It's up more now. Tan has been reshaping Intel's top leadership ranks in recent months, including hiring former TSMC executive We Gen Low, a move that prompted a lawsuit from tsmc so they're not friends anymore. Intel CEO has also also ousted his head of product and hired new executives to lead the company's data center, processor and client computing units, as well as newly formed custom silicon business. He's also invested heavily in Intel's most advanced manufacturing process, known as 14a. And that is the node that intel is hoping that all of the potential customers will jump together and say hey, we're going to buy from this. If you make it, if you build it, we will come. President Trump personally advocated for intel to cook in a meeting, which is funny because it sounds like he's advocating for intel just to cook generally, but he's no, he's advocating for intel to Tim Cook. In a meeting at the White House, according to people familiar with the matter, Trump said I like Intel. In January, he said that the government made tens of billions of dollars from intel deal and that the government's backing of the company had attracted important partners to Intel. As soon as we went in, Apple went in, Nvidia went in, a lot of smart people went in, trump said. Nvidia, the world's largest chip firm, invested 5 billion in intel in September and the two companies announced partnership under which intel would build custom data center CPUs, the processing brains for most computer systems. For Nvidia and Last month, Elon, of course announced the terrafab project. And so Apple relies on TSMC to make chips for iPhones, iPads, Macs and other devices and is under pressure to find additional chip suppliers. There's also chip shortages, so it could be good to dual source independent of the geopolitical discussion. On Apple's last two earnings conference calls, Cook blamed a lack of availability of advanced chips for Apple's inability to meet customer demands for iPhones. The constraints are expected to continue in the current quarter affecting several Mac models. Cook said, quote, we think looking forward that the Mac Mini and the Mac Studio may take several months to reach supply demand balance. And after the earnings call, Apple raised the Mac MINI starting price. And so TSMC's manufacturing capacities capabilities far surpass those of Samsung and Intel. Makers of other kinds of chips for memory and short and storage for example, are more competitive with one another, giving Apple multiple sources of supply, although of course they are memory constrained. Apple's been long, long been TSMC's top customer. But skyrocketing demand for its manufacturing capacity from Nvidia and other designers of AI chips means Apple no longer has as much leverage to secure the supplies that it needs. Starting in 2006, Apple used Intel designed CPUs as its main processor of its personal computers, but switched to its own custom CPUs based on ARM design in 2020. That's the dawn of Apple silicon. And so there's been an incredible economic and financial performance concentrated in just a handful of trillion dollar tech companies. Intel starting to join and starting to perform like that. There's a few memory stocks. I saw one report that referred to it as the Mag 10 and they'd added a few other AI names to that group. But there's just a few companies that are driving the vast majority returns in the stock market. There's a couple slides that we can pull up that show the scope of what they're calling the. It is now a bubble, everyone.
- [09:39]
A
The concentration.
- [09:40]
B
Yeah, the concentration.
- [09:41]
A
Yeah, we went through, we had three sort of beautiful weeks where pretty much everyone started saying there's no way it's a bubble. Yeah, look at the revenue growth. There's no way it's a bubble. But now people. Steve over at Bloomberg, we can pull this chart up since you mentioned it.
- [09:58]
B
Which one is this?
- [09:59]
A
This is the concentration.
- [10:02]
B
Sure. Yeah, I've seen this.
- [10:05]
A
The market, you can see the AI Big 10.
- [10:09]
B
So mag 7 AMD, Broadcom and one other. Who is this? Micron now make up 40% of the market. Railroads for reference during 1835-1910 were 63% of the market. And he comps it to Japan, the nifty 50 tech and telecom throughout various periods of time. And yeah, it's an interesting dynamic. And so I was reading this post by or this piece by Greg IP in the Wall Street Journal and he was trying to disentangle what's happening in the overall American economy from the AI economy. What's, where is the growth coming from? He sort of back the envelope did. He said the economy grew 31% while the non AI economy just 0.1%. And he cites a few economic statistics that are recent. Personal consumption, the biggest component of GDP grew a relatively muted 1.6%. Investment fell in housing, business structures. I think we're building, we're spending more on data centers than housing now or office buildings I guess. I think it eclipsed office buildings and factories and transportation equipment like trucks and aircrafts. Meanwhile, investment soared 43% in tech equipment. Obviously that's chips and GPUs. 23% in software. Surprising. And 22% had a pretty meaningful bounce back certainly on the revenue side.
- [11:39]
A
Datadog Atlassian there's been some acceleration.
- [11:43]
B
It really was extremely widespread where you had, I mean you had like doordash and companies with strong network effects where the software was not their moat. They still had to go and answer to the market. And some of them got in and out of that in week or a month. And some of them, it took a quarter or two to show that there's resiliency. Some of them are still beaten down, but overall software is still doing very well. And so on the flip side, there's another headline that hit The Journal yesterday. US adds 115,000 jobs in April with solid hiring across sectors. Retail, transportation, warehousing and healthcare offshored all showed strong growth and led to expectation beating jobs growth. And so this is the weird dynamic that, you know, you're seeing all of this growth in the, in the AI economy and yet the overall employment is still chugging along. We can pull up the chart from the Wall Street Journal, but the US job market blew past expectations in April, buoyed by gains in, in industries including retail, transportation, warehousing, health care. Not completely unexpected given that those are not particularly AI targeted areas. Whereas you might see, you know, the layoffs in the tech community, but that doesn't, it's just not enough because the US employs over 100 million people. And so a layoff of 4,000 people at some tech company just doesn't really move the needle when you're talking.
- [13:18]
A
Breaking news from Gabe in the chat. Texas Roadhouse is What? Why says 20%. I'm seeing 15% either.
- [13:25]
B
Why is it up so high?
- [13:27]
A
He says you're not bullish enough on Texas Roadhouse. I have never been.
- [13:31]
B
Maybe they beat. Yeah. Everyone says this is a Kiriakou filter. It really. It really is.
- [13:38]
A
It's crazy how. How viral that that meme format has gone. I thought that was just like a thing that I. I sent to you because I'd be dying laughing about it.
- [13:47]
B
Yeah, yeah. No, no. It's quite popular. Broken containment fully. So the American economy added 115,000 jobs in April. This was down from a net gain of 185,000 jobs in March, but it was much better than what expectations were for April. Analysts polled by the Wall Street Journal were expecting 55,000 jobs, and the real number came in more than twice that, which is like. It just breaks the narrative a lot. And it's this. There is this disconnect between, like, the AI economy and the real economy. I was talking to Sager about this. It's like. In some ways, it's like AI is Nvidia earnings is holding up the global economy, and it's holding up the global stock market. It's not actually holding up the economy. If you view the economy as all of the different jobs and activities that happen, even if there isn't incredible growth there, there's actually a surprising amount of strength and resiliency. So the unemployment rate.
- [14:47]
A
I'm waiting to hit the gong until we see what the revisions come in at.
- [14:54]
B
We haven't seen any revisions from March this time.
- [14:56]
D
I don't know.
- [14:56]
A
Maybe it comes like. It comes like. It comes, like, over six months later.
- [15:00]
B
Yeah. But that's why I wanted to have Josh from Gusto on today, because I think that Gusto has data on hiring that is very different from what the labor Department, what's going on at the labor department. It's just a completely different dynamic. And you can look at what's actually going on in small businesses. We've talked to folks at Stripe about this, and there are a lot of other data sources, ADP payrolls and stuff. There's plenty of pretty standardized formats for reporting. And while there might be revisions to some government stats, it feels like there's. There's plenty of places to get data, and it's all pointing in the same direction, which is not. Oh, yeah, they said they beat twice. They delivered twice as many jobs as were projected. And it was actually down no one's really suggesting that. So the April report, coming in after strong job gains in March, shows how the labor market is holding up better this year than last. While health care is still leading the way in job gains, other sectors now appear, appear to be picking up. Businesses are seeing conditions stabilizing and they have weathered the tariffs. So many are hiring, it's looking somewhat better than it did last year. In the first four months of the year, monthly payrolls have averaged 76,000, up from an average of 42,000 during the same period last year. Still, the US has shed jobs over the past six months, and many companies are reluctant to bring on new workers and masks. Companies grippling with changes in everything from trade to immigration to tax policy have for the most part held off on large scale hiring and firing. But it does seem like the chaos of the tariffs and all of the early eos, that creates a lot of business uncertainty, that settles eventually and then you get back to, okay, there were a bunch of changes, but at least we know that there aren't wild changes coming in the future, at least that are expected. Diane Swonk, chief economist at kpmg, says it's still a high anxiety job market. Those who have a job are clearly clinging on, while those who are looking for a job are feeling frozen out. And so I was digging into where is there strength within the real economy once you go outside of AI, what is going on? And there were two recent earnings reports that were sort of disconnected but showed a little bit of the picture of what's going on. So it was Whirlpool and Six Flags. Both had very different reports. So Whirlpool, you probably know from refrigeration and washing machines, dryers, that type of appliance. They've been making an appliance for over 100 years. And they've also been paying a dividend since the 1950s consistently. They've never suspended their dividend. Even through all the Great Recession, the dot com crash. Every year they've been able to pay that dividend. They just cut their dividend, which is a really, really big deal for this company since it's a dividend stock. The stock traded down on the news and the stock is down 80% of the past five years. And they're in some financial trouble. Like they have a lot of debt and they have a lot of competition. But you could play that as like, okay, that the real economy is completely chugging to a halt at the same time you had Six Flags, which should be the thing that is the Most discretionary, like do you go to the roller coaster theme park or not? And Six Flags just reported higher first quarter revenue. They're growing and they're growing attendance and customer spending. And so now Six Flags is not exactly a juggernaut of business. It's only worth 2.3 billion. And the stock is down over the the past two years about 50%. But the business is growing and you wouldn't expect that during a time of like deep economic weakness. And so there's something odd going on there where as you think of refrigeration as extremely necessary, roller coasters as the ultimate. Like you don't, you can definitely skip it if you are cash strapped. But the actual dynamics of the market are very different. So Whirlpool sells big ticket necessary necessities, these refrigerators, but they are deferable. So you can put off getting a new refrigerator, you can repair an old refrigerator if money is tight. But your kids are only really roller coaster age for a limited time. And Whirlpool also faces brutal global competition. And existing home sales are down 3% month over month. And so all of that drives fewer appliance upgrades. And they are not, and they are not necessarily a beneficiary of all the international competition that they face from LG and Samsung and other international players. And so I've sort of correction Six
- [19:40]
A
Flags or new information. Six Flags is. Has been up since, since around, around the, around November. It sort of bottomed. Travis, I remember I was thinking of this because one, they have the ticket ticker fun, which is fantastic. The stock has not been having fun, but Travis Kelsey and a group put in 200 million.
- [20:03]
B
That's right, we talked about that.
- [20:04]
A
The stock traded down pretty substantially. Yeah, I remember that after the investment. But it's basically recovered to where it
- [20:11]
B
was and well, it's still down over the past like five years. Yeah, two years ago it was like a $5 billion company down big. But what's interesting is that at this time of economic uncertainty and all these questions about hiring, questions about economic resilience, they are increasing revenues, increasing attendance, increasing customer spend, and that's what's driving the stock today. And so there's this. I was looking at these two companies and I was like, they sort of fit into this barbell thesis of the AI future which I've been seeing pop up more and more. And the two examples that I always give are, one, the Ellison family is both like long slop and long anti slop. They have a ton of infrastructure investments in AI through Oracle and then they own legacy media like Batman Superman through Warner Brothers. And then Josh Kushner is doing a similar thing with OpenAI and the San Francisco Giants. Like two completely opposite ends of the spectrum. And so you can think of roller coasters as potentially like anti slope because you can't vibe code Space Mountain. But it's interesting to sort of like dig into the weeds of what's going on in the global economy and the American economy and seeing like where the unsuspecting winners are. Ryan Peterson shared a post from the chart from the Financial Times that perfectly illustrates our possible futures. This is so funny that this, this is in the Financial Times, but they fully embraced what happens in the human extinction tech singularity and the end of scarcity tech singularity. In the bull case, real GDP per capita goes north of a million dollars and in the human extinction scenario it goes to zero, of course. But the AI boosted growth path is a steady trend upwards and that is the goal that I think everyone should be working for. Potentially the end of scarcity outcome as well. But it is fascinating and I think this chart that Ryan shares is sort of, I think why there's anxiety in the market because there's this like, you know, the whole like you're not prepared if it's not a bubble concept. But there's a dual anxiety where if AI gets too good, there's mass unemployment. Everyone's worried about that. But if AI is a bubble and it collapses, you go into a recession and everyone loses their job and you're in a similar scenario of economic anxiety for both the, not necessarily the true doomers. I mean obviously they face a similar opinion. But even if you're just like in the it's overheated camp, there's real risk to the stability and I think that's where a lot of the bubble concerns come from. Although we're certainly not seeing very many signs of the bubble. I mean, obviously the valuations are high, but so are the revenue growth charts. So we can go through some of the folks who are arguing.
- [23:20]
A
Are you more of a back of the envelope envelope guy or a napkin math guy?
- [23:23]
B
Oh, that's interesting. I have a napkin here from Wonderco from the event I brought him.
- [23:29]
A
How would you look at that little souvenir?
- [23:33]
B
I like a napkin. I like a napkin. Actually, now that I'm thinking about it, napkin math is insane. It's extremely hard to write on a napkin. It's way easier to write on the back of an envelope.
- [23:43]
A
Yeah, you got a lot of envelopes
- [23:44]
B
and you get a big envelope, you can get an envelope that eight and a half by. It's effective. It's effectively.
- [23:49]
A
An envelope is sophisticated spreadsheet. An envelope is kind of a core business utility. Napkins can be used for anything.
- [23:57]
B
If you're in napkin math, you gotta pivot.
- [23:59]
A
I will say I'm much more of a napkin math guy. I can't, I mean, just, just, just
- [24:05]
B
the phrase that you pull spiritually.
- [24:06]
A
Spiritually. And the phrase that I pull. But if I can't get excited about an opportunity based on what can fit on a napkin, then I'm never going to get excited about it. Right. There's usually like.
- [24:19]
B
So you think a napkin is, is definitionally smaller, it's a less amount.
- [24:24]
A
It's not, it's not, it's a, it's a lesser vehicle than an envelope.
- [24:30]
B
Yeah. Because a big envelope you could do a full spreadsheet on.
- [24:33]
A
Totally.
- [24:33]
B
You could, you could write out totally. And have multiple cells.
- [24:37]
A
No. So I'm, I'm, I. In practice, I'm a napkin math guy. But if something was really serious, I would probably train.
- [24:47]
B
Yeah, we've been doing too much napkin math around here. It's time to pull out the big guns. Let's get out the envelope.
- [24:52]
A
Speaking of envelopes, so this morning I was telling Tyler, I'm incredibly overwhelmed with slacks, emails, DMs on every platform. LinkedIn, DMs, Instagram, DMs, XDM's, all the different messaging apps. And I was thinking, how cool would it be if there was a service where you connected all your inboxes to, and then every day they would print out all of the different messages and then bring them to you. And you could put like a box. Yeah, you could put a box outside of your house and they would just put them in there. And then at a specific time, you know, maybe in a roughly like two hour window, you could go out and grab all the printed out messages and sort of, sort of like leaf through them, decide what you need to respond to. You would potentially, because you're only getting messages once a day, you would probably be a lot more intentional about what you wrote back and forth. Right. And I think that could be sort
- [25:52]
B
of the opposite of Earth Class mail those virtual mailboxes because they'll take your physical mail.
- [25:57]
A
Yeah, but I want everything. I want everything. You want everything in the box. I want to put a box outside of my house that people put. And I think there's something there.
- [26:05]
E
Yeah, Trump, I mean, Trump literally does this.
- [26:07]
B
He does that. Right.
- [26:08]
E
Gets emails printed out and then he'll. He'll write with a big marker his response and then his assistant will scan it and then email it back.
- [26:15]
B
Yeah.
- [26:16]
A
More, more on C. Fry says a napkin is more available than an envelope. Very true. Often the best business meetings were not scheduled as a business meeting.
- [26:28]
B
So there are envelopes around.
- [26:29]
A
There's no envelopes around, but there's plenty of napkins.
- [26:32]
E
Yeah. Also, I mean, you're saying napkins are smaller, but if you're at a restaurant and you have like a fabric napkin.
- [26:36]
B
Yeah.
- [26:36]
E
And you want. Those can be pretty big.
- [26:38]
B
So you're. But I feel like it's bad form to be at a restaurant. Like you're not supposed to ruin the napkins. Or do you carry a pen that has washable ink?
- [26:47]
E
You have the crayons for the kids. Right. People are.
- [26:50]
A
The chat is saying.
- [26:51]
B
They don't ask you to chat.
- [26:52]
A
Is saying, I could call my service they. The United States Post Office.
- [26:55]
B
Yeah. That's a good name.
- [26:56]
A
Thank you, Alex. That's a great name.
- [26:57]
B
That's sort of like the San Francisco Artificial Intelligence Company or something like that. Or the browser company of New York. It would be inspired by that. Yeah. I think that even with the.
- [27:10]
A
Uses crayon. Close fisted grip right on the table.
- [27:13]
B
Right on the tablecloth. Okay.
- [27:16]
F
Yeah.
- [27:16]
B
I feel like if you're given crayons, you are expected to maintain that the child uses the crayons only on the children's menu, which is typically made out of paper. Very disposable. If you see the child using the crayons on a cloth napkin, you are expected as a patron to intervene.
- [27:37]
A
Jackson says digitize this service and call it email.
- [27:41]
B
There we go.
- [27:42]
A
We gotta establish the mailbox first.
- [27:45]
B
I got an idea.
- [27:46]
A
But then we can go there. But I like where you're going.
- [27:49]
B
I got an idea. Well, there's some other posts about the markets. Justin Spitler says them be careful buying semis here. Obviously the market's very, very.
- [28:02]
A
Let's pull up this video.
- [28:03]
B
Overheated.
- [28:03]
A
Travis.
- [28:04]
B
Let's play this video. Oh, this is Travis Pastrana.
- [28:08]
A
Yes.
- [28:09]
B
That's amazing. Nitro Circus we mentioned many times on the the show. Is he drinking a Red Bull?
- [28:14]
A
What a. Yeah.
- [28:15]
E
No parachute.
- [28:16]
B
No. Oh, no parachute.
- [28:18]
A
Yeah. So he jumps out of the plane, no parachute, drinking a Red Bull.
- [28:21]
B
That's actually.
- [28:22]
A
And then he connects. He's got to do some tricks.
- [28:25]
B
He's really taking his time here. The tension is building, building, building. And so he connects with someone who wraps him up and then do they pull the parachute to each other or do they.
- [28:36]
A
They. They. They strap him to them.
- [28:39]
B
Okay, there is a strap. Strap themselves because it seems sort of crazy just to bear hug and hope for the best.
- [28:43]
A
I don't think you're bear hugging.
- [28:46]
B
It's Red Bull. Anything could happen. I would expect him to be delivered a full parachute. His own parachute that he then dons. But he made it. Wow. Well, that's what it's like investing in semiconductors right now. I guess. Anyway, how was your last two days, Tyler? What'd you get up to?
- [29:06]
E
It was sick. I was at the. I went to the trial.
- [29:09]
B
You went to the trial?
- [29:10]
E
OpenAI Elon trial.
- [29:11]
B
Okay, walk me through it. You left the studio on Tuesday?
- [29:15]
E
Yes.
- [29:16]
A
Way before this. The chat is still going. They said if Gary Tan gets a hold of this service, he's going to call it Gmail.
- [29:22]
B
Gmail is good.
- [29:23]
A
Something there.
- [29:24]
B
Gmail. If he vibe codes the full Gmail replacement, anything's possible. So you leave the office at 2 after we wrap the show on Tuesday. You went straight to the airport.
- [29:37]
A
I did.
- [29:37]
B
Okay.
- [29:38]
E
Yep.
- [29:38]
B
And then what time did you go to bed? Cause you woke up really early, right?
- [29:42]
E
Yeah. So I got to the trial. So it's in Oakland. I got to the courthouse at I think like something around like 5:30.
- [29:48]
B
5:30.
- [29:49]
G
Yeah.
- [29:49]
E
Cause you gotta get in line. So basically, you know, it's a public. Cause it's like a federal case.
- [29:54]
B
Yeah, yeah.
- [29:55]
E
There like has to be some room for the public. But there's only like. I think the number is somewhere like 20 or 30. Because I think it's closer to 20 because I think part of the 30 is reserved just for media.
- [30:06]
B
Okay. And when you got there at 530, were there already people standing in line?
- [30:10]
E
Yeah, so they were. I think there were like three or four people there already.
- [30:13]
B
Okay.
- [30:14]
F
In line.
- [30:14]
B
Early bird getting the worm.
- [30:16]
E
But. Yeah, well, because I didn't want to be late because. Because. Did you talk to my guys? That people, a few of them.
- [30:22]
A
Were you there before Mike Isaac?
- [30:24]
E
I was there before Mike Isaac, but
- [30:25]
B
he is a media pass, so.
- [30:26]
E
Yeah, because last time when Mike Isaac came on the show, he said that because each like, news.
- [30:32]
B
Yeah. New York Times got one pass and he was splitting it with Cade Metz. And so they were sort of going back and forth and I guess he had a media pass that day.
- [30:40]
E
Yeah, so that day he had a media pass. He got there pretty late. He got there like 7:30 or 8.
- [30:44]
B
Okay, so you're there from 5:30. He gets there at 8. What time does the trial actually start?
- [30:50]
E
I think around 8:30.
- [30:52]
B
8:30. Okay.
- [30:53]
A
Yeah.
- [30:53]
E
So basically I'm just posting up for like three hours.
- [30:55]
B
Okay. And then this is Wednesday. So walk this through. Like what, what are you seeing? There's these opening remarks from the judge. They sort of welcome everyone in.
- [31:03]
A
Yeah.
- [31:04]
E
Jury comes in.
- [31:04]
B
Okay.
- [31:05]
E
So there are basically three main, like, segments. So basically I'll say it starts 8:30, ends at I think 2:00pm there's two breaks, two 20 minute breaks.
- [31:15]
B
Okay.
- [31:17]
E
So starts off and we watch like the deposition of Miramorati. So this is all live streamed, right? The audio is live.
- [31:23]
B
The audio is live. So you're seeing a video?
- [31:25]
E
We're watching the video of her deposition.
- [31:27]
B
Okay.
- [31:28]
E
It's mostly covering the ousting, the timeline of the ousting. And this is where all the texts came out?
- [31:35]
F
Yeah, yeah.
- [31:37]
B
Were they showing you the actual text? Were they playing a video or were they playing the AI reenactment of the text? Because we have this. Someone turned Sam Altman's text, Amirah moradi, into a 2011 style emo teenage heartthrob anthem. Did this make it into the courtroom or no? Let's play this. We have the audio.
- [32:00]
D
Can you indicate directionally good or bad?
- [32:10]
B
Good. Motion graphics too. Whoever made this, talented. What is this Suno? Can you put in text into Suno
- [32:27]
A
and get a show? Let's play this for an AI skeptic.
- [32:29]
B
Yeah.
- [32:30]
A
And see what their reaction is.
- [32:32]
B
Because how do you see this and not want to build? I've seen these before and I've always really enjoyed them. But with most of the AI music tools, I've never found a great way to put in a full script that I've written to get the results, but does happen.
- [32:50]
E
So the author said his steps to make this. First he OCR the images and turn them into plain text. So images of the actual text messages and then remove the names from the dialogue pasted into Suno and iterated like 20 to 30 times.
- [33:02]
G
Ooh.
- [33:02]
E
And then finally. Thought this one was catchy.
- [33:04]
B
Okay.
- [33:05]
E
Yeah. So I think it's still like this is still Suno, but I think you
- [33:08]
B
still have to iterate about it. I like this. I like this. Yash is really having fun with this. Put up a version of this with the. With the Gen Z brain rot slang. Yo, fam, can you give me a vibe check and am I cooked or Nah. For real? For for real. Satya and the Gang Low key stressing murmurati. Yeah, you're skibidi. This isn't a rare L. This is an Ohio level generational or a loss little bro. Who ratioed me? Sam Altman says Miramarati, some rando, zesty NPC twitch looks Maxing Sigma. I'm not even gonna go further, but Yash is having fun. Lots of people having fun. What else happened? Did someone actually get on the stand or was it all video?
- [33:50]
A
Yes.
- [33:51]
B
So the first, like, it's straight up teacher, played a movie. It's movie day. Went to the movie day.
- [33:55]
E
The judge actually did get angry because, like, she was like, oh, the jury's gonna get bored. And like, she was, like, offering them coffee and stuff.
- [34:01]
B
Okay.
- [34:01]
E
So basically, yeah, the first, like, maybe hour was mirror deposition video. And then we go through some of the actual, like, documents again of the text messages. And then Siobhan Zillis is testifying. So she's physically there.
- [34:14]
B
Okay.
- [34:14]
E
She was a open board member 2020 to 2023. And she steps down when Elon started XAI.
- [34:21]
B
Yep.
- [34:22]
E
So that was. I have no idea how long that was. That was the majority of the.
- [34:26]
B
During the whole conflict.
- [34:28]
A
Yeah. Perfect person testify.
- [34:30]
D
Well, no.
- [34:30]
E
So she actually left the board before the ousting. So she left in February 2023. Ousting is November.
- [34:38]
B
Oh, okay. Interesting. I didn't. I. The timeline always gets so jumbled up because, like, there's like, 2018 battles between Sam and Greg and Ilya and Elon, and then there's the board.
- [34:48]
A
Yeah.
- [34:49]
E
Because all throughout this hearing, there's like.
- [34:51]
A
Yeah.
- [34:51]
B
There's basically two main moments. Right. Yeah. And we're sort of clicking back and forth between them.
- [34:57]
A
Yeah. Chat wants to know what kind of wombos were being thrown around. Was Lorraine being used? People saying, can you please, Lorraine?
- [35:05]
E
This email, Siobhan was actually dropping some good lingo.
- [35:10]
B
I don't know if you remember anything. Like, Mike Isaac always talks about, like, they keep referencing Dota and they keep. Everyone on the stand is using a ton of jargon. How jargony was it? Was it like Dwarkesh Patel podcast level jargon or like actual, like, researchers, like, talking to each other jargon?
- [35:29]
E
I would say it was not very sophisticated, especially among the lawyers. Right.
- [35:33]
C
Okay.
- [35:34]
E
Yes, Siobhan, she did drop some good lingo. Okay. Yeah, it was some referencing, some, like, why she got into AI, but, you know, acceleration, all this.
- [35:43]
A
Okay.
- [35:43]
B
Okay.
- [35:44]
E
And then after that, we watched another, like, hour long video of Helen Toner's deposition.
- [35:49]
B
Wow.
- [35:49]
E
That was just about the ousting.
- [35:51]
B
Okay.
- [35:51]
E
Yeah, it was very fun. I enjoyed a lot. Like, basically, you know, almost Everyone, like, sitting in my section, which was basically the public and the media.
- [35:59]
B
Yeah.
- [36:00]
E
Like, everyone that came as the public were basically also media. They get the media pass.
- [36:05]
B
Sure.
- [36:05]
E
So everyone surrounding me is on their laptop, like, typing, basically.
- [36:09]
A
Yeah.
- [36:09]
E
Just like, taking down what's being said.
- [36:11]
B
Yeah.
- [36:11]
E
And then I'm just sitting there enjoying it because it's like, you know, it's good stuff.
- [36:15]
A
How would you rate Mike Isaac's, like, snack and overall food supplies for the day? Did he burn it? Looks pretty good.
- [36:21]
E
I believe he did bring the butt pillow.
- [36:23]
B
It seems like he's not making progress on the food. Food front. I would expect by day seven of this, he would have a smorgasbord in front of him. Full Thanksgiving.
- [36:31]
A
I mean, honestly, like, Nathan, for you, where he's pressed chili into the. He's got the chili. The chili soup.
- [36:39]
B
You should bring chili soup.
- [36:40]
E
I was. I fasted the whole time. I just kind of. I just knock it out.
- [36:43]
B
Why didn't you figure it? We have talked about the food situation so many times. How did you not think ahead there?
- [36:48]
E
I just wanted to make it more challenging.
- [36:49]
B
Okay.
- [36:50]
E
Yeah. I was like, this is gonna be too easy.
- [36:51]
B
Inspire you.
- [36:52]
A
Should we play this chili. The chili man, if you have it,
- [36:55]
B
we can also play. What's the other video? The. The. The Hamilton musical has been adapted. People are. Are taking this in all different ways. Has anyone done a. I have a video here. A Korn remix of it.
- [37:10]
E
The Hamilton one is new metal.
- [37:12]
B
Let's play the. Let's play the. The Hamilton one. Altman, an AI music experiment from.
- [37:18]
A
Well, let's. Let's. Let's start with chili, because I think the trial is going for another couple weeks.
- [37:23]
G
So later that week, I hired a
- [37:25]
B
professional tailor to help design the chili suit based on a detailed sketch I had drawn.
- [37:30]
A
The easiest way is probably to put the chili in.
- [37:35]
B
Tag my guys in. This is how he can make it through. The plan was to have the chili reservoirs mimic the distribution of fat on a larger man's body so it wouldn't look suspicious under clothing. And with the help. Why is he bringing chili into. Is he going into a.
- [37:50]
A
No, he's helping a local business sell chili. Sell chili in the stadium.
- [37:54]
B
Illegally in the stadium. That's right.
- [37:57]
D
You could probably do, like a cup there, right?
- [37:59]
C
Yeah, Cup.
- [38:00]
A
Skip ahead, skip ahead.
- [38:02]
B
Hold over a hundred servings of fresh chili. 100 servings. Individual reservoirs would be linked by a network of two tubes that would all lead to a battery powered peristaltic pump for dispensing.
- [38:13]
A
If Somebody doesn't do that. The trial is going for another couple weeks, I think.
- [38:17]
B
I think you might. I think the bailiff might actually just send you straight to jail if you did this. I think the judge.
- [38:22]
E
There were surprisingly little hijinks going on. Yeah, I thought there would be way more hijinks. Like, you know, like no one dressed up in a costume or anything.
- [38:28]
B
Yeah. Wait, were there Elon.
- [38:30]
A
I mean, it actually is quite possible someone is doing this and we just don't know because it's such a good idea. And they're.
- [38:36]
B
They might just be going. Chili, were there Elon fans? That was something Mike Isaac was talking about. Like, people. Oh, I guess Elon's not on the stands. So, like, why would the fans go? Right.
- [38:45]
E
Yeah, I don't think there are fans. He was not even there.
- [38:48]
B
Okay. Yeah, yeah.
- [38:50]
E
So. So no one there. I think that there's going to be more testimonials soon with. With some higher named, you know, higher profile is.
- [38:57]
B
Yeah, I think Ilya getting on the stand. So that'll be a big moment.
- [39:00]
E
Yeah, I believe Ilya is early next week. Monday or Tuesday.
- [39:03]
B
That'd be interesting. And did you go two days in a row or just.
- [39:05]
E
I just went one day because Wednesday the people testifying, I think, were like, not gonna be as interesting, but maybe that's a dis. I apologize.
- [39:14]
B
I think everyone's. Everyone's equally deserving of some attention, especially if there's a nice warm bowl of chili being served in the courtroom. Anyway, should we play this? Altman AI music experiment from Daniel Green. The Hamilton recreation of the texts. Everyone has enumerated their various techniques.
- [39:38]
A
Ryan says, john, you should go to the trial dressed as Ilya.
- [39:41]
B
Ooh, yeah. I do have the costume ready to rock.
- [39:43]
A
It only takes 5 hours to set
- [39:46]
B
up, get up at 2am then go stand in line. I think I'd get kicked out immediately anyway. Let's play this. I want to hear this.
- [39:54]
A
How does a startup founder, late stage,
- [39:58]
D
get fired by a board on a Sunday?
- [40:02]
B
Oh, so it's actually like, can you
- [40:03]
F
please officially invite me to the office for a meeting?
- [40:07]
B
So it's not the direct quotes or text.
- [40:09]
F
Agree to a configuration.
- [40:11]
B
Or maybe it is.
- [40:12]
A
He is now saying they need till
- [40:14]
G
end of day Satya. And I said, that doesn't work and
- [40:18]
A
we need to start preparing for.
- [40:20]
B
This is much more Uncanny Valley for me. This doesn't sound like a real musical in the same way that the pop punk one sounded pretty accurate. I don't know. Anyway, we can move on to Some other news, Deep Seek is raising a monster 7 billion dollar round at 50 billion valuation, making it China's largest ever AI raise. But what shocks a jaws here Cryptopunk the most is the the founder. He's personally contributing 40% of the round himself. Wow. 3 billion coming from the founder. Directly owns 90% of the company. Unheard of at this valuation. Deep Seq was founded inside of his hedge fund, one of China's most successful funds. What a beast. He's got to acquire as much compute to push out new Deep Seq models. You know, we saw that chart that showed that Chinese open source models were sort of falling behind a little bit on a different growth curve in terms of performance. But he's certainly betting on getting back in line having a Frontier model within a couple months. Tess has an interesting idea. If I was running a Frontier lab, I would have the model versions count down. Ominous, Extremely ominous. Be like, yes, we're excited. Version three now, next year version two. Somebody would do that. Well, folks over on Amazon are reviewing the Art of War by One Star.
- [41:48]
A
Nothing but common sense. How this became a classic is beyond me. So much of it is common sense to the point where it brought out my snarky no kidding reflex. Like if an enemy leaves a door open, you must rush in. Wow, that was deep. Keep your money if you're thinking of buying it.
- [42:06]
B
That's very funny. Leaving a bad review so my enemies don't read it. Yeah, that's the game theory. Zach Brock says, congrats to Anthropic for defeating Grok in the market and feasting upon the compute of their fallen enemy.
- [42:21]
A
Yeah, basically every time we take, every time we take a day off the show, like something big happens. And that on Wednesday was the, was the big revolution or space x Anthropic deal. Yeah, a lot of people have been predicting that. I, I was not simply because I thought there it was like the rational decision for the parties, but I thought that the tension between, you know, Elon, who had only a couple months ago been, you know, hurling insults at the Anthropic team. I didn't think they would be able to uncover those cultural differences.
- [42:56]
B
Totally.
- [42:57]
A
But demand for compute finds a way.
- [43:01]
B
Yeah, I think that you and others had identified the possibility of becoming a Neo cloud, selling the compute.
- [43:07]
A
Yeah. Last year I was talking about that a lot, all the time. I was like, look, they're incredible at building infrastructure really, really fast, bringing power online. This feels like a very strong, a strong use for Elon Inc. Yes, and, but I, but, but as things evolved, I just didn't see, I didn't see this coming together.
- [43:29]
B
I didn't see this coming together specifically because of the cursor deal. I thought the cursor deal was the long term solution for all that compute and then the compute sort of got sold twice maybe. But of course there are multiple clusters, multiple colossus data centers.
- [43:42]
A
Yeah.
- [43:42]
B
And plenty of work to be done as SpaceX continues to grow their ambitions. Artificial intelligence. Well, we have our next guest, our first guest of the show, Josh Reeves from Gusto in the waiting room. Let's. How are you doing?
- [43:58]
D
Good to see you guys.
- [43:59]
B
Welcome back to the show. Always great to have you.
- [44:02]
D
Good to be here. It's a fun time to talk tech. It's a lot of big moves happening, huh?
- [44:06]
B
A ton of big moves. Yeah.
- [44:09]
A
In the last year we've had a decade's worth of big moves, progress.
- [44:16]
D
There was a time when Silicon Valley was a cottage industry. Not anymore.
- [44:19]
B
Yeah, a garage industry. And now the largest companies in the world for sure. Well, your business is growing. You're quickly becoming one of the largest businesses in the world. And you also power tons and tons of American businesses. Give us the update. How are things going?
- [44:36]
D
Yeah, excited to be here. I think probably what you're alluding to, we shared a milestone. To me it's all a byproduct of helping customers, helping small businesses, but, but yeah, we passed actually a few months back $1 billion of trailing revenue, so.
- [44:53]
A
Whoa, did you hear that, John? Trailing. That's not something you see very often. People want to take today's revenue, multiply it by365 and lead with that.
- [45:04]
D
Obviously talk about there are as is much higher. But yeah, trailing.
- [45:07]
A
Yeah, yeah, talk, talk about that, talk about that decision.
- [45:12]
C
Yeah.
- [45:12]
D
I think for us, I mean we, we when we started this company now over a decade ago, you know, I, I said you always start with a problem, something you need to go fix. You have to create a solution, actually show it works. But then you have to have a, a business model where your unit economics, how much it costs you to serve them, how much it costs you to acquire them. Makes sense. Otherwise it's kind of like gambling. So at Gusto, you know, we've been free cash flow positive too for a number of years. We reinvest that money to build more product, but we knew the potential was there to be at this milestone and frankly much, much further. There are 6 million employers in America and two thirds of them are less than five employees. Small businesses are a huge part of the economy. So yeah, we're excited to beat this milestone, but much, much more work ahead.
- [45:56]
B
So cash flow, positive reporting, trailing 12 month revenues instead of ARR. These all feel like real strengths to eventually go public around. At the same time Monday.
- [46:08]
A
You should report ARR. Come back on Monday.
- [46:11]
B
Yeah, but at the same time, like it's an incredibly tumultuous time in the public markets. There's winners and new favorites every day. Everyone sasspocalypse narratives like how are you thinking about the benefits of being private right now versus potential future in the public markets?
- [46:27]
D
Yeah, I mean our focus is on just execution. I would not expect Gusto to be private in the near term. We'll let all the public. Do you mean public folks, Sorry, public in the near term? Yeah, correct.
- [46:38]
B
What are you announcing here? No, I would not expect a sustained revenue.
- [46:43]
D
I will leave that all to like the SpaceX and anthropics and OpenAI's in the world. There's lots of good noise in that. I would say we're going through this massive paradigm shift in how AI affects not just our product, for example, and how we help small business, but also how a company's run, how a company is built. And so that is where I'm spending all of my time. It's an exciting time to be spending focus there. It actually is a net. Net pretty big tailwind in gusto. But absolutely no plans to be public in the near term. Super focused on small businesses and execution and at some point timing will make sense. But yeah, we haven't had any of that process get started.
- [47:22]
A
We covered the April jobs report this morning. It surprised a lot of people to the upside. What are you guys seeing in your data? You guys basically have a forward view, at least on the.
- [47:38]
B
Yeah. To set the table, the Wall Street Journal had pulled analysts expectations were at 55,000 new jobs and the number came in at 115,000 after a very strong march as well. And so things are. Although there's so much anxiety, so much uncertainty about the future of the American economy, at least over the last few months things have been looking strong, but we've seen revisions before. I think there's a lot of questions about, as you dig into that data, what's going on. And so we'd love to know what you're seeing internally.
- [48:13]
D
Yeah, so two stats we look at quite a bit. There's the kind of state of new businesses being formed, new employers. Obviously companies can exist before having an employee and then there's the state of hiring across the base of Customers, I would say, you know, on the customer small business hiring front, things are more depressed over the last few years and that's continued relative to call it the last decade if you will. We think there's many parts of that, including some of the effect of AI on frankly company growth rates. On the flip side, we are seeing a lot more new businesses being formed and new employers out there and that's showing up in our data. That's showing up in a bunch of third party data. We think that's a really good thing. There's a scenario with what's happening in AI, we call it the happy path where a lot of people decide to take the plunge and become business owners. The non happy path to me is the Pixar movie Wall E which is a feature I hope we avoid.
- [49:14]
B
Yeah. How have you been thinking about getting to customers and small businesses earlier? Have you ever explored trying to meet customers before they incorporate, help them with incorporation, Anything that gets you into the system. His Gusto, I feel like is always one of the first tools, but it sometimes happens like a month after incorporating.
- [49:39]
A
Well, you have, I mean one really powerful dynamic is you're paying millions of employees, many of whom will go and leave companies. Like, I had a good experience on Gusto. How should I pay myself? I'll use Gusto.
- [49:53]
D
Yeah. I mean first off, the biggest way we grow is word of mouth and often it is from not just the employer, but the employee. Also we'll have people join Gusto and ask how you heard of us. And they're like, I was paid through Gusto. So we don't try to recruit from our customers, but sometimes it happens and it's a wonderful dynamic where they truly, truly get the value of the product. To answer your question directly, absolutely. Like we, we think payroll is one of the best products out there. When you're hiring someone, obviously if you don't pay them, they quit. So it's a really important front door for that interaction. But there's a lot of companies that exist as one person, the founder, for some period of time. And so we have an offering we call Gusto Solo that's very focused on that. Turns out the founder still needs to pay themselves. There's specific tax dynamics at play. There's a lot of compliance requirements. We acquired Mozy to help with expanding our work around business compliance. So all those kind of headache compliance things that you still have to do as a one person company, Gusto is to going to be increasingly a great, great partner to help with and then when that Business owner decides to hire an employee, obviously we hope they continue their relationship with Gusto, but stay tuned. We're going to be launching some more products in the next several months. Really focused on building that relationship more deeply with employers or even businesses before they become employers.
- [51:06]
B
Yeah.
- [51:07]
A
Do you think the. Are you expecting the velocity of new product releases from Gusto to go up in order of magnitude in the next few years, or do you see it as a smaller family of products with a bunch of and higher velocity of new feature creation?
- [51:24]
D
Yeah, a few years. Give me the next few months. It's fun to be accelerating. I think of it as this is the most amazing time to build people. Can smaller teams self serve, iterate faster, maintain high quality, but just literally velocity is incredible, increasing dramatically. So we apply that to both breadth and depth, to answer your question. So within the world of what we do, whether it's 401k, health benefits, payroll, people are getting paid when they want, how they want, where they want, faster. We're excited to be doing a lot more work there, but absolutely expanding the breadth of the product. It's still way too hard to be a small business owner. So you will see a lot more new product launches coming from Gusto in the coming several months. Things that we may maybe thought of as like the next two to three years, I now think of as the next three to six months. It's really fun to say that and actually then back it up and see it happening.
- [52:15]
B
One question that I've been having as we've seen LLM usage exploding is the shape of LLM spend across a software engineer who fires off a coding agent, builds something that's new, but it is deterministic code that runs in perpetuity. New features, new Designs, new copy vs. LLMs that are baked into an existing work stream where every time payroll runs, there is a trigger. And it is not something that's fired off by an employee. It is. And I don't know if one of those is more agentic or we need a different word for those. But can you kind of characterize the shape of those two applications for AI right now?
- [53:01]
D
Yeah, I mean, maybe a couple cuts on that. I think new product development generally more accelerated, dramatically higher velocity with AI. If you have an existing product, existing code base, there's just by definition more the dynamic of building the train, rebuilding the train, running the train at the same time, accelerating the train, laying the tracks, kind of all in parallel. So I think it's more complex on existing product surface area. So that's One cut. I think on the kind of cut around the ongoing interaction dynamic, I guess we've always believed, since we're serving a small business, that what we do, it's mostly compliance centric. It's just better done by technology. Right. You think about the potential for human error, whether it's from the customer or a guestie. We want to have as much of that as possible, be into software, into a system that can scale, that can do it repeatedly, that can do it in millions and millions of situations. To your point, some of the newer technology is obviously not fully deterministic. So then having the right evals, having the right feedback loops, having the right review and audit process is super, super critical. So as we do all this velocity acceleration, you can imagine there's a lot of focus at gusto on not just maintaining, but actually improving the quality of what we do and the accuracy and the reliability for our customers. And it does feel like it can have a win, win, win outcome here, where the product gets easier to use, it gets even more accurate and reliable, and actually we can build and ship it faster.
- [54:26]
B
Yeah. Do you have a feeling for which bucket is bigger? Because I can imagine LLMs being very useful to scan every PDF and do translation and do it, do things on an ongoing basis, but baked into the systems that you've built versus augmentation and amplification of engineering efforts and design efforts. Do you have an idea of, in terms of spend, which one's bigger or where it goes?
- [54:55]
D
Yeah, I mean, I would say for us, most of the ongoing systems and how they run, if you're talking about the backend, don't generally involve a lot of LLM technology.
- [55:05]
B
And then those are probably efficient too. Right. Because even if you do use an LLM to process a PDF, well, you can pretty efficiently get to some sort of small language model that's tailored and probably pretty cheap.
- [55:20]
D
Yeah. And that's a world where you want accuracy, reliability, and it's pretty deterministic. I think if you look at risk, if you look at things where you have to make more judgment, that's where I think LLMs can actually complement humans quite a bit.
- [55:33]
B
Yeah, that makes a lot of sense. Jordy, anything else?
- [55:36]
A
Incredible update.
- [55:37]
B
Yeah, congratulations.
- [55:38]
A
The focus, everything is incredible. The intentionality, the humbleness of looking backwards versus forwards. I love it all.
- [55:48]
B
I love it.
- [55:48]
A
Great to catch up, Josh.
- [55:50]
B
Well, thank you so much for taking the time to come on the show. We'll talk to you soon.
- [55:52]
D
Yeah, my pleasure.
- [55:53]
B
Have a good one.
- [55:54]
D
Thank you.
- [55:54]
B
Goodbye. And up next, we Have Jonathan Niemann from Sweet Green live with us in
- [56:00]
F
the TVTN Ultra Dome.
- [56:02]
B
Welcome to the show. How you doing? What'd you bring us today? What'd you bring us today? You brought lunch.
- [56:08]
C
Brought you lunch.
- [56:09]
B
Fantastic. What is it? What are you launching? Give us the updates.
- [56:12]
C
So this week we launched probably our biggest new category in. In company history. We launched wraps, but we've done in in our own way. So we spent about a year designing a custom tortilla with only 4 ingredients. Most wraps in the market, about 22 ingredients full of seed oils, preservatives, etc. So really wanted.
- [56:31]
A
Yeah, I had a devastating moment yesterday where I was eating a box of crackers, like the sort of default cracker in my household. And they, they updated the ingredient list without, without me realizing because, like when I bought them, they were seed oil free. And then some point in the last, like two years, they added it until I looked down at the ingredients list
- [56:52]
C
and I couldn't pronounce.
- [56:52]
A
I was like, no, I just walked over to the trash and. Devastating. But anyways, continue. Four ingredients.
- [57:01]
C
Yeah. So we created a tortilla, four ingredients, and launch wraps this week. So we launched with four signature wraps. Anything can be customized. The response so far has been awesome. We launched it on Wednesday. We brought you three of them to try. If you guys want to give them a look, give them a taste.
- [57:17]
B
Fantastic.
- [57:18]
A
On air taste.
- [57:19]
C
On air taste.
- [57:19]
A
Yeah, we'll give it a try.
- [57:24]
B
Was this customer pull? Was there demand for this? How do you measure that? Now's the right time to launch a new product.
- [57:31]
C
Yeah. So we wanted to reach kind of different customers in different occasions. And one of the things we heard from our guests is they wanted something that was portable. They also wanted food that was healthy but a little bit more craveable. And so this is one way to get.
- [57:46]
A
Yeah, there's like a stated verse, revealed preference.
- [57:48]
C
Yeah.
- [57:48]
A
Just like you want to eat healthy, maybe you want to eat a salad, but sometimes you just need something that feels like a bit more dense.
- [57:55]
C
Exactly.
- [57:56]
B
And how does that feedback actually get to you in practice? Is this like interactions with cashiers and they're taking notes, or are you sending out surveys afterwards, polling customers, doing like, you know, some in store marketing that someone's standing there and they're asking people about their experiences?
- [58:15]
C
Yeah, we do a lot of things in order to get feedback. One is, let's see which one you got there. This is the chicken jalapeno ranch dipping sauce on the side.
- [58:24]
B
What else we got here?
- [58:28]
A
Fantastic. We never Get a lunch break.
- [58:30]
B
We don't. We don't. This is Chicken Caesar.
- [58:33]
C
Chicken Caesar. And then you got a Cali chicken club.
- [58:36]
B
Fantastic.
- [58:36]
C
So a lot of ways, you know, one thing I say all the time is all the restaurants, all the answers are in the restaurant. Spending time in the restaurant, actually talking to guests. We also do a lot of, you know, traditional things like focus groups, etc. Recently.
- [58:50]
A
Do you ever go undercover?
- [58:51]
C
I do go undercover a lot.
- [58:52]
B
Like behind the counter.
- [58:53]
C
Oh, all the time.
- [58:54]
B
All the time.
- [58:56]
A
Wear a mustache.
- [58:57]
C
I spend about, you know, a third, probably like a quarter of my time in restaurants talking to customers, talking to team members. I get all, you know, really, I get all my best feedback there. Both how to make the experience better for guests, but also how do you make it easier to run from a. Easier to run from a team member perspective? We do a lot of social listening from a feedback perspective. And recently, about a year ago, we started working with a company I think you featured here, Listen Labs.
- [59:21]
A
Oh, yeah.
- [59:22]
F
No way.
- [59:22]
C
We were one of their first customers and it's, it's helped us accelerate our time to feedback by about 10x and about half the price. And so, you know, this is actually one of the first things we worked.
- [59:35]
A
It's been a while since we've had Listen labs on, but basically there's like a. Like, what are the core features of the product? Again, it's social listening, but then also like surveying.
- [59:46]
C
It's. It's surveying focus groups and insights. So typically when you do a focus group, it takes a long, really a long time to get it set up and it can take weeks to get that feedback. With Listen labs, we can get a diverse cut of customers by demo age, psychographic group geography.
- [60:05]
A
I feel so rude.
- [60:06]
C
No, come on. And so that's been a really powerful way as well. And then social listening is huge. Just seeing what people are saying online. And then you look at the competitive landscape and where the business need is. You know, we do really well. Our business does really well Monday through Thursday, you know, when people are looking to be healthy. And then you see it on Friday calls off and Saturday. Yeah, so it's. How do we have that cheat meal that you could still feel good about?
- [60:33]
B
Oh, sure, sure. Where in social listening, like, where is the sweetgreen community? Where are people actually having conversations that are valuable?
- [60:40]
C
TikTok and Instagram.
- [60:41]
B
TikTok and Instagram.
- [60:42]
A
TikTok.
- [60:42]
C
I mean, and not Reddit. Reddit as well. You'd be surprised how. How much juicy information is on Reddit.
- [60:50]
B
Yeah, so social listening, how manual is that versus like scrape TikTok and transcribe everything and roll it up.
- [60:59]
C
It's not manual at all anymore. Okay, so a lot of software.
- [61:03]
B
Interesting.
- [61:03]
C
Yeah.
- [61:04]
B
Huh.
- [61:05]
C
There's a lot of companies that are off offering this now. We've tested a bunch of them.
- [61:09]
B
Sure, sure, sure. And is it valuable to find stuff that's not viral but still on TikTok like the TikTok creator or just like person that expresses themselves through TikToks or vertical video to loosely. Just their friend group. Like they don't have a following, they're not a professional creator, they're not making any money on it. But like when, like, like someone would post a photo on Instagram, they post a friend facing TikTok. Is that as valuable as the big creator that goes viral for some like dunk or stunt or tay hot take or something?
- [61:41]
C
You want to find it before it goes viral. You know, like there's this thing going on viral on TikTok right now about this like chimichurri steak bowl, like the Jim Bro bowl. And you know, we should have done something with that, but we should have done it a month ago.
- [61:54]
B
What is there to do with something like that? Like, I feel like a lot of times if there's something that's good, that's organic, that's happening, like and you step in as the brand or the corporation, like you risk like making it uncool in corporate.
- [62:04]
C
Well, for us, if it's something we could actually serve, like that happened to be something that we can put on our menu.
- [62:09]
B
Oh, sure, sure. Okay. So you make a change on your world. You're not trying to go and like your own content around it. Let them do it.
- [62:15]
C
We would create the content and then I serve it.
- [62:17]
B
Yeah, yeah.
- [62:17]
C
So we put a video out this week on it to see what happens and the reaction's good. We'll actually put that on the menu. We have all the ingredients.
- [62:24]
B
Yeah, it's a fact.
- [62:25]
C
Bowl without lettuce or any grains. It's just steak. John's kind of John's steak. Cucumber, tomato, onion. You know,
- [62:36]
B
I've seen random videos of focus groups. Do they really do a one way mirror or two way mirror, whatever that is, where like they can't see you, but you can see them. Is that a thing? Have you ever built any one of those?
- [62:48]
C
When I do them, I actually like to sit in.
- [62:50]
B
Okay. So it's just sitting in a conference room.
- [62:52]
C
Just sit in a conference room.
- [62:52]
A
Okay.
- [62:53]
C
When we, when we tested Wraps we brought in. We brought in customers every day for a week, different groups of customers, and just tasted them. Tasted them a few different ways, really, to understand, you know, how important was the sauce on the side, how important was it to get cut? Like, little details. Like, for example, we decided the wrap was best when you had it mixed first and cut. It sounds like such a stupid little deal detail, but it has huge operational implications.
- [63:18]
A
I mean, mix the ingredients. Mix.
- [63:20]
C
Yeah, yeah.
- [63:21]
A
Because that's a. That's a worse. My least favorite part about any wrap or burrito is you get in there and it's all, like, spaced out. So you have like, an avocado bunched over here.
- [63:29]
B
Yeah, yeah.
- [63:30]
A
All the chicken over here. And then like rice or whatever. And it's like you're. You want to be having the full blend with every bite.
- [63:36]
C
That's exactly right. So it sounds like such a small, minor detail, but it's huge. Arc operational implications. We also had to figure out how to make this work in our infinite kitchens. Over 10% of our fleet and growing is powered by the infinite kitchens. How do you make this work in that operation as well?
- [63:54]
B
How are you thinking about other verticals? I don't even know what you call it. Prepackaged in other grocery stores or consumer packaged goods or product like extensions outside of the retail stores.
- [64:08]
C
We're really focused on our retail stores and the experience. We don't. You know, so much of what we do is based off of the scratch cooking we do in our restaurants. So we're really afraid of extending that out retail.
- [64:18]
B
And you want to get into, like, salad dressing at the grocery store.
- [64:22]
C
CPG is something we'll eventually get.
- [64:23]
B
Okay. You will?
- [64:24]
C
Yeah. We think cpg, like, you know, dressing being an obvious category.
- [64:28]
B
Yeah.
- [64:28]
C
You know, right now, really focus on the core business, but eventually that's definitely a category.
- [64:33]
A
Was. Was my. My immediate reaction to this is. It's. It's. It's incredible. We'll add it to our. Our reoccurring lunch orders. It reminds me of chipotle in like, 22,005 to 2010. Like, some of the best vintages. Like, seriously, back back then, it was. I yearn for. For that time when you could just go into any Chipotle and eat something that actually felt that tasted like fresh and clean and a small number of ingredients. And they had such an extreme commitment to that. And somewhere along the way, I just stopped going because it didn't feel like that anymore. But that's my review. To me, the best prepared food product, product that was widely available in history was that, you know, 2005 to 2010 era Chipotle.
- [65:31]
C
It's funny, that was kind of the inspiration for sweetgreen in a lot of ways. I was in College, you know, 2005, seeing Chipotle take off, eating chipotle all the time. And the idea was to create just kind of a healthier, fresher version of that. And the one you ate, the chicken jalapeno ranch is, you know, it's effectively a burrito.
- [65:48]
B
Yeah.
- [65:49]
C
Some of the other ones are more have lettuce in them. That one is, you know, rice, avocado, chicken, corn.
- [65:54]
A
Saved you a half.
- [65:55]
C
Yeah. And I brought you guys more as well.
- [65:57]
A
Amazing.
- [65:58]
C
And the other thing about the wraps is they start at 11 bucks. So we were also, you know, given where the consumer is today, really trying to create something that can, you know, meet their needs. Today can be something that you can eat every day. So the chicken Caesar in some markets is all the, you know, all the way down to $10.45. And no wrap is above 50, 15 bucks.
- [66:15]
A
Yeah, let's talk about that because we have consumer sentiment right over. I think there was some reporting over the last couple of days. It's like as bad as it's been in like a really long time stock market at all time, highs like insane disconnect. Their jobs report today came in better than expected. But ultimately, consumers aren't feeling great. Part of, I guess part of the approach there is, you know, introducing products like this that you know, are high calorie or not high. I don't know if you describe them as high calorie, but maybe higher calorie and great value. But what else, what else are you doing across the business?
- [66:55]
C
I think the key is around the experience. You know, in restaurants, fundamentals wins championships. So it's really about delivering a consistent experience around quality and hospitality. And that gives you the opportunity to layer in the menu innovation. So right now, you know, we've been very focused on innovation for a while, and right now we're really just trying to dial in the basics. It's what you talked about, Chipotle.
- [67:15]
A
Yeah.
- [67:15]
C
It's like what, what made you love it wasn't all the new stuff that they're now doing. Yeah, it's just being really excellent at the base.
- [67:22]
A
Consistent.
- [67:23]
C
It's consistent. It's making sure the quality is there. You're not out of stock on things. You're staffed at peak. People are, you know, smiling, friendly and warm. It's. It's all those Little things which create, you know, the system behind that is highly complex, but it's all of that that really creates.
- [67:38]
A
Yeah, so. So put another way, sweetgreen, you know, started, you know, as like, my view is like, it's, it has the DNA of a technology company. And maybe that drove you guys at some point to focus too much on, on, on like on the things that ultimately don't drive, always drive. That consistency does that feel like somewhat.
- [67:59]
C
Yeah, I'd say the, you know, early on in like the 2000 and tens, the technology side of the business from a consumer perspective was an advantage. It was, was a competitive advantage. We were the first ones to have digital ordering, delivery, mobile payment, all that stuff. It's, it's pretty much been commoditized for the most part. All restaurants have it. Yeah, you know, I think our digital experience is better than most, but at the end of the day, you can order online, anywhere. Yeah, we do have a technology infrastructure that allows us to scale. So if you look at like our gna, it's been flattered down for five years. So as we continue to scale, a lot of the tools we've put in place and now with how we're leveraging AI, like we can continue to scale without adding headcount and our data and reporting and all the tools we have to manage our kitchens are really incredible. Like, we have tools that, you know, kind of predictive ordering for how we order food. You know, predictive apps that teach us, teach our teams, or guide our teams as they make the hot food in the cold food. Because what's different about sweetgreen is we pretty much make everything in the store every single day.
- [69:01]
A
Yeah.
- [69:01]
C
But to your point, at the end of the day, you know, it's, it's
- [69:04]
A
about you can't get, you can't start spending. Maybe you're spending half your time on innovation, half your time on like the basics.
- [69:11]
C
It's. Yeah, it's probably even more so on the basics.
- [69:14]
A
Right now.
- [69:15]
C
It's on delivering on a great consumer experience. Someone asked me the other day about how we build our brand and you know, we do a lot of stuff, culture, social media, events, all great. But the way we build our brand at the heart of it is just creating an experience that people love.
- [69:30]
A
Yeah.
- [69:31]
C
It's an environment, a store that looks beautiful, clean. It's, you know, a team member that goes out of their way and is friendly. It's a culture of like saying yes when someone wants something that's a little bit, bit weird. Just figuring out you Know that, how to say yes to that, to that customer?
- [69:46]
A
No, there's nothing worse as a customer if you're in like a fast casual experience and you can see everything that's in the store, you can see all the ingredients, and then you ask for something and they say, well, that's not available. And it's like, well, you're just making a call on the fly not to combine those things in the way that
- [70:02]
B
I put the pizza in the deep fryer. I know you have.
- [70:07]
C
So our team knows it's a culture of saying yes. Figure out a way to say yes, take, you know, take care of the guests. And I remind our team all the time our number one marketing channel is word of mouth. 80% of people find out about Sweetgreen through a friend, family. You know, it's the, you know, they walk by a store, they had a good experience and they come back. Like our critical metric that we use in tracking, like the health of how we're doing is what we call our comeback rate. So of customers that come in, what percentage come in within 30, what percentage come back within 30 days? And we can now track that across analog and digital customers. And we've kind of gotten the whole organization aligned around, you know, from the people making the menu, people running our loyalty and CRM, and most importantly, the people in the, in the restaurants. Like, your number one job is delivering a great experience where that guest comes back, because that's what drives that lifetime value.
- [71:00]
B
Where was the first store?
- [71:02]
C
First store was in D.C. in Georgetown.
- [71:04]
B
Yeah. What was the key to making it work?
- [71:08]
C
Key to making it work? I think originally was probably the simplicity of it. Originally the idea was create a healthy fast food restaurant. But the size of the first store was 500 square feet.
- [71:19]
B
So small store, but great location.
- [71:20]
C
Small store, good location, good traffic. Had to keep the, the, keep the operation pretty simple in order to execute it. And I think just, you know, we were maybe lucky in the naivete. We had never worked in restaurants.
- [71:34]
B
Yeah.
- [71:34]
C
And so we were able to approach things a little bit differently and create a brand that just didn't look like the other brands. Whether it be how we sourced, you know, how we went to market in the early days. It's, you know, like one of the things that helped build the brand was we threw a big music festival for eight years.
- [71:49]
B
Wow.
- [71:49]
C
It was like, you know, we threw 25,000 person music festival with, you know, like the Strokes headlining. And the idea was, how do you make healthy eating cool?
- [71:57]
B
Is that within reach for some brands? Like what's the cost to throw a 25,000 person music festival? Did you charge tickets?
- [72:03]
C
We charge tickets, yeah.
- [72:04]
B
Okay.
- [72:04]
C
So it was effectively a break even event.
- [72:06]
B
Oh, that's cool. Okay.
- [72:07]
C
It's called the Suite Life Festival.
- [72:09]
B
Okay, great. Yeah. Any plans to bring it back?
- [72:12]
C
We are looking at it. We have our 20 year anniversary next year and so we have some, some, some ideas.
- [72:17]
B
Yeah, where's, where's the store footprint these days?
- [72:20]
C
Almost 300. We'll end this year with about 300 restaurants.
- [72:23]
B
And do you have a target? Do you want to grow that?
- [72:25]
C
Yeah, if you look at Chipotle, they're at about 4,000. They give a target of 7,500 stores in the U.S. so you know, our first target is 1,000 stores but we think over time should be able to get to about 5,000 units in the United States.
- [72:40]
A
Yeah, the. I always appreciate looking at the forecasts for some of these, some of these brands, whether it be Starbucks or Chipotle, because in our, in our chaotic modern world, everyone's like, it's so hard to plan more than two years out. Especially in tech. Everyone's like, it's hard to see two years out. And if you look at like the forecasts of store growth, it's like Chipotle's like in 2030 we will deliver exactly, you know, 200 new doors across these Geos. And it's like you actually can like think long term. Like consumers are still going to want to eat food, they're going to want to eat healthy food in 2030, 2035, 2040. And you know, taking that long view
- [73:20]
C
I think is super important in the Adams development world. Like you, we're already building our 2028 pipeline and we're looking at 2029. You have to get, get ahead of it, especially when you do a lot of ground up development, building drive throughs. Yeah, that sort of thing. You're about two years ahead.
- [73:35]
A
You're doing drive through.
- [73:36]
C
We do, yeah, we have a handful of drive thrus. They do really well. We actually just opened our first drive through featuring an infinite kitchen. It's close by, it's in Costa Mesa here and it's awesome because now you have the power of the infinite kitchen. From a speed perspective with the drive through stores. Store's doing awesome. So it's a model for the future.
- [73:55]
A
Why have you never done anything in Malibu?
- [73:58]
C
I would love to do Malibu. It's probably my favorite place in the world. It doesn't really have the density from a consumer perspective.
- [74:06]
A
That's been my so I've lived there for coming up on five years. And early on, every time I would leave Malibu and go to somewhere else, I go to Montecito. There's like much better restaurants, in my opinion, in Montecito, Santa Barbara, some of these other beach towns. And I was trying to figure out why, and it's entirely because the town is spread out, like you said, the density, it's spread out on this huge line. That was my intuition. Obviously I don't have restaurant experience, but you can put a restaurant in Malibu, but It's still like 40 minutes in the car for like the average person in town potentially, because you have to like get on PCH and drive like miles and miles and miles. And if there's a little bit of traffic, then that's 20 minutes there, 20 minutes back. And then it's like, I'm not, you know, people are just like, they'll just. I'm not driving 40 minutes for.
- [74:54]
C
The bigger challenge in Malibu is actually the staffing side of it.
- [74:58]
E
Yeah.
- [74:58]
C
Oh, interesting. Nobody in Malibu wants to work for 20, $25 an hour. And so the people that are going to be coming there are coming from really far away. And that's what becomes a problem. You can probably feel it in a lot of the places you go to. Yeah, I'm sure you can see the staffing.
- [75:14]
A
Yeah, like labor, labor, you know, turnover will be higher. General. You just have to, to pay significantly more for somebody to be willing to spend the money.
- [75:22]
B
But yeah, that, that's somewhat limited.
- [75:24]
A
So.
- [75:25]
C
Yeah, limit limited on the Malibu kids. Working, working at the restaurant.
- [75:29]
A
What. How do you, like, I'm. I'm curious how you spend your time with, with like the capital markets and investors today. You've been public for a while now. I'm, I'm sure you have, have had investors come and go, but also some that you've developed like, you know, really meaningful relationships with. What is that like as a, you know, public company CEO?
- [75:53]
C
Yeah. You know, I think we were fortunate. Before we went public, we had a lot of these kind of crossover investors.
- [75:58]
A
Yeah.
- [75:59]
C
Investors like T. Rowe or Fidelity Bailey Gifford that we were able to build those relationships with and, you know, took long, long term positions, believed in what we were doing. So that really helped us as we made, as we made that transition. You know, I talked to them about, you know, our large investors every quarter. And it's like I always say, with investors or with anything, it's really the people behind it than just like the name, name on the door.
- [76:23]
A
Totally.
- [76:23]
C
And building trust and confidence in what we're doing. And you know, I think there's investors that are looking for short term results and there's other investors that really, you know, kind of believe in the long term, believe in the management team and the goal is to just build, build trust with them over time.
- [76:38]
A
Yep. Makes sense.
- [76:40]
G
Yeah.
- [76:40]
A
So no, no, no, we can't expect like an allbirds, like local inference provider provider out of sweetgreen.
- [76:48]
G
I was going to ask who's going
- [76:49]
B
to bid on ebay.
- [76:50]
C
I mean that's an amazing story to watch.
- [76:53]
B
A couple hundred stores, you could go to sweetgreen and verify your baseball cards, auction them off on ebay.
- [76:59]
C
Funny. Our coo, Jason, who who's been with us about a year, spent a long time of his career working at. He was actually the COO at GameStop for a while. He ended up more recently he was at Chipotle, but before that he was at gamestop from like 300 stores to 3,000 stores. So it's been fun watching all of the developments now and talking to him about it.
- [77:22]
B
Yeah, yeah, we were looking it up. It seems like GameStop has shrunk the store footprint a ton of about half since Ryan took over five, six years ago. But like wildly different business because of digital distribution. Like there's a direct competitor, you can't, you can't download a wrap on your phone like you can a game. And so they've had to change the business model. Significant amount for me.
- [77:46]
C
It was interesting learning how much of that business model was dependent. All the margin was dependent on the trade business. So the new releases, they didn't really make as much money on all the leverage the market came from.
- [77:58]
B
Gamestop's famous for being like, oh, you bought that $60 game, we'll give you $5 for it. And they sell it for 20. And that's not bad.
- [78:05]
C
And that was the business.
- [78:06]
B
Yeah, yeah.
- [78:06]
A
What are general real estate trends? Not just restaurants, but that you're seeing across different markets across the US
- [78:20]
C
Drive throughs are hot.
- [78:22]
B
Yeah. What makes for a good drive through? Like there's a few in la, but it doesn't feel like a drive through city.
- [78:28]
C
LA is more about zoning.
- [78:29]
B
Okay.
- [78:30]
C
And getting the right permits for it. And they're not adding more drive throughs. So you have to take over an existing drive thru in order to get one.
- [78:37]
B
I typically only see big chains.
- [78:39]
C
Yeah. And so you're competing with, you know, like some of the drive thrus that we've taken recently have been like, like old Jack in the boxes.
- [78:45]
B
For sure, sure.
- [78:46]
C
But you know, who are you competing with? Chick Fil A? Raising canes. And they have very high auvs. You know, some of these, you know, Chick Fil A, I think that's 8 or so million dollars per restaurant.
- [78:58]
A
So they can, they can just out,
- [79:00]
C
they can outbid you and they need, you know, the acre lot with the, you know, with the egress for the parking. So drive thrus are really hard to come by. But we do see a pretty significant lift when you put a drive through in. People are really looking for convenience and that's only accelerated since COVID And do
- [79:18]
B
delivery drivers prefer drive thru? Is that faster for them?
- [79:21]
C
No, we actually, most people actually don't want the delivery drivers going through it. So you have to build a model where you have a separate area. So if you go to like McDonald's
- [79:28]
A
or any place, because they're not doing the ordering, they're just doing the pickup.
- [79:32]
C
Yeah. So you have like separate parking spots or for the delivery drivers having them come in. Some places are experimenting with the double drive through.
- [79:40]
B
I was going to say, is that possible?
- [79:42]
C
It's possible, but you need so much space in order to do it.
- [79:47]
A
Don't have a lot of space in
- [79:48]
C
California and you don't have the space in California. Other. I think the thing about what we see in retail is the best streets continue to be the best streets. So like, you know, and we see it like if you look at la, you know, Melrose Place, just like rents continue to go up because there's more customers and there's more, more customers. It's, it's like there's this, this thing about, like this, these moments and these, you know, condensed attention and traffic in these great high streets and so on those areas, prices have only gone up. Everywhere else the market's been pretty stable. We haven't seen things, you know, collapse. We have seen. The biggest challenge we've seen is around what's happened with interest rates because landlords have left less money to give from a tenant improvement perspective.
- [80:36]
A
Oh, interesting.
- [80:38]
C
So that's been the issue there. Even though there's probably less demand, you know, when they're, you know, interest rates from zero to where they are today, they're not giving you as much improvement allowance.
- [80:49]
A
Last question. How is the drone delivery market market evolving from your view as, as an operator, as a brand? Like, are you, are you, do you expect drone delivery to be, you know, double, double digit percentage of delivery orders nationally or in specific markets? Like, how do you see it evolving?
- [81:12]
C
I think I Think what Keller and Zipline is doing is really interesting. We've done some early work with them. Looking at it. It's not something we're hugely focused on right now, but they have had some breakthroughs and so I, you know, if you said 10% in the next year or two, I probably don't see that. But over time. Absolutely.
- [81:30]
A
Yeah. It's very market dependent. Right.
- [81:32]
C
Very.
- [81:33]
A
Like, let's work on something in Texas.
- [81:35]
D
It's.
- [81:35]
C
Yeah, it's. It's Texas. More suburban. You do have to change how you build your restaurants a little bit having like the portal where you, where you put the food in. But when you think about the drone portal. Yeah, when you think, yeah you need like a drone portal. But when you think about the speed of delivery, you know, for delivery, getting sub 30 minutes is huge. Is a. It's. That's like kind of like the number when people are opening the app. You want to be sub 30 minutes. And what you can do with drones in getting there, it's. It's pretty amazing. They still can't do really large orders. So like the.
- [82:05]
B
Sure that yeah, what it fits in is ordering for a whole office. But going to work.
- [82:09]
C
I do think over time they will, they will figure it out and more of restaurants will be built around that.
- [82:16]
B
Yeah.
- [82:16]
A
Yeah.
- [82:16]
C
I mean what's amazing to think is how much your radius can increase and what that means for your footprint over time. If that's true. Like.
- [82:23]
A
Yeah. I mean in la. La, since, you know, we're here. When you look at this market, the craziest thing is putting, when you put, you know, you go into maps and you put an address in and you're like, it's seven miles and it's going to be 50 minutes is like a normal, a normal thing. It sounds crazy to anyone else in the country, but it's like a drone going seven miles is like, you know, feasible.
- [82:45]
C
Yeah. And you've seen the videos, they like they're able to just drop the, you know, just literally have it parachute like right on your doorstep. So it's, it's pretty amazing. It's coming, but I think it's a little early for us.
- [82:56]
B
Yeah.
- [82:56]
A
Yeah.
- [82:56]
B
Very fun. Well, thank you. So thank you for the food.
- [82:59]
C
Great to see you guys. Let's hang.
- [83:01]
A
Let's hang soon.
- [83:02]
C
Thank you guys.
- [83:03]
B
Put this in here. Throw these back on and we can click over to some of Jason Fried's favorite watches. He was a fan of Nike's early 2000s era watch lineup. Have you ever seen these before? Jordy?
- [83:19]
A
I have, I have.
- [83:20]
B
I have not.
- [83:21]
A
I've never seen one in person.
- [83:23]
B
Never seen one in person. Where's my other headphone? There we go. Yeah. Has this design aesthetic come back yet? I feel like there's enough posts about it at some point. I feel like a tech startup is going to really run with it on some device.
- [83:38]
A
Someone will do it. I like the kind of soft lines.
- [83:41]
B
Yeah. Avi from Friend was sort of doing that with the material that went into the Friend pendant. But he was talking to Paul Graham about like, how do you differentiate from Apple? Like if you launch a product like this, it does seem stand out from an Apple watch, certainly aesthetically. And so you at least create some distance that's maybe defensible. Even if they come out with a competitive product, if you become iconic with this design or a similarly bizarre design, you don't stand as much chance of getting commoditized. Ryan Cohen's in the Financial Times. People have been processing his bid for GameStop to take over ebay. Very fun story. Not too many updates on what's actually happening. I think there's sort of a wait and see approach here. But the Financial Times has a little bit of a deep dive here. Ryan Cohen spent much of his career being dismissed. It's a pattern that has turned the chief executive of video game retailer Gamestop into a leather jacket wearing outlier with a loyal base of meme stock investors. I saw someone commenting the legal, the leather jacket, very Jensen esque. Although has he been dismissed? I feel like through the rise of Chewy there was a ton of respect for him all along that journey. Since e commerce is such a challenging business, there's so few companies that have exited at north of a billion dollars. There's been a ton of attempts to make e commerce companies, but they often run into Amazon headfirst. The Canadian entrepreneur turned activist investor has built his cult like reputation by repeatedly betting on ideas that many institutional investors labeled as absurd. First it was online pet supplies. Okay, I guess that fits. Then a dying strawberry.
- [85:33]
A
Why has Chewy. Why has Chewy. So Chewy is a $9.6 billion company.
- [85:39]
B
Yeah. That's great.
- [85:40]
A
Why has Amazon and Walmart not eaten them alive?
- [85:48]
B
I don't know. I mean, I think a lot of people have pet food on subscription and they just don't change it. And so you acquire that customer and then even if it's on Amazon, so
- [85:57]
A
the CAC equation just works.
- [85:58]
B
I think the, I think the retention is remarkable because the dog doesn't tell you, hey, I'm bored of this particular food.
- [86:05]
D
Food.
- [86:05]
B
Just like awesome.
- [86:06]
A
I feel like my dogs would grow up and they'd be like, but how would they communicate? They had this look and I know I'd had to go take an egg, go get an egg from the chickens and crack it on their.
- [86:17]
B
But that's not. You're not getting that egg on Amazon, right? They don't make egg laced dog food on Amazon. Like switching e commerce platforms isn't going to solve your belligerent dog problem.
- [86:32]
A
Yeah. One response from eBay could be Focus on yourself King, because Chewy's down 40% the last year.
- [86:40]
B
Well, is Ryan?
- [86:41]
A
No, he's no longer the CEO obviously.
- [86:45]
B
But still, I guess it's down since IPO. It had a quick rise in 2021 the stock was up at $101, now at 23. So. So some sell off there. But let's read through more of this. Now he is focused on this $56 billion attempt to acquire ebay and our e commerce giant, roughly four times the size of GameStop. Wall Street.
- [87:09]
A
Right?
- [87:09]
B
Wall Street's immediate response to the hostile bid has ranged from skepticism to outright mockery. Analysts have questioned how GameStop, valued at roughly $11 billion, could realistically finance the takeover of a company worth around 46 billion. None of that has dissuaded Cohen. The more ebay fights me, the more I'm going to not take no for an answer. I'm not going away, he told the financial times. The 40 year old billionaire first set out his his stall for the takeover bid in an awkward interview with CNBC on Monday that instantly ricocheted across social media. He responded to questions about the financing details with a blank expression and clipped monosyllables, explaining that it was half cash, half stock. Of course, that became a massive meme. The interview generated waves of memes online as X and Reddit users speculated the Cohen looked drunk higher.
- [87:57]
A
Do you think the half cash, half stock meme is good or bad for the prospects of successfully acquiring the company?
- [88:05]
B
I think once it clicks it sort of makes sense and. But it is a big jump to ebay shareholders to, you know, to roll into that at that, at that price.
- [88:15]
A
Yeah, the decision or the sort of like situation would be wildly different if ebay hadn't been performing since Jamie became CEO.
- [88:22]
B
Yeah, it's outperformed the S&P 500, right. Which is pretty remarkable. So yeah, it's not beaten down and the shareholders seem pretty happy. The Financial Times says the script has played out before. Born in Montreal in 1986 to a teacher, mother and father who imported hoarded glassware. Cohen did not attend university. As a teenager, he built websites for family, friends and small businesses, teaching himself coding now in 2011, now in his mid-20s, he founded Chewy from South Florida with the ambition of creating an online pet supply business that could compete with Amazon. Investors thought the idea was ridiculous. I knew that significant capital would be required to finance the growth, that's for sure. A lot of inventory, very, very, very capital intensive business, he said. He approached dozens of VC firms, even flew out to Silicon Valley, went door to door on Sandhill Road. Can you do that? I don't think you can just knock on Sequoia Capital.
- [89:21]
A
Someone should. I think some of these Neo labs are hot enough that they actually drive down, no meetings and just go door to door. I'm here unannounced. Be like, cool, I'll come back in like 20 minutes.
- [89:33]
B
I got a coffee and grab a coffee. Yeah, yeah, probably. So he went door to door on Sandhill Road. Probably a turn of phrase. He was probably scheduling meetings. But he was explaining how Chewy would succeed by delighting customers in running an ultra efficient operation. But everyone turned us down. He couldn't raise money. The skepticism was understandable given the failure of pet supply website pets.com and the. Com bubble era. Pet food was bulky, low margin and expensive to ship. Venture capitalists doubted consumers would buy 40 pound bags of dog food online. Cohen ignored them all. Instead he focused obsessively on customer service. Chewy employees wrote customers handwritten cards and sent flowers to grieving owners who animals had died. That's really sweet. That is a good touch. He doesn't come across as that. He comes across as, you know, this like ruthless operator, but, but that is really, I mean obviously that.
- [90:24]
A
Well, he comes across as an absolute dog, John.
- [90:27]
B
Yes. He was raised by dogs. Essentially grew his career through the dog economy. The approach helped to create unusually intense customer loyalty. By 2017, private equity backed Petsmart bought Chewy for 3.35 billion, then the largest e commerce deal on record. It cemented Cohen's reputation as a rare entrepreneur who had successfully challenged Amazon on its own turf. Flush with cash and credibility, he launched Ryan Cohen Ventures, RC Ventures, his investment vehicle that led to his most controversial bet in 2020. While much of Wall street viewed GameStop as a doomed relic of the physical retail era, Cohen quietly amassed a large stake in the company and began pushing for a radical digital transformation. I really wonder if there's going to be. What is this? Just horses why are we just playing horses?
- [91:16]
A
The team had this on this morning. I thought it was incredible.
- [91:19]
B
Yeah, the horses are good. I wonder if they will be subway servers.
- [91:23]
A
Some people do subway servers.
- [91:25]
B
We do horses.
- [91:26]
A
Horse B roll.
- [91:27]
B
Do you think there'll be a Ryan Cohen gamestop type deal of the sass apocalypse era? Like the most SEAT based sass, you know, like laggard who's getting absolutely beaten up, shorted. Huge short interest. And some wild card operator comes in and. And turns it around or at least turns it into a meme stock.
- [91:47]
D
I don't know.
- [91:47]
A
Yeah, it feels. I mean, it's just felt like so many people were, were too, too afraid of. Of, you know, catching a falling knife basically. But sure, we'll see some of that.
- [91:59]
B
We can dig into it more. We can ask Zol about it because we have my good friend Zol from Refractor in the TVP and ultradome. Let's bring him on down to the seat and I don't know, are those plates left over? We can of course, work. Anyway, great to see you, Jerry. Good to meet you.
- [92:15]
G
Thanks for having me.
- [92:17]
B
Let's introduce yourself and then I want to hit the gong. I want you to hit the gong. Tell us the news today. Introduce yourself and then tell you the news.
- [92:23]
G
Yeah, no, I'm Zahl from Refactor Capital. As you know, I led Lucy's seed round a decade ago.
- [92:28]
B
That's right.
- [92:30]
G
And it's been a pleasure. Yeah. And I run a deep tech hard tech fund.
- [92:35]
B
Hey, there we go.
- [92:38]
G
I like it.
- [92:39]
B
Yeah, yeah.
- [92:40]
G
And by the way, I do have swag for you guys. So that is for you. And I got to.
- [92:43]
A
Guys, how did. How did John. How did John pitch Lucy to you back then? Because this.
- [92:48]
G
Oh, that is a long.
- [92:49]
A
This is what, four years before Zins were like a common, you know, word?
- [92:55]
G
Well, no, no, it was a. It was a smoking alternative.
- [92:57]
B
Jordy.
- [92:58]
G
That's what he was pitching.
- [92:59]
B
It was before Zinn launched. It was before Zinn launched.
- [93:02]
A
No, I know. Yeah, I know.
- [93:04]
B
Yeah, yeah. I mean, the pitch was basically just.
- [93:06]
A
The main thing is like if I was getting that pitch. I mean, you're very compelling. I probably would have invested if we knew each other back then, but you know, a pouch. Yeah, it's popular in Europe.
- [93:19]
B
Well, we weren't doing pouches yet.
- [93:21]
G
No, it was just the gums.
- [93:22]
B
And I think we maybe had mapped out loss and. But the idea was basically that Nicorette was a horrible product. Horrible product. But had grown into like a billion dollars a year in Revenue and had been around since the 70s, had been really well studied by the scientific community, had been loosely effective at getting people to quit smoking, but delivered in the most medicinal way and had not had a consumer or brand wrapped around it. So very much in like the hymns and Roman telehealth, like let's take the, you know, the scientific backed product that we know works, but then package it in a new way and go direct to consumer.
- [94:02]
A
Yeah.
- [94:02]
B
Chewy style.
- [94:03]
C
Yeah.
- [94:03]
B
It was fun.
- [94:04]
G
It was a blast. It was a blast.
- [94:05]
B
But you were doing a lot of other investing at the time. What else stands out from your investing career?
- [94:10]
G
Yeah, well, I've been. Can I first say, like, I have had a fantastic week. Can I tell you about the week? Yeah, I think it's my best week adventure in and I've been doing this for 12 years. So let me paint the picture. Right, so Wednesday, astranus raises a 450 million Series E. This was the first check I wrote after I left Andreessen Horowitz.
- [94:31]
B
That's right.
- [94:32]
G
Very first check I wrote.
- [94:33]
B
Great company.
- [94:34]
G
Led the seed round, introduced them to a 16Z which did the Series A. And they've also done subsequent rounds. So that was on Wednesday morning.
- [94:41]
C
Yeah.
- [94:42]
G
That was amazing.
- [94:43]
A
Right?
- [94:43]
B
Yeah.
- [94:44]
G
You know, $2.8 billion valuation. And then on Thursday, of course, I launched my new fund. Fund five. Right. And so I've got 300 million under management. Yes.
- [94:56]
A
10.
- [94:56]
G
10 years and change. Yeah. At this point. So another 50 million. I just raised $50 million funds again. We can talk about that in a sec. And then, you know, I'm like waking up yesterday morning and I'm like, you know, getting all my Tweets and my LinkedIn posts together for the launch at like 6am and something pops up in my feed, in my emails. It's like one of my companies just got acquired. I'm like, okay, well, I got to take a look at this. What happened? Well, Path AI got acquired by Roche for a billion dollars.
- [95:24]
A
There we go.
- [95:27]
G
And I have two founders that have had kids this week, babies this week. So it is just tons of good stuff. Best week ever, guys.
- [95:34]
B
Yeah, I mean, I know.
- [95:37]
A
So it's time to rest on your Laurel.
- [95:41]
B
I'm thinking about getting a bed of laurels that people here could physically go and rest on.
- [95:45]
G
Yeah, yeah, exactly.
- [95:46]
B
One day. What were you doing before Andreessen?
- [95:49]
G
Yeah, yeah. So, you know, originally from the Midwest, born and raised in Indiana. My dad had a computer business growing up, so I built computers in our basement and we sold them to People's homes, doctor's offices. 80s and 90s pre dope.
- [96:01]
A
This is basically like, we'll build you a customer PC for your use case.
- [96:05]
G
Do you remember 386s and 486s and Pentiums and all that?
- [96:08]
A
I was born in 95.
- [96:10]
G
Okay. So maybe not, maybe not. But this was like the era of the 80s and 90s before Dell, like really kind of took over.
- [96:17]
A
Yeah.
- [96:17]
G
And so, you know, we built that business. So I got into hardware software coding at a young age. And then I went to Penn and I did the Wharton undergrad program. What I didn't understand when I got there was that 90% of my classmates would go into banking and consulting, and I did internships in both. And I hated it. So I didn't know. A senior year comes around and I'm like, what am I going to do with my life? And I was so lucky. I was so lucky, guys. Microsoft came to campus and recruited me to be a product manager.
- [96:43]
B
Okay, there you go.
- [96:43]
G
Right? And so that started training under the goat.
- [96:45]
A
Yeah, go Satya.
- [96:47]
B
Yeah, Satya. Running it.
- [96:48]
G
No, no, it was Steve Ballmer at the time.
- [96:50]
A
But wasn't he like a rising pm?
- [96:52]
G
He was a rise. He was. He was VP of Product at that point.
- [96:54]
A
Yeah, yeah, that's what I'm saying. You're working under the.
- [96:57]
B
In the org.
- [96:57]
G
Yeah, People knew him. Of course people knew him.
- [96:59]
A
Insane aura. Yeah, that's awesome.
- [97:01]
G
But it was like a time when like Microsoft stock price was just between 25 and $30.
- [97:05]
B
Yeah.
- [97:06]
C
The baller era was rough.
- [97:07]
G
It was rough. And they didn't understand the Internet. This was 2004 to 2006.
- [97:11]
B
This was like, while they're missing mobile, basically.
- [97:13]
G
Correct. They missed the whole. And I was like, this is not happening. I'm wasting my career up here in Seattle. So I moved to the Bay Area 20 years ago and I got a job at Google. I was a PM in 2006 on the AdWords 2 team that. Let me tell you a story on that real quick. So this was Susan Wojcicki's team. So Google was founded in her garage.
- [97:33]
B
Yep.
- [97:33]
G
You know, back in the day. She passed away a couple years ago of non small cell lung cancer. But she was an amazing leader, amazing human, and the talent that she had put together on that floor in Building 42 on the second floor was amazing. So I'm in this cube. Elad Gill is sitting next to me. Satya Patel from Homebrew over there.
- [97:51]
B
Whoa.
- [97:52]
G
Josh McCarthy Farland, who was At Greylock and now at boom. Ben Silberman from Pinterest.
- [97:56]
A
No way.
- [97:57]
G
Kevin Systrom from Instagram. Dan and Zeke from Optimizely.
- [98:01]
B
Wow.
- [98:02]
G
Tom Tungas now from Theory Ventures. And Goku Rajram, who's on the board of Coinbase and Pinterest, not father of adsense. We were all within three cubes.
- [98:11]
B
That's incredible.
- [98:12]
G
It was insane. I don't know what was in the water there, but like, that's remarkable. We've all done pretty decently.
- [98:16]
B
Yeah, for sure. What is, what is a PM on AdWords do? Yeah, because there's a, there's like the ad matching product. That's a product. But then there's like the dashboard with which brands go to advertise and sites and there's like, you know, two sided marketplace effectively. Like what is a day in the life? Like.
- [98:37]
G
Yeah. So there's, you know, back in the, in the core product offering at Google back then was AdWords and AdSense.
- [98:44]
B
Yeah.
- [98:44]
G
So AdWords was like, okay, I go to google.com, i type in a search query, the ads show up. AdSense is putting ads, whether it was text or display image ads, video ads on other publisher sites so that publishers can start making money from the Google search algorithm based on the content of scraping that page, surfacing the best possible ads. So my job was to kind of think about new concepts or new product launches to make more money on the AdSense side.
- [99:08]
B
Sure.
- [99:09]
G
But they were AdWords products on AdSense.
- [99:11]
B
Got it. Yeah, yeah, yeah, yeah. Right.
- [99:12]
G
But then we acquired YouTube and so I was asked to move over to YouTube and another OG hunter walk.
- [99:18]
B
Oh yeah.
- [99:18]
G
Satya's partner at Homebrew. Yeah, Homebrew brought me over.
- [99:21]
B
That's awesome.
- [99:21]
G
So I was working at YouTube for three years.
- [99:23]
B
Cool.
- [99:23]
G
And this is when we were bleeding cash. Everybody was suing us. The Sony, the music labels, like the movie studios. We were burning cash like crazy. Yeah. And Eric, Larry and Sergey come to us and they're like, hey guys, this is like 2007, 2008. Hey guys, like we need to turn on the revenue spigot like yesterday.
- [99:45]
B
Yeah.
- [99:45]
G
Like, go figure out how to get the Fortune 500 to feel comfortable putting their brands on YouTube. Because remember with UGC, user generated content, people were really worried about it again 20 years ago. Right. So we had to build all new products to get them more comfortable with making that, making that their, their brands and their products available on YouTube.
- [100:04]
B
Yeah.
- [100:05]
G
And so, you know, we built that business to a few hundred million in revenue. I Think they did like 20 billion last year.
- [100:10]
A
Linear.
- [100:10]
B
Linear or something.
- [100:12]
A
Linear TV execs at the time, were they saying like no one's ever going to put dollars against YouTube because of that UGC risk? Were they, Were they, Were people saying that?
- [100:21]
G
Yeah, I mean they didn't, they didn't, they didn't want to, but they didn't want to risk because honestly YouTube's filters for content were not as good back then. So like weird stuff was like popping up.
- [100:33]
B
Totally. This was like the tagging age where like you could upload a video and tag it like skateboarding Los Angeles and then you hope that that's what's in the video. But there wasn't an AI model that could look inside every frame, listen to the audio and actually understand what it was. At least not yet.
- [100:53]
A
What were the. I'm super interested in, you know, every, every platform, every platform that has traffic and usage is like different. So like you know, just screen space is on, on a website is different than screen space on Instagram is different than screen space on, you know, a Google search result. Right. What, what were the unique challenges with building an ad product for YouTube given the consumption habits and the wit and, and the state that people come to YouTube in.
- [101:23]
G
Yeah. So, you know, they actually had to create a new role for me because they called me the big ads PM. So my job, to your point, on linear TV, you know, execs, they wanted a, they wanted their 30 second TV spot on YouTube, but we weren't, we weren't gonna do that. That's just like so orthogonal to the YouTube thesis.
- [101:41]
B
Yeah, yeah.
- [101:41]
G
So what we did instead is we let them buy the homepage.
- [101:44]
B
Oh yeah.
- [101:44]
G
Do you remember this? So we had homepage takeovers.
- [101:46]
B
Yeah, yeah, yeah.
- [101:47]
G
So you could put like these, these big banners up.
- [101:48]
B
Yeah.
- [101:49]
G
And our sales team would have this calendar of 365 days and they go sell them and they go sell them for a million a pop.
- [101:54]
B
Wow.
- [101:54]
G
Two million a pop. Whatever it was on Christmas or whatever. Like three million.
- [101:57]
B
Yeah.
- [101:58]
G
Right. And so. And it got tons. I mean hundreds of millions of views.
- [102:01]
B
Yes. So much traffic.
- [102:02]
G
Yes.
- [102:03]
B
Everyone lands on the homepage. Then it was pre app.
- [102:05]
G
Yeah. Apple ea all the big brands.
- [102:08]
E
Yeah.
- [102:08]
B
They would all do l'.
- [102:08]
A
Oreal.
- [102:09]
G
They wanted to put their brands up there and you know, that was getting them comfortable with like a known display ad product that they were using on AdSense or somewhere else and bringing them on. But with that million dollar ad buy, John, we were forcing them to actually spend money as part of that million on YouTube.com through the search engine. So when people search, they'd have to have their display ads and their other search products there and running. So it got them comfortable with using YouTube because otherwise it's like, okay, it's just a 24 hour pop.
- [102:40]
B
Yeah.
- [102:40]
G
Don't you want to continue the campaign for the next month and leverage all the gains that you made from brand awareness by coming people coming to YouTube.com yeah.
- [102:46]
B
And they're not tied to a specific video at random, which is where the real brand risk comes. Because it's like your ad was shown on this particular ad. It's like, no, we were on the homepage. There's a lot of things. If you're mad at that one video, take it up with. With Google or YouTube.
- [102:59]
C
Right?
- [102:59]
G
Yeah, exactly.
- [103:00]
B
Interesting.
- [103:00]
G
And then after YouTube, I went to Netflix and I was their first head of mobile.
- [103:04]
C
Yeah.
- [103:05]
G
Yeah. This was like six months after the iPad launch from Steve Jobs and we had to rebuild all the phone and tablet from the ground up. They were growing like a weed and they look like this is like Forrest
- [103:16]
B
Gump level Silicon Valley lore. Then I was at Andreessa and then I was at Netflix. Then I was at.
- [103:23]
G
There's definitely gray hairs here.
- [103:25]
B
Every big tech company, there's like three on the Mag 7 that you still gotta go do tours at at some point.
- [103:30]
G
Well, I did LinkedIn after that.
- [103:31]
B
Okay. Yeah.
- [103:31]
A
I got Microsoft there, so.
- [103:34]
G
But it was. I was basically at LinkedIn.
- [103:36]
A
Please tell me you were. You were pitching them on short form vertical video on LinkedIn back when you were there. Because they finally added it.
- [103:43]
G
They finally did. They finally got comfortable with it. Yeah, we were talking about it back then, but they were not. They were not going for it. They were going for it.
- [103:50]
B
So. Yeah. What was the experience?
- [103:51]
A
Too addictive? I spent. I spend four or five, sometimes six hours on LinkedIn just scrolling on the vertical video.
- [103:58]
G
Oh yeah, yeah, it was great, man. I mean, a very small team back then. This was 2011, 2012. Only 8PMs in the whole company. I was responsible for all the phone and tablet apps across Android. Like the Tablet Nook. Remember those? And the Kindred Fire. Kindle Fire.
- [104:14]
F
Rather Kindred Fire.
- [104:16]
G
And so we had to build that.
- [104:19]
B
No one really gives Bezos enough credit for using the Gen Z slang. He made a fire phone and that's what kids would say, oh, we should make a fire phone. Like this phone is fire. And he. And he roll it into an entire brand. But today people would like call that out as like Being too, like, patronizing to the younger generation.
- [104:37]
C
Yeah.
- [104:37]
B
But the fire phone just sort of slid under the radar.
- [104:40]
G
Oh, well, it was a complete flop too.
- [104:41]
B
I know.
- [104:42]
A
Yeah. Well, he also has like, you know, he clearly picked up, people say, you know, the derogatory, like, oh, you're basic. Yeah, he flipped that around.
- [104:51]
B
Yeah.
- [104:51]
A
And tried to turn it into a positive with basic.
- [104:54]
B
Oh, yeah.
- [104:55]
A
Like, it's okay. It's okay.
- [104:56]
B
Gen Z slang over there at Amazon.
- [104:59]
G
That's right. Anyway, yeah, so that was, that was a great time. And then, you know, Andreessen Horowitz, about 12 years ago, we're looking for kind of junior folks of like their general partners who are product people, like me to come join the investment team and help them source, evaluate and work on investments together. Right. And so this was 2013. I ended up joining. The firm was only four years old at that point. Investing funds three and four, which were each 1.5 billion in size.
- [105:26]
B
Yeah.
- [105:26]
G
At that point they had not launched any sector specific funds like the crypto fund or the gaming fund or anything.
- [105:31]
B
Or American Dynamism.
- [105:32]
G
Exactly. None of that was a lot. It was just one fund.
- [105:35]
B
They didn't even have growth yet. It was just one.
- [105:37]
G
It was just one fund. Yeah, it was just one. One vehicle.
- [105:39]
B
But they were doing weird deals. Like, the Skype thing is always so funny to me.
- [105:43]
G
Were you doing SPVs? Yeah, we were doing SPV into Pinterest, Facebook even.
- [105:48]
B
Yeah, yeah. It was very creative deal making. It was not. Which I think was like to their massive benefit.
- [105:53]
F
Yes.
- [105:53]
B
Just so many experiences obviously of financial wins, but also just like the breadth of activity at the firm allowed for those expansions to go successfully.
- [106:02]
G
That's right. That's right. And so, you know, obviously you can imagine it was just an amazing whirlwind of an experience. And then I got to meet you and Rob Soylent Journey.
- [106:10]
B
Yeah, that was.
- [106:11]
G
Well, yeah. And then, yeah, with Chris Dixon. And that's when I started actually getting into hard tech.
- [106:17]
B
Yeah.
- [106:18]
G
So I was kind of bored trying to find the next Netflix, the next Google. It wasn't as interesting to me as like the world of atoms. And so I ended up starting getting into digital health, computational biology, synthetic biology. I don't have a bio background, but like very smart people like you guys got me excited about where some of these directions were going, especially in food and chemicals and materials. And then a few of us convinced our LPAC to actually launch the Bio fund.
- [106:43]
B
Yeah.
- [106:43]
G
Which was the A16Z's first sector specific fund that they launched in 2015, which was a $200 million fund to go after the intersection of life sciences and computer science. And that's when I spun out and started Refactor, you know, about 10 years ago.
- [106:56]
B
Yeah, yeah.
- [106:57]
G
And so, yeah, if you recall, I started it with David from SV angel, and then he ended up retiring, but been solo since. And. Yeah, I haven't hired anybody. It's literally just me running the firm.
- [107:10]
B
So you have SaaS for the tooling on the back end, fund administration, like lawyers. There are certainly things that in a day you can't do. Do you not even have an executive assistant?
- [107:25]
G
I had an executive assistant for nine years part time, who would just do a. All my scheduling, all that stuff. But there was a new company that Sequoia. Sequoia Partner actually ended up leading called Block It.
- [107:36]
B
Okay.
- [107:37]
G
And it's an AI calendaring tool.
- [107:38]
B
Okay.
- [107:39]
C
And you use that.
- [107:39]
G
And I just. I just. I just made that change three months. It has been amazing.
- [107:43]
B
Amazing.
- [107:43]
G
So I have, like. I don't know what SaaS is, John, but I know what AI is. But I've got, like, a whole bunch of.
- [107:49]
B
Yeah, yeah.
- [107:49]
G
You know, setups that I can actually, like, amplify my time. I have my fund admin. They do all my compliance. I have a tax and audit team, fractional back office. Yeah. But as far as employees go.
- [108:00]
B
Yeah, no employees.
- [108:02]
G
You're looking at them.
- [108:03]
B
That's amazing.
- [108:03]
A
Yeah.
- [108:04]
B
That's wild. Then why keep the fund size the same? Why not at least inflation adjust it?
- [108:10]
C
Yeah, yeah.
- [108:11]
G
I mean, so this. I get this question all the time because I had more than 50 million of interest in this fifth. Right. And so if I increase the fund size, then one of a few things might happen. A, I have to write more checks. Right. And so that's, you know, as a single person, like, it might be a little stressful and, like, I won't be able to give as much time to my founders.
- [108:32]
B
And marginally you might be adding, like, the company that wouldn't make it on the margin in the previous fund. So you're just adding in, like, worse performance.
- [108:39]
C
Adverse selection.
- [108:40]
B
Yeah, adverse selection.
- [108:40]
G
Exactly. Exactly. And so if I. I only do 20 or 25 companies, every fund, it's a concentrated portfolio. I'm writing one to $2 million checks. All things hard tech, bio, energy, aerospace, critical materials, robotics, physical AI. So all of these things, like, need my time. And so if I. And I. And I want them to actually be a fun driver.
- [108:58]
A
Right.
- [108:58]
G
If I'm starting out with 10% ownership in a business, maybe after dilution I have 5% of that company. Well, what's 5% of a billion dollar unicorn? $50 million.
- [109:08]
B
Yeah.
- [109:08]
G
What's my fund size?
- [109:09]
B
There you go.
- [109:10]
G
So it makes it work, right?
- [109:12]
B
Yeah.
- [109:12]
G
And I hope there are much bigger outcomes. But, but like Path AI, it was a great outcome. Right. And so I wish I had more in it.
- [109:19]
B
But on dilution, is the level of dilution increasing? I mean you look at the capex numbers from AI labs, you look at the capex and spending and dilution that's going on at large scale hard tech companies that are growth stage. Is that dynamic shifting? Are you worried about that or are you changing your strategy based on what's happening with, with some example companies?
- [109:41]
G
Oh yeah. I mean we're seeing it in the last six months in the areas of hard tech, deep tech, whatever you want to call it. Right. Like we're seeing it, the valuations increase both at precede and seed. And so I have to be able to make certain adjustments to make sure that I can get into the highest quality companies. Right. Because like 10% of a bad company is still a bad company. Right. And so to your point, on average selection, you know, I want to avoid that. So you know I've made very conscious decisions and exceptions on being below my target ownership to get into some of the very best companies.
- [110:14]
B
Sure.
- [110:14]
G
Right. And so especially in this world of AI where I feel like the outcomes may not be a billion, they might be 10, might be 100. Right. And so I mean we're seeing that constantly.
- [110:23]
B
Even with like there's so many markets, I mean the coding agent market across cursor, windserve, cognition, like you have three companies that are all north of three decacorns in the same category. Effectively that is a wild situation. I can't remember the last time that happened in a competitive industry. Certainly you had Uber and Lyft, but there was a divergence there at some point. And the third company was, I forget what it was called, but I think it topped out at like 100 million in valuation, maybe a billion. But it was very power law driven. And we're seeing many more categories be more oligopolistic early on, which is, I
- [111:04]
G
mean the venture math is pretty simple for a seed fund like mine, like refactors. If I want to be a top decile investor, you go back and look at 40 years of venture capital history, what is a top decile investor? A net 3x fund. So I take my 50 million and I turn it into 150. But in order to do that, there's a lot of ways to make money in. But the vast majority of those funds that hit that top decile net 3x mark had at least one fund returner, meaning they had one company return the entire fund. More likely many multiples like a 2, 3 or 4x. Right. And so you only can do that if you have sufficient ownership at entry. Which is why I've done more of a concentrated strategy versus I love SV Angel Box Group. I co invest with them a bunch of other seed funds that I love working with. They write smaller checks and they probably do like 100 investments of fund and they get into some amazing companies. So I love working with them because they're a great source of deal flow but they've got a whole team that can go and just meet companies all day long. That's all they do, they just meet companies. And I want to really spend more time building foundational relationships with my founders. I just get a lot out of that.
- [112:17]
B
Yeah. How are you sourcing and vetting hard tech founders? It feels like there's the SpaceX alumni mafia and then there's the wildcatters, the college dropouts who are doing a bunch of interesting stuff.
- [112:31]
G
Yeah, yeah. I mean I have one here in LA General Galactic ex SpaceX Exvarta team building the fastest engines in space using electrolyzers. So water on board. You guys remember the TV show the Expanse?
- [112:44]
B
Yeah.
- [112:44]
G
And for all mankind.
- [112:45]
B
Yeah.
- [112:46]
G
The whole idea is like can we use water as a fuel?
- [112:48]
B
Sure.
- [112:48]
G
Right. And we think that there's water at the south pole of the moon. Imagine being able to actually have a satellite powered by water that can go all the way out to the moon and refuel and come back. And if we're going to have a civilization on, on, on, on the moon, you know that is going to be required water as propellant.
- [113:04]
B
So what is your diligence process for that? Because that sounds like amazing but I don't feel feel equipped to answer. Will that work or not? How are you deciding to make the investment?
- [113:15]
G
Yeah, I mean when I invested in their pre seed like you know, what was it, three or four years ago? Three years?
- [113:20]
C
Yeah.
- [113:20]
G
Three or four years ago it was just Halen and Luke with a deck and an idea. And I just love them as founders and they were extremely well referenced and I feel like they had a sense of urgency about them.
- [113:33]
B
Sure.
- [113:33]
G
And that's what I look for. I mean I don't really have a market map saying I need to invest in this idea. I want it to exist. That's where I make mistakes. Right. If I fall in love with an idea more so than a founder, I have to really watch myself. It's more of the other way around. I need to fall in love with the team. I may not know what the business model is going to be or the product. This team has pivoted once before. Right. But they figured out this new idea and it's going to work. And the space Force loves them and getting some great momentum there. And so that's an example of one. But like, you know, when you're doing hard tech investments across all these categories, I leverage my network of professors, advisors and founders and I don't have to pay them because they just want to meet these companies for free because sometimes they may want to invest, sometimes they want to be an advisor. And so it's great to be able to do that.
- [114:20]
B
And so they can sort of fact check some of the scientific claims. Jordy, any other questions?
- [114:25]
A
Is the, Is the business getting more, you know, running the firm getting more fun now that you have over a decade of investments that are kind of, you know, some of the most exciting companies in hard tech are the ones that were started 10, you know, 20 in the case of Space X 20 years ago, things like that. I think of companies like, you know, Zipline, Keller's company that are, that were incredibly exciting, probably in a deck back then, but then now are like actually coming to real, like fruition.
- [114:54]
C
Yeah.
- [114:56]
A
At least here in the US and you mentioned like weeks like this with a lot of activity. I've certainly felt that way as an angel investor. It's like the more time that passes, the more fun it is to be an angel investor because you're not just making new investments, but you're getting to see the, see the sort of stories of all your previous investment play out and you start to like, really notch wins. And it's a great feeling.
- [115:18]
G
It's a really great feeling. You know, I've got six unicorns in the portfolio today. I've got many more hopefully coming, you know, nuclear energy company and all of these things. So it's.
- [115:27]
C
Oh yes, it's great.
- [115:34]
G
And so, you know, there are just so many amazing companies to be built and I'm ready to work with them. And I don't know if you guys saw my new website, but I've got my call sign.
- [115:43]
B
Okay, what's your call sign?
- [115:44]
G
Better call Zol.
- [115:45]
B
Better call Zol.
- [115:46]
F
Better call Zol.
- [115:49]
G
And it's the call sign I was built for.
- [115:51]
B
Well, thank you so much for taking the time.
- [115:52]
A
Thank you, John.
- [115:53]
G
I appreciate it.
- [115:53]
B
Jordy, always good hanging.
- [115:55]
G
I'll give these to the guys.
- [115:56]
B
Fantastic.
- [115:56]
A
Perfect.
- [115:58]
B
We have Brian Chesky from Airbnb joining in just a minute. We can pull up what else is in the timeline. In the meantime, someone. Oh, there are some birthday announcement. David Attenborough turned 100 years old. He's called the most consequential broadcaster of our time. And if you haven't gone down the David Attenborough rabbit hole, I highly recommend it. Major inspiration for travel. And I'm glad we have Brian Chesky from Airbnb here to talk about all things travel and his business. He's in the waiting room. We'll bring him in to be the ultra one right now. Brian, how are you doing?
- [116:38]
A
Good.
- [116:38]
F
How are you guys doing?
- [116:39]
A
We're doing great to see you.
- [116:40]
B
Welcome back.
- [116:41]
A
It's been too long.
- [116:42]
B
Yeah. Why don't we kick it off with just, you know, an Update on how 2026 is going, how the business is going. What's new in your world?
- [116:51]
F
Yeah, I mean, things are really good. We've accelerated growth. For the first time since the pandemic, we grew 10% last year in revenue. And this quarter we announced that revenue was 18%. So from 10 to 18%, which is a pretty big acceleration in marketplaces are really, really hard. Right. Like, it's kind of like gravity. Once a marketplace at our size doing around $100 billion in gross bookings a year start to come down, it's really hard to tip that curve. So this has been a pretty big feat for the team to be able to do it.
- [117:20]
B
So how'd you do it?
- [117:22]
F
You know, it's kind of funny, a number of years ago, we. So I noticed that, you know, as we got bigger, we started losing a little bit startup intensity, that startup energy. And I asked myself, like, how can we get that energy back? And we basically took a very small team. We named the team Project Hawaii. The name doesn't really matter, but basically I took a very small team of people and we said, we're going to focus on a very, very small service area. And we decided to focus on conversion rate the guest journey. And I basically tried to work with a team, like, as if we were at Rouse Street. Rouse street was the apartment that Joanie and I started. And I said, we're going to act like a startup. And the team just grinded really, really hard. We weren't working like a big company. We were very small team, grinding really hard, focusing on obsessing over the Customer experience, really looking the data and we really got a lot of points in the board. Then we really started taking these pods and we really started working with the teams, trying to coach them how we worked in the early days. And I just think the pace increase, the intensity increase. We really like tried to bring in like world class people onto the team. They got very, very focused. We try to get all the management and bureaucracy out of their way and it's been a couple of years. In fact, we're doing this last year. But one of the things is, as you know, financial results are lagging indicators, especially when you're a big company. So it takes sometimes a while to get the financial numbers to reflect what's happening inside the company. But I feel like we're a startup again more than ever before, like where we feel much younger and smaller than earlier. And I think with AI that would be the other thing is 60% of our code is now written by AI, which is twice our benchmark of our competitors and peers. And it's really, really helping us, I think is a huge boon to us. I don't know if it's helped the OTAs, but it's helped us with customer service. The customer, customer Service tickets down 10%, 40%. People who contact me, the AI solves the problem for them and we brought us the entire journey. So everything is really accelerating.
- [119:17]
B
On the special team that went in to optimize conversion rate, I imagine you have folks internally whose job was basically the customer journey already. Then you bring in your special team and is there some sort of culture clash there? How do you set people up for success to not get too political in that environment, actually see it as an opportunity for a win, Some fresh eyes, some fresh ideas. What is required to actually have success? Because I think a lot of big companies that bring in McKinsey and they put together a big deck and everyone freaks out and thinks they're getting fired and maybe some of the good ideas are surfaced, but it never really goes through. So what do you have to communicate to the team that is receiving information from this new KWAI team?
- [120:06]
F
Yeah, so it's a great thing. Really what I'm talking about has played out over like really five years. And the term founder mode, what it really meant that Paul Graham wrote was about me, like skipping layers of management and going into the details with teams and instead of trying, I mean, here's my advice. If you're a CEO or a leader of a company that's big, don't try to change the whole company. Try to change a corner of the company. It's kind of like don't renovate the whole house, Pick like one room and make it perfect and then go room to room. So I really told teams like, hey, like I actually didn't replace the team. I took the team that was already working. I like handpicked some people on the team, but I really just taught them the pace. And I would review work very regularly. So I'd sometimes review the work weekly or even daily to just teach them a level of intensity, a level of perfection. And it was unfamiliar and uncomfortable. And I will tell you that not everyone liked it. And some people didn't stick around the company because they didn't like that way of working. But those who stuck around and the vast majority did, they realize, wow, actually when the CEO is involved, it's actually easier, there's less bureaucracy. I try to make the work better and I try to clear all obstacles. And so I basically did that group by group. And the great thing is I was in massive number of details reviewing everything for years. But eventually it's like muscle memory. It's like I know an instructor or a coach, you have to teach them, but once they learn that and they have the muscle memory like a golf swing, you can step back and now it's muscle memory. So now I don't have to be in all these details, all these teams. But I would just say a couple of things. People listening. Like great leadership is presence, not absence. I think a generation of management's consulting or management school management teaching taught us that CEOs should trust their people and get out of the way. And I don't think trust and get out of the way are the same thing. Again, I think leadership is presence. And if you, if you should actually be partnering with your people, you should be on the field with them. If you're a, if you're a cavalry general, you should be on a horse, you should be on the battlefield. You're not like you're not overseas somewhere else just writing out blueprints. You got to be on the field with them and you got to be leading from example. I think leadership from the front, not from the back. And so these are just some of the things we do. And I think AI is going to create like the equivalent of an AI founder mode, which is now you got to be even more hands on. And I don't think in the era of AI there should be any pure people managers because you're so close to the details, to the data that everyone has the opportunity to be hands on. And it's hard to imagine only managing people and not agents. So I think this hands on approach is for everyone inside of a company.
- [122:40]
B
How did Jordy, please.
- [122:43]
A
Do you think a lot about what you would do if you were you for from 10 or 15 years ago and you were trying to disrupt Airbnb? Like is that a helpful exercise?
- [122:53]
F
And yeah, and it's slightly, and it's slightly scary because I think that I've told. I had a meeting with our team recently. I said that 26 year old me and Joe and Nate could, could F us up if we wanted to. And so I told them this is what I would do.
- [123:08]
A
But that's because. But that's because you understand this market better than anyone else. You know, you know the key drivers. Like I don't think 26 year old
- [123:17]
F
off the street, hey, I'm paranoid. I think that if everybody sits still, I think a different group of 20 or 30 year olds could also disrupt us. And I think that's true of every one of us. And maybe the by the way, maybe that's not true and maybe that's not reassuring because we do have a brand that's a noun and a verb. We've got a global network effect.
- [123:34]
B
Yeah.
- [123:34]
F
I think it's actually hard to build a network that may not be possible. But the software and the app, I told our team like we can't sit still. Like our app is beautiful, it's really nice. But we got to be in a world of AI native. And here's a key point I would make. I do not think a chatbot is the right interface for travel or E commerce. That might be a radical statement. ChatGPT launched third party apps last year. In March they shut them down. I don't think a chatbots right interface. It's got four or five problems. Their first problem is it's text based, you know, photos are an afterthought. The second is there's no direct manipulation. You have to type every single prompt, which is fine for a conversation, but you can't like add filters. You can't cook around. The third problem is it's hard to compare. A lot of E commerce and travel is comparison shopping. If you have thousands of options, the AI has to know exactly what you want to be able to show you one or two things. But you usually want to see more choices and you get lost. And most AI single player, it's not not collaborative, let alone the fact that Airbnb it requires people to have an account that 85% of people send a message. So what I'm trying to do is,
- [124:42]
A
Sorry, sorry to push back there but I'm. Yeah, my wife's planning a vacation and there's like three ish hotels that we're looking at and we're looking for specific dates and she. Theoretically, I'm not saying the products are there yet, but theoretically you could ask a chatbot. I'm interested in staying in this location, I'm interested in these hotels, I have XYZ number of people and it could go and the agent could go and pull together like relevant sort of like listings or room types etc. Pricing, show me pictures and then actually do an analysis of the trade offs based on all the information available as well as information in other parts of the Internet and pull it together and then she could share that chat with me and we could both review it together.
- [125:34]
F
Yes, I agree. That's the future. And that's not a chatbot we described. Yeah, that's not a chatbot. It's going to be a completely different interface. It's going to be. Well, I guess you'll have to wait and see. But I think the future are not apps. The future agents, but I don't think they're going to be tax forward. I think they're going to be really rich user interface.
- [125:53]
B
Okay.
- [125:54]
F
And so I think the current chatbot paradigm, what you're describing, it can do it. I just don't think it's the best way to do what you just described. I think there's a much more immersive way to do that.
- [126:03]
B
Yeah. As you think back, I mean it feels like at various points clearly there's been like always, is all of this going to move to chat bots or is all of is are there going to be a million million competitors that are all vibe coding exactly what you have. And so is software remote. The stock has not been beaten down during the SaaS apocalypse. But have you had to process those with investors? Have you had to walk people through Airbnb strengths again?
- [126:31]
A
I'm sure you did. After the like the 2028 intelligence crisis. Thankfully they picked DoorDash instead of you guys. But thank God.
- [126:39]
F
I really appreciate them picking on someone else for once. But yeah, no, absolutely. I think there were entire. Actually when ChatGPT launched a third party apps, their stock probably went down like 7% and by the way, I thought it was a really good idea for them to do third party apps. I think it could have been successful but they would have needed a richer SDK for it to work and it would have like, like the App Store. Apple's App Store was good for Airbnb and was good for every company because they had a really rich user interface. Maybe this is the point I'm making.
- [127:06]
B
Yeah.
- [127:06]
F
That imagine using iMessage on your iPhone to do everything when in fact like every other app has a unique interface. So what I think is E commerce. You want a very rich user interface.
- [127:17]
B
Yeah.
- [127:17]
F
It would be agentic. You could be able to have a conversation with it. You can talk to it, I could talk back to it. But I think the point is it has to be more visual.
- [127:24]
B
Yeah.
- [127:25]
F
I think a text based interface is for some solutions.
- [127:27]
B
Yeah.
- [127:28]
F
And a chat bot, that's visual. Yeah, yeah, that would work. I'm just saying today's chat bots are. Aren't the right solution. But the answer to your question, I had a couple new employees. Like one of our, one of our new team members is a name, a guy named Ahmed who is CTO of Airbnb now. He led the llama models at Metta. And one of the comments he made to me was he said, wow, yeah, Airbnb is so much more than the app. And in fact the app that you see is like 20% of Airbnb. I mean, we have typically 4 or 5 million people staying in an Airbnb every night in more than 100 countries around the world. So there's a lot more to do around payments, around customer service, adjudicating everything we have to, you know, we have a $3 million guarantee against thefter property damage for a million homes a night. That's $3 trillion. You know, there's just so many types of things around managing five and a million. Host. We have a host app, so there's a lot of things that are beyond the guest app. I actually think the guest app would be pretty easy to copy and I think in the age of AI you can make a better app than ours. And we want to make that app before anyone else does and we want to be agentic. But I think the key is most of Airbnb is not the app that you use. It's mostly the offline experience, it's the operation.
- [128:34]
B
Yeah, that makes sense. Yeah, the imessage example is good. It's even, maybe even going deeper into like old SMS because iMessage has hydrated so many things with the reactions and you paste a link and it hydrates it and it's becoming more of a visual tool. But it is a long road question.
- [128:52]
F
Things yeah, yeah. Well, before you say, one of the things is the last time I was on I did make a comment and I'll make it again. Almost every company is enterprise company. And the last stat I think I said on TVPN was like three months ago and stats haven't changed. I think it was 175 companies in a YC batch and I think 16 were consumer. That trend hasn't changed. Almost everyone is going into coding, almost everyone's going to enterprise. I think that's great. My whole point though is that the consumer experience hasn't fundamentally changed that much beyond a chat bot. And I think the consumer is a massive opportunity for AI and I think it's going to need to be a richer user interface. I think people that want to do more things, they wanted to be. I mean the modern AI today is text, photos and buttons, mostly in some videos. And I think there's a more breakthrough visual paradigm that can be much more immersive. And with the new image and video generation models, you could do something so much more immersive. So the exact example you gave a trying to book a hotel, yeah, you can do it on chatbot, but there's probably some more breakthrough way to imagine that, to visualize it, to see the neighborhood, to see the map, to be able to talk to it, to understand where it is, to be able to compare photos, different hotels. There's something I think richer on the horizon.
- [130:04]
A
Yeah, that was, I was, I was surprised at the push into browsers last year from some, some different AI companies specifically because I always, I already felt like an LLM, like the chat, the chat bots were helping me browse around the Internet. It was like just double down on that and helping me find information all over the Internet. I like that it's brought together in a standardized way. Right. If I like the example, you know, of comparing like different hotels. Right. It's nice to have it just like formatted the same way. So I'm looking at it. I'm not being like influence. I want to be influenced by the images and like the property, but not necessarily the design of the website. Right. Because that they can be disconnected. I wanted to ask if you think LLMs will impact travel in the world in the way that social media did in travel trends, specifically because Instagram nowadays, like a place like Marfa and like Texas, right. It's like it's random town or you know, artistic type town in Texas and then it just becomes a cool place because of Instagram and then, and then, and then the entire kind of town evolves because of that. I can imagine, like people researching like places to go with LMS could ultimately drive some of that and then ultimately reinforce each other. But. But what, what do you think?
- [131:26]
F
Yeah, maybe the simple framework is think about a travel journey. Step one is destination discovery. Where should I travel? Step two is flights. How do I get there? Step three is where do I stay? Urban beer hotel. Step four is like what do I do when I get there? Restaurants, activities. And step five is typically like logistics, car rental services. And then step six is you're in the city and then you might want to do things that time spontaneously. I think lm, or let's call it the LM technology applied to, I'm arguing a slightly richer user interface than typical chatbot will be revolutionary for step one. Step two for destination discovery and flights also because flights are not very hard to build. There's just three global distribution systems. Anyone can pipe in an API and have a flight booking app pretty, pretty quickly. And so I think the LMS are really, really good at destination discovery. If you want to say like, hey, I want to go somewhere that's like Paris, but it's a little more affordable, it's good to go in August. I like Opera. Like it's going to give you like a very rich like suggestion of places to go. Now I think the current chatbot doesn't have really rich maps. I think it could do a lot more visually. I think eventually chatbots or this new interface could be very video based. So imagine a chatbot that was actually video based or very, very photo based rather than a little less text based. I think that would probably be what you'd want for travel. I think what I'm describing would disrupt travel more than the current chat bots which are more acting like Google sending referral traffic and they actually, the traffic from these chat bots are converting higher than Google. And so actually the chat bots are actually additive to travel companies, to remediate travel company. You would really want to disremediate the travel journey. It would have to be much richer.
- [133:10]
B
Yeah, I feel like. Do you remember that company Hipmunk? I think it was founded.
- [133:14]
F
Yeah, it was really cool.
- [133:16]
C
Yeah.
- [133:17]
F
Actually the CEO of Reddit.
- [133:18]
B
That's right, Steve Huffman. Right.
- [133:20]
A
Wait, are you talking about hip camp? No, Hipmunk.
- [133:24]
B
Hipmunk was Hipmunk.
- [133:25]
F
It was a really cool flight booking.
- [133:26]
B
It was a really cool flight booking dashboard. And it would rank the flights by pain, like a pain index. So it would say, well, you're not gonna face financial pain because it's a really cheap flight, but you do have a stopover or you do have to get to the, the airport super early. And so it would blend all these things together. And it's something that it feels like could be vibe coded over a weekend now. And yet it is weird that when I open up the app store charts, I see chat, cbt, Gemini, Claude, and then it's just AI chat apps. And we haven't had that breakout moment. Demis was on stage at YC actually talking about the fact that it feels weird that we have this super intelligence in the enterprise and vibe and 60% of Airbnb's code is written by AI now. And yet we don't have like a new AAA game or a new video game that everyone's playing. And I'm wondering, like, is that a lack of creativity? Is it a lack of risk taking, that there's just too much money, it's too obvious to be able, if you're good in AI, just go into enterprise because you'll just raise a bunch of money, get a bunch of customers. It's really easy to make money and it's more risk on in consumer. Like what do you think needs to happen? Or is it just the intelligence isn't quite there and maybe the next model is the one that unlocks it?
- [134:40]
F
Was this such a good question? I think there's like three or four factors going on. The first factor is I saw a tweet recently that I think there's like 60 new Neo Labs being formed.
- [134:51]
B
Yeah.
- [134:51]
F
And I've met a number of these researchers and a lot of them are interested in doing the same things. And I think what I'm noticing is more most of these teams are purely AI people.
- [135:01]
B
Yeah.
- [135:02]
F
They don't have product people or designers on their team are picking things like science, they're picking things like coding, they're picking things like we want to create a different type of model. This is great. But I don't know if all 60 companies should be doing the same problem. And so what you're seeing is you don't see a big focus on people wanting to do consumer. I think some people think that the AGI can just figure out consumer. I think that's maybe simple, overly simplistic thinking. I do think you have to have a point of view about consumer. So I think that's the first point. I think the second point is, you know, Silicon Valley has become more vibe and trend based than when I came to Silicon Valley. 2007. Although it was back then. And one of the things I think when people see enterprise companies doing well, they go into it and I think there's this like natural flywheel. The other thing is, I know in Y Combinator for example we teach the Y combinator companies use the other companies as their customers.
- [135:53]
B
Yeah.
- [135:53]
F
Now back in the day we would get them to use our product as consumers, but I think everyone's figured out actually it's way more efficient to get them to, to become customers.
- [136:02]
A
Like 5 code. 5 code gen startups all being like you code with me.
- [136:08]
F
Yes, exactly.
- [136:08]
B
Yeah.
- [136:09]
F
And so I think enterprise makes a lot of sense until everyone does enterprise. And then suddenly like everyone's saying consumer is hard but you know, it's also hard like competing with 10 other companies and enterprise doing your idea too. That's also hard.
- [136:22]
A
Yeah.
- [136:22]
F
I think additionally the image in. In video generation models are having a. I think Image two is almost like a cloud code. Breakthrough is so good now in the sea dance video models that I think we're about to enter this new era where suddenly going beyond text is possible. I think you're going to even be able to see real time AI generated interface and next year or two. So my prediction is this has been the era of enterprise, even chatbots breakout success. I would predict that most of revenue is going to be going to Codex. That's my. I might be wrong, but just given how much money is going to cloud and I think in the next two years you're going to see a massive revolution in consumer and I think you're going to need a couple of companies to lead the way. I think we need reference points. I mean even me saying the chat bot is not the interface people probably asking well what is? And someone needs to be the one to do that. I mean we're going to try to do our part. I'm not sure Airbnb will be the company to blaze the trail, but I think someone has to. I think in another era it would have been Apple, like the Steve Jobs era. Apple would have done that. Maybe Apple will do that, but they would have been typically the company that would have done that.
- [137:31]
B
Yeah.
- [137:31]
A
Are you worried that everyone is working on the exact same thing? And when I say exact same thing, I mean agents that can do things on your computer and or you know, look at the iPhone app store chart or look at, look at the early stage startup market. Right. A lot of companies, you know, the full spectrum of companies from hyperscalers to labs to Neo labs to I'm Not.
- [137:59]
F
I'm not too worried. I might be a little worried for the entrepreneurs if they're like the 10th company doing something. I mean, like Peter Thiel was one of our earliest investors and he used to. He wrote this book zero to one that a lot of people listening probably read. He has a saying, competitions for losers. And you want to kind of try to do something that no one else is doing. Back in the day, that was enterprise because everyone was making an app. And now I just, I had to be careful about being the 10th company doing something unless you're going to be a lot better decisively. It's just really, really hard in a crowd and market. So you want to kind of zig when everyone else is zagging. So I'm not too worried, but maybe a little bit. And I would just encourage entrepreneurs to try to claim a space for their own. And by the way, Airbnb was that it was kind of accidental. But like in 2007, everyone wanted to do like a social networking something. And we were like, it was tempting for us to try to be a social networking something something as well. It was accidental that we inflated through airbeds one weekend and created airbed and breakfasts. And people thought it was the worst idea ever. I guess it became like the worst idea that ever worked. But we carved our own space, as did Uber, and we weren't trying to be anything else. And so I kind of. And by the way, OpenAI wasn't trying to be anyone else either. Maybe they were a little bit like DeepMind, but I was not the thing when OpenAI Anthropic got started. So I do think there is something about not chasing trends. I think once it's a trend, it can be a little crowded and to try to claim out your own space. And I would say there's so much of the US economy that AI hasn't yet touched.
- [139:34]
B
Yeah. On that note, how are you talking to. Or how would you talk to designers or RISD students about applying non technical skill sets or non AI disciplines, non computer science backgrounds to consumer. The reason I ask it is I feel like there's a lot of pushback against AI in design communities in colleges right now. Broadly, AI is not particularly popular and there's a lot of fear of job loss. But at the same time, I'm sure the tools have changed throughout the history and I feel like you believe that there's an incredible opportunity. But I'm worried that some people are not jumping on that opportunity because they have hesitations about various. Oh, well, does this displace this tool or how was this made or is this the right tool for the job? Should I even be using this?
- [140:41]
F
Yeah, I'm really worried that an entire generation of designers, artists and creative people are going to decide to kind of sit out AI. And I think it's the biggest opportunity for creative people in my lifetime. By the way, this is not the first time designers were late. We creatives and designers were really late to the Internet and web. So going back 30, 35 years, most of the prestigious designers did not get into so called web design design. Web design was considered a lower tier, lower status design and they stayed with print, all the established people. And so they did not really chase design. What ended up happening was because you had a lot of these really excellent designers that did not go into web design. I think what happened was a lot of, there were a lot of good people that became web designers, but I think there became a gap between web design and engineering. You had people that, you know, maybe didn't have the full, the full design skill set. And what happened was this function emerged called product management. Product management, I'm not arguing against we have product managers, very important. But in industrial design there's no product managers. For the most part, they're industrial designers. In architecture, the architect is the product manager. And so I think what ended up happening in web design designs are very, very narrow. There was a void, product managers filled that void. I'm not arguing against that model, but I think if designers sit out, we're going to see see as engineers and product managers designing for them. And I think the counterpoint is designers can be engineers and product people for intensive purposes. Said differently, if I were starting Airbnb today, I'd be vibe coding in cloud code. And I'm trying to do it now. I'm doing it for fun. I'm not really going to be doing anything productive Airbnb, but if I was 26, I would have thought of myself as a technical person. I think all designers should think of themselves as product people and as front end engineers, if anything. And so I think this is a boon. I don't think that the future of the Internet has to be all text based. I think it can be very visual. I think the photo and video generation models allow you to design incredibly rich interfaces. And this is, I think, the best time in the world for designers and creative people to get involved. I think they're just a little bit afraid and whatever we can do to enlist them to just get their hands to sturdy. Many Designers are. I just think more should get involved and not begin getting out of typical wireframes. Just start coding.
- [142:58]
B
Do you think? So? Assuming we solve this problem of not enough creative people working in consumer. We build, we as a society build cool new consumer products and experiences and apps. There's something that deserves to be at the top of the App Store. Store has distribution and the acquisition of customers and users changed structurally. You were very good at SEO and Google and there was a referral program. Do those patterns still work? Are they broken? Is there a new playbook that needs to be rolled out? For anyone who has something that's great, but they need to get it in the hands of consumers broadly, I think,
- [143:44]
F
I think we need new patterns. This is a funny saying. I think the highest turnover job of any executive at Silicon Valley might be the cmo. I don't know for sure, but like, you know, you don't see a lot of turnover CFOs. You don't see a lot of turnovers of CTOs. Yeah, but like Amazon was famously. And I don't know if part of my theory is that like what works in marketing changes every few years and you have to be adaptable and you're old playbook gets outdated. So marketing is unfortunately one of the hardest functions. I have a huge amount of respect for people marketing because once something works, it almost becomes stale because then everyone does it. And so like influencer marketing was really successful until everyone did it and then people kind of tune it out. It's this thing called banner blindness. Banner blindness is after you see something over and over, you tend to be blind to it and so you need a new tactic. So I think that a couple of thoughts. Number one, I think that just like we need new interface design, new creative experiential approaches, we also need new types of marketing. I don't know off the top of my head what that is. It's probably something we haven't done before. I mean, Joe and I, we did like bizarre things. Like we sold collectible cereal boxes and we like when social media was new, we were all over social media. And when, when there were newspapers, we would hunt down reporters and like try to get them to write about us. But people ask me, how do I get users? I'm like, all the tactics from 2008 aren't relevant anymore. So you gotta be relevant and you gotta find new tactics. The second thing I'd say though is you're right, the distribution is mature. When we started Airbnb, the App Store was young. In fact, the Internet was still young. I mean we could just, you could launch a website and just ride the growth of the Internet. And it's hard. The Internet is not growing like it did before. In other words, people were coming onto the Internet and hundreds of millions a year new users. So I do think distribution for consumers mature at the same time. The top apps in the app store are new apps. They're mostly chat bots. So what that tells you, if you do something truly breakthrough and revolutionary, consumers will still probably find it. And so that's, I think, yeah, I think the two principles are do something so revolutionary, consumers will find you. And number two, you're going to have to find your own new tactics. And you know, anything that is standard is probably stale.
- [145:59]
A
Well said.
- [146:00]
B
I have one.
- [146:01]
A
But yeah, looking out into the future, you've ridden the growth of the short term rental market. You dominate it now is is the biggest opportunity to just continue to make the best product in the category, ride the continued growth of it. Or do you think there's another str size market for Airbnb?
- [146:23]
F
I think the biggest opportunity for me is to go beyond our core business. I do think, you know, core business does close to $100 billion in gross sales. If you net out all the other businesses gross booking value, the total amount of reservations going through the site, I think that could probably double one day. I don't know how long that one day is. But for every person who stays in Airbnb, eight or nine stay in a hotel, I think we can get one extra person in an Airbnb eventually. And I think that can get you to 200 billion. I think there's a market much larger than Airbnb, which is hotels again, hotels about eight or nine times the size of Airbnb. I don't think we'll ever be a hotel dominant site, but we are going more aggressive into hotels. A fun thing is that about half the hotels in the world are independents and boutiques. They're not chain hotels and they're not really happy listing on the OTAs because they pay a higher commission. The chains. A lot of these independent hotels are being forced or they feel like their hand is being forced to franchise to Marriott, to Hilton, other brands because they have loyalty programs. They can, you know, they can negotiate lower commissions. So we think we can be a distribution channel for these boutiques independents. That is a multi billion dollar market. I think another one would be services. There's no Amazon for services. And you know, think about like you can hit a button and A car can pick you up, you can hit a button in of front, food can be delivered to you. But what about hitting a button, having 80 other things possible? You know, I don't know if any one market is large, but if you add up the 80 different service verticals, that to me is another pretty big market. And then maybe the last one is like living stays longer than 30 days more and more people have a job via laptop. More people are nomadic, more people are moving around. That's another really big market. So I think for Airbnb, we're looking at really expanding, expanding to a lot of different categories. And I think that's where most of the growth is going to be in the future.
- [148:09]
B
Do billboards in San Francisco work on you going back to marketing? It's one of these old things where like, yes, I completely agree with you. Like the CMO who did the first billboard campaign back in the 50s, probably printed legend books written about them.
- [148:25]
A
Well, the billboard campaign that we're fell off now, couple years was just buying so many billboards that people are like, wait, how did this company buy so many billboards?
- [148:35]
B
And so I'm wondering, is your, is your thesis on marketing that these things go in cycles? And yes, we might be out of the influencer marketing meta right now, but it's going to come back in five, 10 years. Who knows that's possible? Or is it rise and fall and then never again?
- [148:49]
A
I think these channels though, they don't go away. They just stop being like, if you do this thing, you're going to grow like crazy.
- [148:56]
B
Yeah.
- [148:56]
A
Still need to do that thing, but it's no longer an art.
- [148:59]
B
Okay. Yeah. What do you think?
- [148:59]
F
You guys are okay. So, so one thing I have to acknowledge.
- [149:02]
B
Yeah.
- [149:02]
F
Is I think all of us advertising works better on us than we admit.
- [149:06]
B
Yep.
- [149:07]
F
I honestly, I, we, we spent a lot on like all these ads and I kind of think to myself, does do these really influence people? Like, and if they do, and if they didn't, we wouldn't be spending like a billion dollars here in advertising. So I want to admit that advertising does influence us more than we realize. There's, I think a lot of studies that show like, oh, that doesn't change my mind, but you see something seven times and it probably, even if you don't want to believe it, you might start believing it.
- [149:29]
B
Yep.
- [149:29]
F
So clearly it works.
- [149:30]
B
Yeah.
- [149:31]
F
I do think though, the ROI of new ideas works a lot better. Just for example, we spent a lot of advertising, but the most popular marketing we've ever done Is like when the Barbie movie came out, we took a house in Malibu. We turned the Barbie Malibu Dream House, and it was like, new. And the whole Internet talked about it. And the ROI of that was better than any ad we've ever done. Right. We've done things like that. So I do think doing kind of crazy, slightly unhinged things that people notice. Like, I mean, you guys are successful because you're different. Right. I think. I think being different is the key to marketing. And so if you have a billboard, people are probably influenced, but if you. If it looks different than another billboard, it's going to work. So I think the key to marketing is to be different because you got to stand out.
- [150:14]
B
Yeah. Yeah. The Red Bull Baumgartner, the skydive from space is, I think, like a thousand x roi. I think they spent less than a million dollars on that, and I think it has a billion views or something like that Space. Great example of that space.
- [150:31]
F
But if, like, people did that, then
- [150:33]
B
the tunnel wouldn't work as well, and that's probably why they didn't do a sequel. And then another one. Another one. They went and found something else. Let's put a plane through a tunnel or fly a, you know, a hot air balloon upside down or something.
- [150:43]
A
Get a blimp. Get a blimp on Airbnb that I can go on a sky cruise.
- [150:47]
B
Sleeping in a blimp.
- [150:48]
F
Yeah, so that would be. That would be. That would be a very good idea. Maybe we can talk about.
- [150:52]
B
I'd love that.
- [150:53]
A
Ship it. We'll podcast.
- [150:55]
B
Yeah. I don't know how long it takes to get to. Yeah. La. To San Francisco. If it takes me two days, three days. But I have a beautiful view.
- [151:01]
F
Would you guys do that?
- [151:03]
B
I would 100% do that for us and you. So. So. So we say things like this all the time where we're like, oh, yeah, like, we'll definitely, like, got your do something. And oftentimes it's hard to schedule. But we have been talking about blimps for, like, over a year.
- [151:17]
A
We know we can broadcast. We've been having to happen with a blimp. We've been having guests call in from their jets with Starlink, and the connection is perfect. It is so much so that people in our chat sometimes don't know that the person's on a plane.
- [151:32]
F
Okay, you're. Airbnb and TBPN are going to work together on a blimp partnership. You guys are going to broadcast across the country on a blimp.
- [151:40]
B
I love it.
- [151:41]
F
He's going to be a part of that.
- [151:42]
B
Yeah.
- [151:42]
F
I want to make this work.
- [151:43]
B
Let's, let's, let's, let's figure it out. Fantastic. Well, thank you so much for coming. Have a great weekend, and we'll talk to you soon.
- [151:50]
A
You're the man.
- [151:50]
B
All right.
- [151:51]
F
Bye, guys.
- [151:54]
B
We're going on a blimp, baby, for blimp maxing. Let's go. We've been waiting for this for so long. We've been thinking about blimps, what we could do with them, and we finally have the perfect that.
- [152:05]
C
I couldn't be more excited. Thank you for tuning in to TVPN today.
- [152:10]
B
It's a fantastic show. Thank you to our guests and our audience. And leave us five stars on Apple podcasts and Spotify. Sign up for our newsletter@tvpn.com and I
- [152:22]
A
was supposed to cut to John.
- [152:23]
B
There we go.
- [152:26]
A
All right. Anyway, we love you guys.
- [152:28]
B
Sign up for our newsletter. Leave us five stars. Thank you. We'll see you on Monday.
- [152:31]
A
Have a great weekend.
- [152:32]
B
Goodbye.