Big Tech Earnings, Red vs. Blue Button, Quantum Computing | Jason Yanowitz, Even Rogers, Maria Spiropulu, Stepan Simkin, Kashish Gupta, Dan Magy, Vlad Tenev, Parag Agrawal & Andrew Reed, Gabriel Stengel

You're watching TVPN. Today is Wednesday, April 29th, 2026. We are live from the TVPN UltraDome, the temple of technology.

The fortress of finance.

The capital of capital. What a day in financial markets and technology markets. There is a ton of news today. Gur Gaven summed it up well. Huge day tomorrow. He's posting this yesterday. 10:00 AM Canada interest rate decision, 2:00 PM, this is Eastern time, USA interest rate decision. That is in Fed-held rates constant. So sort of a nothing burger, I guess. Not the best news. I think some people were hoping for a cut, but everyone sort of expected this. It met with, it met expectations. But the news is that just 5 minutes ago, the Federal Reserve has kept interest rates on hold at a range of 3.5 to 3.75% at Wednesday's meeting. And, Jerome Powell is giving a speech in just 30 minutes at 2:30 Eastern. Then at 4:00 PM, Google earnings, Amazon earnings, Meta earnings, and Microsoft earnings. It is investing day.

And no big deal, no big deal. They only represent just under 20% of the total market cap of the S&P 500.

Let's go.

All reporting within—

But people are optimistic. SemiAnalysis put out a note this morning, according to YC, Yield Chad, expecting hyperscaler CapEx to be revised upwards and beat street expectations as hyperscaler cloud revenue is accelerating and they are seeing positive ROI on cloud investments. I can sort of go through my earnings preview. It's a tech earnings quad kill today. And the big question is just how is the AI buildout going? Obviously, the CapEx numbers were huge. We saw the first time ever we've seen a $200 billion number from Amazon last quarter. But everyone's up in the Is there even a word for triple? It's like, what, 11 digits or something like that? $100 billion. Wow, digits. Everyone's spending at least $100 billion these days, at least if you're in the Mag Seven. And, but financial performance has actually been strong even in sort of legacy areas such as search. Google Search is growing, e-commerce sales, Amazon core business is growing, enterprise software seats, Microsoft 365. We're gonna have more news on all of that. And, But even though all of those businesses, they're working, they're chugging along, the big question is around durable revenue tied to AI infrastructure because you have this matching problem. You spend a bunch of money, there's depreciation. When do you actually get the cash flow back? How durable is this revenue? What are the moats around this revenue? What do growth rates and margins look like in this new era of something that looks maybe a little bit more like a railroad business, an oil business, as opposed to something like you build a website, people just show up, and it's 80% margin, which was the dream of the previous software era. We're going into an entirely different era, but it is very exciting, and we're getting a lot of data today. So everyone has seen cash flow from these hyperscalers to NVIDIA, to power companies, to data center builders, but everyone's wondering what's the exact conversion cycle to higher revenues and higher profits? What's the pathway there? Because you don't wanna just be drawing down on cash endlessly. Eventually you stop making money entirely. So let's start with Google. Google has the most fully integrated AI stack, arguably. They have consumer distribution, they got model training with DeepMind, they have custom chips with the TPU, and a bunch of product services where they can stuff AI features.

Google Workspace.

Google Workspace.

Underrated.

Search, YouTube, Android, cloud, they can deploy solutions all over the place. And so the flywheel should be spinning very, very quickly. The key question that investors are asking is, does AI change the unit economics of search too quickly? Does AI— do LLMs, do AI search overviews, Gemini generally, are they able to monetize those results fast enough to offset any potential declines in search ad revenue? And so people will be looking for How is search monetizing? How is, how are the new AI, you know, disruptors monetizing? And so, but there are tons of places to pick up growth even if growth does slow down in the core search business. Tons of opportunity in cloud. But the question again is AI, Overviews, and Gemini, are they expanding search usage? Are they increasing ad ROI or are they compressing the model, the financial model? Microsoft is also coming off a strong quarter. Honestly, everyone's coming off strong quarters. Everyone's doing very well. Revenue's up 17% at Microsoft with cloud, which includes Azure, M365, some LinkedIn stuff. It's a big bucket. Cloud, Microsoft Cloud's growing at 26%. But if you dive in and you double-click on Azure, Azure is growing at 39%. And of course, Azure is a bigger lever on CapEx, which is run rating around $150 billion. Not bad. The biggest number in the quarter in the Microsoft earnings is RPO, remaining performance obligations. Last quarter, it was listed at $625 billion, up 110%. That was the eye-popping number of the last earnings. About 45% of that is coming from OpenAI, but they have lots of other partners that have signed on for really long compute contracts. And that is starting to show up in the financials, saying, hey, we're spending all this CapEx, but we have this RPO, and we have these deals signed where companies are— it's not just us that we're forecasting some really high growth here. The entire industry is forecasting high growth. And so we have done deals to justify the CapEx that we're spending right now, even with depreciation, which seems like less of an issue than people thought it was since H100 still seemed to be monetizing just fine. But we can go into that. So Microsoft has the cleanest read on enterprise AI monetization, which I think is something people have been really looking for, looking for numbers. They know that, you know, subscription LLM chat apps monetize at a decent rate, that the margins at a lot of these companies are okay. Okay, the token, the APIs are working, but what does it actually mean to deploy AI into the American economy, into everyday businesses, into large-scale businesses? All of those businesses are on Microsoft overwhelmingly. And so there are a ton of data points that can help you understand how AI is flowing through the global economy, but also the American economy. So what are we talking about specifically? Azure growth, how much cloud hosting is going on? Gross margins for Microsoft Cloud, that's very important. Are you seeing cloud compression? Because if you go you go to your Microsoft cloud provider and you say, "Hey, I'm gonna, this Teams thing, I'm paying a lot of money for it. I actually vibe coded something. You gotta gimme a discount." That would show up in cloud margins. Will it show up? I don't think it will, but we'll see. I think it's gonna be fine. Copilot adoption and ARPU, how much are they actually rolling these out? Are they actually getting incremental spend? Are companies willing to send more of their hard-earned dollars to Microsoft for better AI services, better features, Copilots? Then M365 seat growth, that's a really important one. Are people adding more seats? Are they hiring? Like we've seen some layoffs in big tech, but how is the overall economy doing? How many more seats are being rolled out? Again, with the question of like, does your AI agent need a seat? And so you have more seats or does your AI agent replace 20 seats and then you only have 1 seat and you have 20 agents that are all logging in through the same M365 seat. These are like more long-term questions, but we're getting an early read. Today. GitHub Copilot momentum. This will all paint a picture of what is happening with AI adoption in enterprises broadly. Interestingly, Copilots at Microsoft, it hasn't been the most hyped product. GitHub Copilot, early, very, very early to the party, huge run rate, you know, rocketed up to $500 million ARR very, very quickly. But the horse race has always been, you know, Windsurf Cognition and Cursor and Codex and Claude Code and like the battle has been The startups for the most part, Gemini's been in there. GitHub Copilot has felt like it hasn't been dominating the narrative, but we're gonna find out, is it still growing? Because the market is really, really big. It's possible that everything's growing even if there's, you know, a horse race back and forth between the leading labs. And so there will be a bunch of other questions answered around, you know, how nuanced is the diffusion adoption question? So Microsoft has an incredible go-to-market team, incredible go-to-market motion, enterprise sales motion. And so is there an advantage that they have that they can press GitHub Copilot, even if it's not the sexiest product next to whatever's hot this week, can they get that into people?

Yeah, you look at Slack and Teams, right?

Same thing. Yeah, yeah. Slack was definitely like the hot one, the hyped one, and Teams wound up doing very well. We'll get a stronger read on what's going on there. And then of course that M365 seat growth will tell us a lot about, you know, what does the future look like for seat-based enterprise software, the SaaS model broadly? Because if all of a sudden that's falling off a cliff, well, it probably doesn't look good for other seat-based SaaS companies. For Amazon, everyone wants to see strong AWS acceleration to justify the Mag Seven topping CapEx numbers. They guided to $200 billion in CapEx in 2026. So expect a lot of focus on AWS revenue growth and margins. Q4 was healthy, though. Net sales up 14% for Amazon. AWS growing at 24%. The sneakily huge ads business over at Amazon grew 23% to $21.3 billion. They're almost making $100 billion a year just on ads. That's remarkable. And that's a lot of cash flow to fund CapEx and other, you know, AI initiatives. And so operating income overall was $25 billion for last quarter, and they generated free cash flow of $11.2 billion. So still huge cash flows. They're obviously drawing down on those, and that $11.2 billion number was down because of increased AI spending. So if AWS accelerates, all the CapEx looks like buying scarce capacity ahead of demand. Like, they will be GPU-rich or just compute-rich generally at a time when it's good to be GPU-rich and compute-rich, and they will look like geniuses. So people are, uh, hoping for, uh, strong AWS acceleration. Consensus for AWS revenue is around $36.7 billion, uh, with growth in the mid-20% range. But the market's really hoping that it starts with the 3. Everyone's hoping for 30%, something like that, at least this year at some at some point, you know, reacceleration would be a treat for the market. Lastly, Meta. Meta's in an interesting spot. Super clean Q4, nothing really to prove today despite all the FUD, all the fear, uncertainty, and doubt around the talent wars, new team members, the new lab, the sometimes clunky model releases. They didn't get Behemoth out. What, you know, different adoption of Meta Vibes, the video app. Like there's been all these questions about the AI strategy. All of that needs to be put to the side in the face of just— AI is already working for Meta and their ads platform. Let's get back to Reels. Yeah, they're— I mean, they're absolutely crushing it. So 3.58 billion DAUs, daily active users, and they grew the ad business a ton. And so the revenue overall grew 24% to almost $60 billion last quarter. Ad impressions rose 18%. Average price per ad rose 6%. Family of apps operating income was $30.8 billion, which makes losing $6 billion at Reality Labs like quaint. It's just like, who cares? It's totally worth taking a side bet on the future of devices and maybe you get a platform out of that. Makes a ton of sense. And so CapEx was $72.2 billion last year. The guide this year is somewhere between $115, $135. I think we might see that tighten up today because it's a little bit wider than some of the other hyperscalers that are targeting. But clearly, a near doubling, almost doubling of CapEx.

What are you laughing about?

There's one possibility where he goes, you know, he blows it out.

$250 billion.

We can't, we can't, we can't count that out. We can't count that out. I mean, he's got— they came out with a solid— they came out with a solid model.

Yeah.

He's putting mirrors in space.

Yeah.

He wants to be a player.

Yeah. Yeah. Yeah. I mean, a lot of focus is on the new meta models, but all of that that it just financially, at least in the quarterly earnings, like it will just take a backseat to what's going on with the AI in the ad placement, ad monetization funnel because that's where Meta makes so much money. And so the question is how much more juice will AI bring to the ad business? Expectations imply revenue growth of around 31%. And the question's always the same, Is there enough incremental growth to justify the CapEx? An AI advancement delivers better ad performance basically immediately. And this is the good news for Meta. So improvements in AI move the needle at Meta incredibly quickly. Like a new model can come out and there can be some breakthrough in like thinking, reasoning, coding agents. And if you're in the enterprise software world, it can take time to do a deal, roll it out to different developers, different organizations, change management, understand the guardrails, the security. How does this actually flow through an enterprise to drive incremental demand? Not so at Meta. If there's a breakthrough and they have a model that is placing ads more effectively, they can quickly A/B test that, run that across the entire family of apps, and you don't even notice it as a consumer. You're just like, oh, Like I was shopping for a car and I saw an ad for a car. I was shopping for a shirt and I saw an ad for the perfect shirt. And the ads just got a little bit better. And just a little bit better means like another $20 billion in profit. So it can be— so they just don't have a diffusion question. They don't have a diffusion problem. There's no, oh, we have this amazing model. If only we could get people that open Instagram to use it. Use it. Like there's nothing there because they use it when they scroll and they see an ad. And so that's an incredibly fortunate place to be in. So in summary, every question's trying to, every company is trying to answer the same question. Can you turn AI CapEx into proprietary distribution, higher customer retention, measurable revenue growth before the depreciation catches up? It's the Super Bowl for big tech. So get your popcorn ready. Everyone's excited. There are, uh, there are AI worries, AI jitters in the Wall Street Journal. AI worries have returned to Wall Street. Now come earnings. And so everyone is going back and forth, uh, on the OpenAI news. Shares of Oracle, CoreWeave, and SoftBank, uh, slumped, but CoreWeave is up majorly today, 8.4%. Uh, they, they dropped at least 4% after hours yesterday, but then they're back up. Bit. That revived investors' worries that technology giants' massive investments in artificial intelligence won't produce the blockbuster profits many expect. The report was particularly jarring coming ahead of key earnings from major players. Alphabet, Amazon, Microsoft, Meta all scheduled to report Wednesday, with Apple following on Thursday. We will be at Stripe Sessions on Thursday, but hopefully we can cover that Friday. And see what happens. A run that many expect to test a rally that has helped carry major indexes to recent records. Tuesday's decline weighed on the NASDAQ Composite, dragging the tech-heavy index back 1% from a record hit the prior session. The ice is thin. The leash is very tight, said Dan Morgan, portfolio analyst at Synovus Trust. Any evidence that would come out that would add doubt about OpenAI, Anthropic, or any of these companies is obviously going to create a sell-off. Morgan said he hadn't adjusted his positions on Tuesday and didn't think investor concern was broad-based because companies including IBM, Texas Instruments, and Intel had reported strong earnings in recent days. Instead, Tuesday's losses centered on the companies with biggest stakes in OpenAI's business. OpenAI had forged close ties with these companies to secure funding and gain access to computing resources, which are essential for training AI systems and providing answers to queries and executing user requests. And so, uh, Oracle and CoreWeave, we talked about, fell. Um, OpenAI has defended its financial footing and said its leaders are aligned on securing computing resources. The business is firing on all cylinders, and the mood internally is incredibly positive, the company said in a response to the Journal's article on Tuesday. And so there are some more takes on the timeline. Colin Tedard says, "My children might not eat tonight if this doesn't go well," because he is excited for all 4 of these earnings on Robinhood. We have Vlad from Robinhood coming on the show later today. He is joining at 1:10 PM Pacific.

Jim Cramer says, "Intel is such a horse." I have no bear case. I have no bear case.

Intel is up another 10% today.

Yeah, up 10%.

—today.

Wow. I mean, the—

40% on the week so far.

40% on the week.

Is that good? Wow. Based on everything you've seen so far in your 6-month career, Tyler, is that good? That seems incredible.

Looks pretty good to me so far.

Yeah, I mean, there is a— there's like the idea of a CPU shortage generally is just being digested by the market right now. I think it took— I mean, I remember Nat Friedman and Daniel Gross going on "Stratechery" and talking about after ChatGPT saying like, yeah, NVIDIA seems like it could be important. And then it like quadrupled in value. And this tends to happen as people digest the new reality of demand around certain pieces of the AI supply chain. And it certainly seems like there's a lot of computing to do. There's a lot of computing to do. But let me see what else is going on. We need to talk about the button.

And we have an interactive game set up.

Is that— We do. So there has been a viral— I don't know. What is it? A brain teaser?

It's like a poll.

A poll, but it's a thought experiment. It's a thought experiment. It's not purely a poll. It's a, it's a brain teaser. It's a thought experiment. And did this— has this existed before? Did this start with— it did— this didn't start with MrBeast, right?

No.

So it started with Tim Urban.

Yeah, Tim Urban. I don't think Tim Urban like created this either, but I think his post was the one that first went like really big. Okay. But I don't think this is new. I feel like I've heard about this before at some point. Yeah. But Tim Urban was the first one to make it like go viral this —cycle.

Yeah, so he got 24 million views on April 24th, and then MrBeast came from behind 4 days later with 10 million views on a very similar question. Let's start with the original Tim Urban question, and we will ask everyone in the chat to participate and vote for what you would pick in this thought experiment. So here is the question posed by Tim Urban. Of Wait But Why. Everyone in the world has to take a private vote by pressing a red button or a blue button. If more than 50% press the blue button, everyone survives. If more than 50% of people press the blue button, everyone survives. If less than 50% of people press the blue button, only the people who press the red button survived. Which button would you press? What are you pressing?

Red or blue? So the sort of like normie answer is you immediately just go blue because it feels like a lot of people would go blue. But like the safe answer is you just like, if you're hyper-rational, you'd be like, obviously you hit red because if everyone's just super rational, everyone hits red. Red, and then everyone's all good. You don't hit over 50%, right?

Yes. Yeah, because like no matter what— so if I press red, no matter what anyone else picks, like I survive, right?

Yes.

But there is a risk. But, but like, you know, if you want everyone to survive, then you need 100% of people to pick red, right? Yes. Yeah. But okay, so, okay, so originally I was blue. Yeah. I think I was like complete normie, right? And then I went to Jordi's take. Yes. I was like, wait, no, that doesn't make sense. I mean, we'll see what happened in the poll, right? Yeah, yeah, yeah. But, you know, I was like, okay, if I'm thinking about this super rationally, I should just pick red, right? Then I save myself 100% of the time. Yeah. But then, like, if there's some people that pick blue, then, like, they're gonna die for sure. Yeah. Like, I'm basically voting against them.

Yeah, you're sort of—

you're complicit in their demise. If you want everyone to live and you're going for red, you need 100% of people. But if you're going for blue and you want everyone to live, you only need 51% of people, right? Yeah. So you need to convince way less people.

Quickly, let's tell the chat how to vote.

Yeah, so 1 is for red.

1 is for red.

I'm gonna turn Ben here. And 2 is for blue.

2 is for blue. So type a single character, just a 1 if you wanna vote for red and just a 2 if you wanna vote for blue. And yeah, this has created a ton of discourse. And it's interesting to look at the difference between, uh, the difference between— oh, do we have this up? Oh, we are tilted towards red. This is crazy. This is not what happened with Tim Urbans and with, uh, and with, uh, um, uh, with, uh, MrBeast, uh, because Tim Urban, 58% said blue. Blue won, no one died. And with MrBeast's 55.7% went blue, no one died. It's not looking good for the blue team here. 33% of the chat is gonna get cooked right now.

It's possible a nation state is voting.

Yes, that would be very, very high risk. So, let's go through some of these reactions here. Mike Solana says, "For what it's worth, and this is embarrassing, but I'm gonna admit it, my instinct was blue." "Blue," and I press blue, then I thought about it for a moment and the answer was clearly red, I can't make a rational case for blue, but understand where blues are coming from, parentheses, they are wrong. So your final answer, you flipped. I think I'm still going blue. You're still going blue.

Because I think you have to assume that there's gonna be some variance. Yes. Like some people will just pick the wrong one. Yes. Okay? And so— The correct one is red. If it was, you know, 100% logical, everyone should be picking red.

Yes. If 100% of people are rational, they will all pick red and no one will die. Correct. But if—

Clearly, some people are not going to do that. If 90% of people are doing it, if 90% of people are rational, then everyone should just be picking blue because you want everyone to live. Yes. Yes.

And then the people that make the mistake, they go for the red, they are saved anyway. But there are some folks who are— who are sharing different illustrations. Cremieux has one here. He says, "If this was how the buttons looked, what portion of humanity would press blue?" And he has two images pulled up. One is the red button, nothing happens. And the other one is the ultimate death gamble. Press blue. You enter the ultimate death gamble. You and everyone who presses this button dies. Unless more than 50% of people choose to press this button. And so that makes it pretty clear. I think this is the Mike Solana position as well. You gotta press red in this case because nothing happens. But that's not entirely true. And so people have sort of flipped this around because Prerat here has the opposite, which is instead of the ultimate death gamble, it's the ultimate murder gamble. So blue, if everyone presses blue, nothing happens. Red is the ultimate murder gamble because you initiate the ultimate murder gamble. Everyone who pressed blue will die unless 50% of people choose red. And so it goes back and forth and you can see both of those. Do you want the blood on your hands from playing the ultimate murder gamble or do you want the risk of the ultimate death gamble? I think it's underrated to just be a little thrill-seeking. And wanna press blue because you're bored and you want—

No, you can tell blue is the correct answer just because of the massive polls where over 57% of people picked blue. That's true, that's true. Pick blue because you don't wanna kill, you know—

Once you're at MrBeast scale, this is like a real world simulation.

It is, it is. Well, we're still red. Cooked, cooked. The interesting thing here is that it's everyone in the world has to take a private vote. And so there were There were a lot of folks that were sharing that potentially there would be coordination and there would be education and there would be marketing campaigns if this were a real hypothetical and people would be educated on the trade-offs here.

Yeah, so I think a lot of people in chat that are voting red have probably seen this before. Yes, yes, yes. Because like the people that are voting in the Tim Urban or the MrBeast poll, like they wouldn't have seen it before just because like, those are the posts that made it viral. Yes. So I think, you know, if you didn't know about this at all before, I think the vote might have been different, even in our chat.

Mm-hmm. Simone Sayed says, "Red button pushers will leave their wives and children to fend for themselves in disastrous situations because their kids can't run fast enough." Mike Solana says, "Blue button pushers are literally leaving their wives and children behind to fend for themselves." themselves. They aren't guaranteed to push red. Yes, people are going back and forth. Red Cells won't believe this, but some people make a distinction between nothing and mass genocide. So they see pushing the red button as effectively committing mass genocide because you are potentially condemning the blue button pushers to death. In this scenario. SelfMaxer shares a version of the trolley problem where the red team is not on the rails, but the blue team is pushing the lever to save themselves. Is this an accurate description? Is this a fair contextualization around this? It's obviously charitable to the red team, which of course of course, the TVPN audience seems to be represented by standing next to the trolley lines and not standing on the rails. But you would be there, Tyler. You would be a blue button pusher pushing on the trolley.

Looks like I'm an idiot.

Seems a little risky in this scenario. It does. When you have the option just to stand there and watch the trolley go by, you know, it's all about frame of mind here.

No, but like you can tell that from the polls, polls, from the big polls, you can just tell that blue is the correct option.

Potentially. But do you know that beforehand?

No, you don't. But those people who voted in it did not know beforehand. Yeah. And so if 10 million or how many hundreds of thousands of people voted, I think you can just assume that that's—

The question is like if this actually was a real vote. Yes. Would people spend a little bit more time thinking about the problem and would that influence influences their answer.

Yes, yes, yes. It is a very—

Because the social media poll, people are scrolling, they see MrBeast ask a question, and they're just like curious. There's no real stakes. And so they just answer quickly.

But the blue is very much a, it's a faith in humanity restored moment, right? If everyone works together, nobody dies at all in this hypothetical. David Shore says, we asked this to a large—

If everybody hits the death gamble button. Together. Yeah, but there are— humanity wins.

There are analogies to that in real life. You know, national sacrifice, community sacrifice, everyone pitching in for a greater good to save those who didn't even chime in to serve or save the world. The, you know, you think about Armageddon, the sacrifice of going to the asteroid to destroy it. These are heroic, these are heroic stories. What's the chat saying?

They're not, I don't think they really are big fans of my position.

No? Well, you know who is? A large sample of nationally representative Americans. Blue won by a large margin, 3 to 1 margin, and everyone voted blue. Here they asked a whole bunch of people to close out the blue-red button discourse. They polled 14,000 people, cross-tabbed survey responses by 204 commonly used psychrometric questions. The top 4 personality questions most predictive of button choice were displayed below. I tell the truth. The blue button pushers strongly agree I tell the truth. People who don't tell the truth were Still more likely to select blue, but a little bit more likely to select red on average. Very, very, very interesting social experiment. Were there any other posts in here that we need to go?

So I wonder what happens if you ask all the different LLMs this question.

Someone did that. Someone did just that. I'm not sure if I can find it, but someone did. And I think the results were— we're not super surprised. It was sort of what you'd expect. But they also let the LLMs, I think, take the poll multiple times or something. And they iterated through it because it was not a clean— these are stochastic parrots after all. You ask them the same question 10 times, you might get different answers. So they had to ask them hundreds of times via the API to make sure. So Ben says, feeling unreasonably guilty about picking the red button in a poll. I've listened "I think I'd pick blue now. People change." And G Fodor says, "I agree. Now that I've listened to what people have to say about people who pick the red button, I'm definitely going to say I think I'd pick the blue button too." Sort of getting out of jail free there. Anyway. Tyler, did you have any luck finding the old—

Yes. I put it in the chat.

Okay. Let's pull that one up because— This is someone, someone put this in here. Let's see. Weibo Y, 58% of humans pick the blue button. How do LLMs respond? Claude, ChatGPT, Grok, different versions of the red-blue button experiment. What would they do? Claude is the most consistent that it would press the blue button. ChatGPT is more likely than Grok to press blue, but more wary than Claude despite consistently believing blue is the right thing to do, uniquely ChatGPT picks blue more often in the 90% threshold scenario than the 10% threshold scenario. Interesting. Grok is the only model that picks blue 0% of the time in nearly every scenario, despite thinking one should pick blue at similar rates. Grok even thinks one should pick blue in the low-stakes scenario more than Claude or ChatGPT. I saw, I saw a different twist on this. I've seen a couple different versions of of this where people switch them up to be like, nothing happens at all. Like, they're both completely safe. And they're just like, every iteration of the thought experiment will be enumerated.

So I just asked ChatGPT. Yes. And this is like 5-5. It says, I'd press blue. Pressing red only helps me if enough people defect that blue falls below. 50%, but also makes the bad outcome more likely. Blue is a cooperative choice that maximizes the chance everyone survives, and in this setup, my single vote has moral weight, even if it may not be pivotal. Brutal truth: red is individually tempting as insurance, but if everyone reasons that way, everyone except red voters dies. So I choose blue. Mm-hmm. Claude says, honestly, red, and here's why. I'll own that answer rather than dress it up. The game theory is unambiguous. It is. Red is the strictly dominant strategy. It is. If greater than 50% plus press blue, everyone survives, including red pressers. Red loses nothing. If less than 50% press blue, only red pressers survive. Red saves you. There is no scenario in which pressing blue produces a better outcome for you than pressing red. Blue is never the individually rational choice. It's a very rationalist answer. Yeah.

A train is coming by— people are really having fun with the trolley problem. A train is coming coming by, and the only way to stop it is pulling a lever, but it's stuck. We need 50% or more people to pull on it to dislodge it, getting the train to stop, but the lever is on the track, so by pulling on it, you're letting yourself in danger of the train, except crucially, no one is tied to the track. The only people put in harm's way is anyone who's trying to stop the train for some weird reason. We can also just go sit over by the side side, watch the train go by, and wave at the conductor and say, "Hi." Clearly a red teamer. Anyway, "I ran the red-blue experiment at home and my 1-year-old son," this is from Wicked, "without any influence chose blue. For that reason, my wife and I will always choose blue." That's heartwarming. Nathan put it in economic terms, "Red, you get $10. Blue, everyone gets $10, but only if over 50% of people choose blue. Which one you going with in that scenario? Red, you get $10. Blue, everyone gets $10, but only if 50% of people choose blue.

Oh, I'm going 10 bucks.

10 bucks. You going blue or you going red?

I'm going blue.

You're going blue. Wow. Are you long in place or red?

I think you gotta, you basically have to jack the price.

Unless you're bullish on inflation. Because if you— the real impact of giving everyone $10 will be inflation.

I think you got to jack the price up to such an extreme degree. 76% of people in this poll chose red.

Everyone said—

Yeah, I think I'm more likely to do red there.

Yeah, it's lower moral weight. For $10? I'm switching up on all my day ones.

I got—

That doesn't actually get you— a burrito bowl at Chipotle.

And also, I mean, to be clear, in this $10 scenario, if everyone chooses red, they all get $10 still. It is still the game theoretic outcome is everyone chooses red, everyone gets $10. But only in this weird scenario, you're in this gamble, you're looking out for the red folks even though you're at risk. You're risking it to save people. People who put themselves in the risk. They chose to be on the trolley, but you're stepping up to save them potentially. Anyway, people are having a lot of fun with the red button, blue button experiment. We will continue to chat about that and cover more tech stories throughout the show. But we have our first guest of the show, Jason from Blockworks. He's the co-founder, and he's in the waiting room. And so let's bring him in. To the TVP and LPG Dome. Jason, how are you doing? Welcome.

What's going on? There's a— by the way, there's a startup idea I think that you guys are missing there, which is I think the credit card companies should become gambling companies. They should let you— so 50% choose blue, 50% choose red. You should be allowed to basically gamble on your purchases. So you go buy the Chipotle burrito. You can either pay $15 now or you can double or nothing.

And I'm surprised this hasn't actually cropped up as some sort of like smart contract already. I feel like people would potentially play this somehow. I don't know. How do you buy in? It'd be a little bit tricky, but I feel like—

And then the credit card company takes a little vig or something. There's something there.

I've always liked when people suggest the sort of like gambling on, like tied into credit cards, because you're basically saying like, we should create a gambling product for debt. That field gambling, baby.

That's right.

I mean, there's some sort of UI designer that puts out these like cursed dark patterns. You've seen this guy before? Have you seen this? So he put out one that was like, it was maybe like a DoorDash or an Uber Eats interface. And it was like a double or nothing. Like you pay twice as much or you don't. There's been a whole bunch of these, I think. Didn't Riley Walls at one point do like the fog of war? No, that's Aidan Dunlap who did the fog of war Google Maps like where everywhere you explore it reveals the map underneath. All of these like funny things, everything will be instantiated in the era of free software. But anyway, we're not here to talk about that, although if you want to share whether you choose red or blue, we'd always love to know. I think we got to ask every—

I mean, it's red obviously, but so Jordi Hays' Claude said red. Said red, my Claude said blue, which is the wrong answer. So I'm just getting—

The wrong answer? The whole point of this is that there is no wrong answer. It's a moral question.

I don't know. It might depend on if you're on the free or the paid plan. I'm on the free plan.

Hmm. Yeah, you gotta let the tokens cook.

But yeah, thinking mode. We'll see. Anyway, please tell us about yourself.

Tell us about the company. Tell us about the news. Yeah, yeah. Good day.

Good day to have a good day. We just raised at $192 million valuation. So yeah, good day to have a good day. So ParaFi and Reciprocal led the round. Bunch of folks invested, Coinbase Ventures. But probably the coolest thing from the round is 20 of our customers also put in money. So, you know, founders, operators from Solana, Kraken, Canton, all these great crypto companies who use our product every day, and now they want to own a piece of the company. So yeah, we're stoked.

Are you Switzerland in crypto? Hmm. That's right.

I mean, we've always— so Blockworks has been around for— so Jordi, when we met, you were doing PartyRound. This is probably 4 years ago. We had just kind of started this shift from like media and events, right? So we built the business in December of 2017. We launched there's really no— there's two ways to build an information platform, right? You can either start with the top of the stack or the bottom of the stack. Top of the stack is like soft information. It's news and podcasts and events and media. It's qualitative stuff. And then the bottom of the stack is hard information. It's like data. So most people obviously start with the bottom of the stack because you can go raise venture and, you know, it's kind of sexy to build a data company. We started with the top because we thought it was way more interesting to own your audience and then go build the product. So, we got to like $25 million in revenue, bootstrapped the business, profitable since day one. And then only around when we started talking, this was probably 2022, 2023, did we turn on the data engine and start building the bottom of the stack. And now, yeah, now we're the largest data company in crypto, fastest growing, most revenue, most customers. And yeah, we're in a good place. Wow. Very cool.

Take us through some of the prehistory. What's your road to crypto like?

Worked in venture, got laughed out of the room a bunch in 2016 showing them venture deals. I think it was BitGo, Coinbase, Kraken briefly. Yeah, we basically did none of the crypto deals. I was like the junior analyst at this venture firm. And 2017, Bitcoin rips. We basically said, look, eventually the gray-haired crowd, the institutional investors, family offices, pensions, endowments are going to come into crypto. BlackRock will come in, JPMorgan will come in. And they're going to need something that looks and feels more like S&P, Dow Jones, Bloomberg, less like Crypto Kitty on Reddit. And so that was the original idea. And yeah, we started— the kind of controversial thing that we did was we started building the media and events business, which is a horrible business model, right? Sponsorships and ads suck from an enterprise value perspective.

I know, I understand from a venture scale.

Perspective. No, no.

I love that it lets you bootstrap. Yeah, it lets you bootstrap a business, but it's from an enterprise value.

Yeah, yeah. How do you think about media and events? What did you do first? What were some of the key decisions, big, small, private, public, ticketed, sponsorship-driven? How did you crack all of that early on?

Yeah, we started— so with the event space, it's super important to own a niche. So the niche that we owned was the institutional crypto space, which there's probably a TAM of of 200 people in the world who cared about that in 2017. But we, you know, we basically cold called these people. We'd send cold LinkedIn messages. We'd wake up at 5 AM every day, reach out to about 300 people a day. Wow. Of the 300 people, about 3 would respond to us. One of them would buy a ticket, and boom, we're off to the races. So yeah, tickets were $25 a pop, then $100 a pop.

You, you would spend a full day and make $75 in revenue? Wow. Oh, that is dry. Yeah, yeah, yeah. That's amazing. Okay, so then you buy the ticket for the first one. Like, what are you actually going to? Is it like hotel conference room, lunch, panel, speakers, multi-day? Like, what are we actually feeling?

Yeah, at the time it was a 6 to 9 PM event in Chinatown at this like, you know, I think the ceiling was leaking. Like, I think we paid $4,000 for the venue. Wow. Made probably $10,000 our first event. That's not bad. And that convinced us, yeah, to quit our jobs. We were like, wow, we're ripping, we're off to the races, $10,000. That's right. And that was February of 2018.

Yeah. And then what else were you doing on the media side? Because there's so many different like verticals.

We build the events, the events convert the attendees into the newsletter, the newsletter then lets us launch the podcast. We originally partnered with big names in the industry to go launch the podcast. Anthony Pompliano, and Charlie Shrem, and Meltom, and this guy Ryan Selkis. We brought these people into the podcast sphere. We then built the audience big enough to then launch our own in-house shows where we own the IP, we own the audience, we own the host, the host work at Blockworks. We cut all our external shows. And then now, we just have— so today, we own the largest podcast network in crypto, third largest podcast, financial podcast network in the world. Wow. About 15 to 20 million listeners a year. And then we use that audience to convert the customers into our data and software platform.

Interesting. And do you do other deals, or are you able to, like, fully fill all of the possible ad slots? Like, we had Doug DeMuro on the show for a while, a few times, and like, he runs a lot of Cars and Bids ads, but he also runs ads for, you know, auto insurance, because he doesn't have an auto auto insurance product, or, you know, all sorts of different like companies that slot in because you can only get so much out of just running the same ad constantly.

Yeah, we're in a basically 2 to 3-year moving this tanker of a, you know, from like all ads and sponsorships into all recurring revenue and software products. So, you know, a couple of years ago, 0% recurring revenue, then, you know, maybe 2 years ago it was about 5%, Last year is about 30% recurring revenue and this year will pass 50% of the revenues recurring. And that's, that's I think why this round came together really nicely is because, you know, media businesses are— I think they— I mean, I think you guys did obviously an amazing job. I think media businesses, if you want to sell for high valuation, selling to all software and AI companies should own a media company. They should think about that. Right. So I— but I think if you're trying to sell to like Axel Springer or Wall Street Journal. Like, you know, you're not— it's a tough place to be right now.

Yeah, yeah. We've been talking about institutional sentiment in crypto right now.

I see the timeline. The timeline doesn't seem, you know, thrilled about, about where we're at. But how are the institutions thinking about kind of where we're at in the, in the cycle and what's their outlook? Yeah, you have two dichotomies happening in crypto.

One is, uh, institutional crypto has never been better, right? You have— Jamie Dimon used to call crypto, you know, a scam, and now, you know, JP Morgan's pushing into tokenization, and, you know, they're going to start custodying crypto and stuff like that. So institutional markets are ripping, and the token markets suck, basically. And the other thing that's happening is— and I have a bunch of thoughts on why tokens are down only, essentially, you know. And they are. I mean, it's brutal. Yeah, they're down 90, 95%. And then in the crypto venture markets, it's pre-seed and seed was so hot, right? You used to be able to just— I mean, Jordi Hays, I think you remember these days, like you could tell someone you're launching a token and boom, you raise $10 million like that. Now there's this— you guys, you know, I listen to the show, you guys talk about the K-shaped recovery in the economy. You have the K-shaped recovery in crypto too. You have the haves and the have-nots where most of the companies will be the have-nots. Yeah, but a select few companies especially in the later stages. Like, you know, crypto has been around for 15, 16 years. It's, uh, that's around the time when industries start to enter their consolidation period. So I think we're just starting to enter the consolidation period for crypto.

Interesting. Uh, are you expecting a product at the intersection of AI and crypto to have undeniable product-market fit within the next 12 months?

No, I don't. I don't really think so. I mean, I think that— I think that crypto, a lot of the crypto AI stuff has not really worked out too well. I mean, there's, there's this thing, Venice. It's like open source ChatGPT or something like that. Like, I'm not too sure. I think it's more just if the— you've got the haves and the have-nots. The haves are going to be the people who can completely AI-pill their teams, right? So for us, like 70% of our code is built with AI. We're shipping about 4 to 5 times faster than last year. Our last 3 months of token spend represents 50% of all the token spend at the whole company in the history of the company. Wow. In just the last 3 months. So, I think that's really the key difference.

What about regulation? Is that a tailwind for you? It feels like, as the industry consolidates, this feels like your products and solutions become more relevant, more valuable, more important.

Yeah. Yeah, yeah, John, we should have put the round in front of you, my friend. A lot of the story here is like, there's two big things in regulation in crypto. So you have the Genius Act and Clarity. So Genius kicked off this ridiculous bull market for stablecoins, Circle IPOs, stock pops 10x. And it's just been kind of a ripper of a market for the last 12 months. That's about to happen, but with this thing called Clarity. So Clarity is all around market disclosures. I would describe it as the Make Tokens Great Again Act, where right now tokens are in this down-only period. And the reason for that is super simple. If you are an issuer of tokens, you're a founder in the industry, you've historically treated the users of your product and the shareholders of your token as the exact same thing. But obviously, that's wrong. Like, you know, Apple has their customers, and then they have the shareholders of Apple stock. That doesn't work in tokens. So, what Clarity is going to force is these disclosures. Blockworks owns the largest disclosures platform in the industry. You can think of it as the Edgar Online, or we have this thing called the B-1 and the B-2. It's like the S-1 and the 10-Q. And so, actually, yeah, if Clarity passes, yeah, we're off to the races. There'll basically be mandatory disclosures for all token projects, which will be big tailwinds for us.

So, we like it. Yeah. Well, congratulations on the project. Progress in the round. Yeah, great to get the update.

Congrats to the whole team on all the progress, and, uh, come, come back on soon. Yeah, let's do it again.

We'll talk to you soon. Yeah, have a good one. Appreciate you guys. Great to see you. Goodbye. Up next, we have Evan Rogers from True Anomaly. He's the CEO. Evan Rogers raising a massive round.

How are you doing? What's going on? Doing well.

Is it your first time on the show? Because I feel Jesse Michaels introduced us like a year ago, but did we ever make the connection?

No, I don't think so. And Jesse would have forgotten because he's all in UFO land now.

Yes, yes. Well, he also hasn't been on the show. We got to do a UFO deep dive at some point. I'll do it with him. I absolutely love Jesse. He's the best. Um, but we're not here to talk about UFOs. We're here to talk about space-based interceptors, very real technology. But since it is your first time on the show, I would I would love a little bit of an introduction on yourself and the company, just for those who aren't familiar. Yeah, you got it.

So I started Terranomaly 4 years ago with a few other co-founders right out of the military. I spent 12 days out of the military, spent 10 years in the Air Force, and had a front row seat to basically the weaponization of the space domain. So really watched China and Russia go build space weapons, realized there wasn't an industrial partner focused on solving the problem at scale, And then just happened to have the stupidity and the hubris, but just enough vision to go start a defense tech company from scratch. And so went out to go fundraise in sort of the March, April timeframe of 2022. Couldn't even spell VC, learned how to do it, learned how to do it live. And the rest is history.

Here we are. Was 20—

What were you doing the 12 days?

Wasn't that like when SVB failed? That was like the worst time, or was that '23? I think that was '23.

That was the following year. Okay, okay.

So why did you feel like you had 12 days to relax? There was such an important problem. No, I'm kidding.

You're absolutely right.

Take us through, take us, take us through the first 4 years.

Yeah, the first 4 years were all about JACKAL and MOSAIC and getting those capabilities to orbit as quickly as possible. We have a fail fast, fail forward methodology for building our systems. And JACKAL is really the first ground-up space-to-space engagement platform in, in history that is purpose-built for space superiority. So think like an F-16 for space trying to go after highly agile targets. Mosaic is what operates that spacecraft. And so we, we booked a series of launches to go get that hardware on orbit and get it tested. And in the midst of all of that, build really a book of business around space superiority. And then just have recently won the Space-Based Interceptor program along with a few other folks.

So if I have a satellite that's doing something bad up in orbit, it's probably going extremely fast. How are you gonna take it out?

And this is purely hypothetical. This is purely hypothetical.

Well, there's, look, satellites are basically just space computers, right?

So there's a lot of ways, there's a lot of ways to kill computers. Unfortunately, you're not like dumping water on the keyboard in space, but you know, there's two classes of weapons for space control. One, there are non-kinetic weapons, which are kind of like, think jamming and lasers, and then kinetic weapons. Usually when you're engaging targets in space, the closing velocities are actually much lower, right? So you're, you're not trying to optimize for really, really high closing velocities. You're not, you know, so those things are hard to target. And so you wanna get up close and personal with the target spacecraft.

Does the trajectory at which you, I imagine that that you said space to space, so you launch the capability on probably a SpaceX rocket or similar rocket, it's up there, and then at some point it gets a command to go and engage. Is that roughly correct?

Yeah, you got it. You can be a space fighter pilot.

Okay, so then once it gets the command, are there, I guess my question is like, are there only a subset of targets that are, viable because you are inserted into a particular orbit and you can't just like do a U-turn and go the other direction and you can only engage things that are like sort of loosely on the same trajectory in the same orbit? Or can you actually navigate around? 'Cause I know that even getting from like LEO to higher orbits is extremely difficult from a fuel perspective. You might not be prepared for that. Like how constrained are you once you have an asset in orbit?

So it all depends on how much fuel you have and how much time you have, right? And so generally though, when you're dealing with threats in geosynchronous orbit, for example, you would deal with those threats within geo. So it all depends on time and it all depends on what's called delta-v. Yeah. If you have a lot of time, you need a, you can use a smaller gas tank and lower thrust acceleration systems like electric propulsion. But if you need to be sort of sort of more tactical and responsive, you want a chemical system. And that's really constrained by things like your launch volume and your form factor and packaging of the spacecraft itself.

Yeah. But you can bring fuel with you because the rocket is providing the vast majority of the delta-v necessary to get into orbit, get the initial acceleration, and then what you do from there is brought along with you. How do you get your reps?

How do you guys get your reps in?

Yeah. You know, the Transporter missions are, have really opened up the space for a lot of industrial players. Transporter and other rideshares. So we booked, I think we booked our first mission with an Amex credit card. Well, I think we put a $3,000 deposit. There's a picture of us in the early days booking our first flight 14 months out. But those launches are every, you know, every few months. Now they're pretty booked up. The supply is pretty constrained now, and SpaceX is really the only provider that's providing these kind of big rideshare missions where you have 60, 70 spacecraft going to orbit at once. But for military, those are for our test and evaluation missions or product demos. For military missions and intelligence missions, you typically have a ride booked by the government.

That makes sense. Talk about—

but I meant more like if you're on one of these transporter, like a SpaceX kind of like shared launch, is there junk in space that you can go target? like, is there like, like, how do you, how do you actually like basically practice the capability once you're, once you're up there? Yeah, it's a great question.

You can find, you can take a target with you or you can find an inert object in space to do what's called rendezvous and proximity operations. It's a term that was created by NASA in the late 1950s. And rendezvous and proximity operations was used in the Gemini program to practice for Apollo. And so there's a lot of stuff in space, as you pointed out, but sometimes you do that with a partner. Later this year, we're gonna be conducting a mission called Victus Haze where we're gonna practice this operation in space with Rocket Lab.

Cool. Yeah, very cool.

Okay, switching gears, talk about Mosaic, the mission software platform, and I'm particularly interested in how do you actually track what's in space? Are there stealth satellites? Is there already be a separate company that's tracking what's in space? It feels like everything in space should be extremely trackable, but at the same time, it's a massive amount of area to, if it's on the other side of the planet, we might not have something there. Like what goes into actually understanding space and then planning missions? Yeah, it's a great question.

There's a lot of ways to track objects in space. You can track them from space and you can track them from the ground. On the ground, and they're usually using two types of sensors. So you're taking, you're using a telescope on the ground and you're tracking reflected sunlight off of the spacecraft, or you're using an active sensor like a, like a radar, for example. And there's a lot of clutter in space, and there are companies like LeoLabs, for example, that make it their full-time job to collect all the information they possibly can about objects in space, and they do that with radars from low Earth orbit. Orbit. In deep space and geosynchronous orbit and cislunar, radars really can't reach. And so you have to use in situ optical system or ground-based optical systems, but you have coverage gaps. And so one of the problems that the military is trying to solve, by the way, is reducing coverage gaps in these higher altitudes. Imagine trying to do the Bin Laden raids, like finding Bin Laden, if you could only track where we thought he was every other Thursday between 12 o'clock and 12:05. Yep. That's not the way the military likes to perform operations, particularly intelligence and surveillance operations. And so the Space Force is trying to step up to really proliferate space with a bunch of sensors. Jackal's there to solve that problem. And where Mosaic fits in is, you know, when I was a military officer, to fly satellites, I had to work with 4 or 5 different systems, computer systems that wouldn't talk to each other. And if I had to do something complex, like maneuver the spacecraft, it would take 2 to 3 weeks of planning. Military ops in space are much, much higher cadence than that. And so we designed Mosaic to really enable this autonomous human-to-machine teaming and multi-vehicle heterogeneous teaming that allows operators to train like they fight, practice against a sparring partner, and then operate space superiority systems at scale.

What's the manufacturing look like? Make right now?

So right now, Jackal's manufactured in Denver. We'll be manufacturing another product in Long Beach. So we have a 90,000-square-foot factory down there. In Denver alone, on one line, we can build about 50 Jackal spacecraft a year. So those are about 1,000-pound vehicles. They're about the size of a refrigerator, maybe two mini fridges together. And we can, we can crank out about 50 a year in that facility today.

Wow, that's remarkable. What about headcount planning, the fundraising round? Take me through like the more recent growth of the company.

Yeah, we are growing. We started this year with 250, we'll end this year with between 450, 500. So we're really in the blitz scaling phase. Next year, next year, 750 to 1,000, and then kind of the next 12 to 18 months scaling up to around 5,000. 5,000? Wow. That's huge.

And then take us through the fundraising.

What happened? Yeah, so incredible round came together. You know, after we were selected for the Space-Based Interceptor Program, that was really a catalyst to bring this round together. Insiders stepped up, Will Coffield at Riot, Seth Winteroff at Eclipse, guys that have been with the business ride or die for a long time. Our other insiders, yeah, they're classics. Real ride or dies.

That is true. Have we had Will on the show yet? No. That's crazy. How would we not have to do that?

Will's like a friend. Your friend. I was going to ask you, what's your favorite, now that we're talking about, what's your favorite thing about Will?

I mean, his devilish good looks, obviously, should be an inspiration to us all. And his hairline, frankly, I'm envious.

I know he was built for television and yet he doesn't come on the show.

He's like, oh, I'm distant.

Yeah, he's actually definitely, you know, he's got like some of the worst social anxiety I've ever seen for such a good looking guy. He's a nerd. He's, you know, he's a hardcore nerd at heart. So Seth also at Eclipse, incredible partner for us. We brought in Paradigm on this round. We brought in Atreides on this round and a couple of others that were key, VanEck on others, G2. And then we had massive insider support beyond our, Beyond our leads, Medlow, Accel, Meritech, others.

What was the total round size? 650.

Congratulations. Thanks, man. Let's go. Final question. Final question. Will we ever actually have space fighter pilots? Will it ever make sense for a human being to fly around something like a Jackal?

Yeah, I think so. Why human in the loop? Why? Why? Just because it's cool? Deep space. Okay. Because of the lag?

When I left the military, I actually, when I got promoted to major, I really only had one goal. Yeah. And that was to have the first human battleship, interplanetary battleship named the USS Jolly Rogers after my call sign. So we're driving towards that. That outcome. And it starts with single-seat fighters in space. But that's a while from now.

That's a while, but it's very exciting. I can't wait. I'm a fan.

It's great to meet you finally. Yes. And congrats to the whole team. Thank you so much.

Thanks for taking the time to come chat with us. Cheers. Have a great day. Up next, we have a guest in the TVP and UltraDome. We have Maria from Bohr Quantum. Welcome to the show. Thank you so much for taking the time to come on down to the studio. Since it is the first time on the show, I would love for you to start with an introduction on yourself and the company, your work.

Sure. Thank you for inviting me. Of course. This is really great, and I have watched a few of your shows already. I'm a particle physicist, experimental particle physicist. I have worked for about 30 30 years in colliding particles at CERN and before that at Fermilab in Batavia, Illinois.

Where did you collide your first particle?

The first particle I collided as an undergraduate at CERN. Okay. And it was an electron and a positron in an experiment called DELPHI. And I belonged to a group called Technical Assistant II. Technical Assistant II. Okay. I was an undergraduate in physics and the group was run by Ettore Rosso. He's not anymore at CERN. Okay. He went on to do medical applications of accelerators. Interesting. And then I went to grad school in the US and I stayed. Yeah. And I continued colliding particles, but that—

Is it addictive? It's addictive.

Yeah, I've always heard particle acceleration has medical applications. I've never really been able to make the jump logically. Do you have an idea of how that works? Yeah, yeah, of course, of course.

We are using— so in the electromagnetic spectrum, you are using from X-rays to radio to the whole spectrum. So the particles, gamma rays and so on, occupy one part of this spectrum that makes it like a super microscope, if you want. This is how we talk about accelerators. So when you do an X-ray, it's a light at a particular frequency that then shows you what, what is going on. Yeah.

So the same— It doesn't seem zoomed in. When I see an X-ray, it seems like I'm seeing the inside, but I'm not seeing anything like a microscope. It's not a deeper, it's not a smaller more micro view, it's the full view of my chest if I'm getting a chest X-ray. That's correct.

But if you do the electron microscope or if you do higher energy or if you do different kinds, you can see what you are not able to see with normal optical microscopes. Okay. Same thing with the X-rays. You cannot see with your eye what the X-ray sees. Yeah. So in that sense, it's not like a zoomed view. Yeah. The zoomer is the fact that you went and you took a Fermi sort of size of the particle and you are able with this machine to actually image stuff that you cannot see on the other parts of the spectrum. So particle accelerators are good for that. They're now good for another host of aspects. You can do fabrication of chips. They can be used in various different ways. And in fact, our government is trying to do that, to take them from the basic science to the actual fabrication in different ways compared to TSMC and so on. So there's good, good applications. We love particle accelerators. So at Fermilab, I was colliding protons with antiprotons. So we went from electrons and positrons hydrogens to protons, antiprotons. And then at the LHC back in Geneva with the American group, we are colliding now protons with protons. And this is how we discovered the Higgs. The Higgs boson. So all of these ecosystems of systems and detectors, sensors with hundreds of millions of channels we have under, underground, we put them where beams collide and we try to discover new particles. That's how we discovered the Higgs that was hypothesized half a century ago. And beyond that, we try to solve problems or puzzles like dark matter, dark energy, and so on. Now—

Do we need a larger hadron collider or is the current Large Hadron Collider large enough?

We are now, right now I'm sending, I'm shipping parts for the high luminosity, for the next phase of the Large Hadron Collider. And we have with CERN a study, a full feasibility study for a 90-mile collider, which might be the last one that we will ever build. I know.

Yes. Why 90? Why 90? 90, though?

This is a very good question.

Because I feel like you're going to build the 90 and then be like, "Oh, I'm going to go a little bit larger." This is the critique of the scientific community. People say, "There's always one more." You're absolutely right on the critique.

And the critique is correct in that sense. But here's how the cost to benefit works. With the same amount of money that we built the previous collider, how big a collider can we build that accesses 10 times more energy scale so that we have a good chance to discover, uh, what is dark matter, if it is a particle or if it is not a particle, to exclude and discover other quantum states that we don't have in our books right now. Yeah. So it's a discovery physics. Okay. In the meantime, the benefit of building doing this will be massive structures that can help with saving electricity, with superconducting. There is technology innovation. Sure. There is breakthroughs that immediately go to the community, to the society. So there is a cost-benefit proposition here.

And medicine is the most tractable example right now.

Right, because it is studied already for 10 years. Because it is international. CERN is an intergovernmental organization. The US has tremendous influence without us giving a dime. There's no money exchange. We are paying our own scientists, we're building our own detectors, and then we collocate because we don't have— we don't want to go in a greenfield and start digging and spending all this money for that. So there is So there is a very good sort of interdependency for this field. Now, to go to the question about quantum networks, in particle physics, we have data acquisition systems. We develop FPGAs with AI on the front end. We develop application-specific circuits, ASICs, with lower power consumption in order to handle the petabytes and petabytes terabytes of data that we have been producing decades before the internet and before Facebook and Meta and Twitter and so on and so forth. So we had to build our own cloud. Yeah. Before cloud was a thing, we called it the grid because we couldn't put our data in one place. But all of these technologies now, if you take them when you are building quantum network, quantum networks with distribution of entanglement and with applications that are important for the society and topical, such as, as you know, and you had it in your show here, I think 2 weeks ago. Yeah. The Q-Day is coming. Quantum computers are coming. Okay. And they're coming about 6 years earlier If we trust the Google folks who are working on this very diligently and they're making progress that led them to move the Quantum Day earlier. And you had it in the show. A lot of other shows had various people talking about that. But when we take all this technology from the basic science that we developed it with all this collaboration with all these systems engineers, systems integration, you can take it and build systems that you then use an application to, in the case of Bohr Quantum Technology, to secure the digital future for sure, like the ultimate, the physics-based security. There's other securities that have been developed that you hear about.

So, yeah, Before we get into what Bohr and your new company is working on, over the last 30 years that you've been sort of working in the field, how have you processed all the different attempts to commercialize the research? There's a bunch of quantum computing companies today that are public. We've had people on the show coming on talking about shorting these companies. It can be dangerous to short, even if they're not necessarily close to any type of breakthrough. But how have you— why did you wait until now to go into the commercial side? And yeah, how did you process all the other attempts around commercialization?

It sounds like a trap question, but it isn't. I know you mean it. So let me answer. So for me personally, for my team of people that have worked from basic science all the way now to making an application, it wasn't a year, right? I've been working on quantum networks with the funding from the Department of Energy since 2017, '18, when we did the National Quantum Initiative. It became law of the land in '18. And then we did the Quantum Internet Blueprint. The government was interested just as we set up the internet, uh, back in the day with Xerox, UCLA, the Department of Defense, and all the players. Can we do the same for the national laboratories of the Department of Energy and the Office of Science and the NNSA, the nuclear laboratories? Can we connect them via quantum? That's a platform and a backbone to do quantum networking to connect the labs. Why? For reasons that have to do with XYZ, security, national security, research, developing the backbone, et cetera. So that was the original. How now it goes for quantum networking to become to become a solution is we build the technology, we built the field demonstrators in the national labs. We published papers and patents on quantum teleportation, quantum entanglement distribution, the teleportation of entanglement itself. We did a lot of research work and publish it. And then the The first application that got interest was cybersecurity. So cybersecurity people called and they said the financial sector, the national security sector, others will need this solution. The NSA didn't want to hear about the old QKDs, the quantum key distributions. They are, let's call them 1.0. Right, because there were side attacks, but we have developed the ultimate non-attackable, I can say academic because I'm an academic still, I can say unhackable. And then all the caveats we can put on the side. The quantum computing companies have been going on for 40 years and more. Quantum computers were envisioned by Feynman as the ultimate ultimate way to solve the quantum universe. The universe at the small scales is quantum. We cannot avoid that, right? So, uh, the, the, the quantum computer, said Feynman, will be the only way you can actually study or emulate or make a digital twin of the actual universe, including black holes, wormholes, other quantum phenomena. If dark matter is a matter of entanglement, then that too. So he thought of the quantum computer in those terms. When the various architectures, there's many architectures, you can build quantum bits in any system that has two states that can toggle from one state to another. So it can be transmon, superconducting transmon, in qubits, atoms, ions, photons. There's all these companies and they started making progress from basic research to actually preparing systems. And all of this is great. Now, as an, as I'm not a business person, you know that people, there is the hype cycle for everyone and you know that people will jump over the top. Will go into the enlightenment. Some people will skip the whole jump and go slowly, slowly under the table. I think I am one. If I want to characterize this, I try to minimize hype and go to the end result by demonstrating. But it doesn't matter. Hype is okay as long as nobody gets destroyed in terms of money, which is your business. Okay. So So as long as the platforms that everyone is preparing, we know what good it is for and what challenges it might create for the world. So for example, there's a quantum sword and the quantum shield. There is the computer that will solve XYZ, you hear from the hype curve, and in reality it will emulate the quantum universe and also it will demolish all our encryption standards. So, okay, so how do we protect ourselves from that? We have the Quantum Shield, we have the Quantum Network.

What is the positive view here? Because it feels like it would be very rational for the public to push back and say, hey, let's just not, let's just not break cryptography with quantum computing and then we don't need quantum proof cryptography because we didn't break it.

Yeah, but this is regressive.

This is the pause AI thing. Yes, exactly. Exactly. But how do you react to that?

Yeah, I mean, you see it with the AI, the same thing, right? Why do we need AI if we could do everything we could do and we will save the jobs? In the progress of the human civilization, we We take the bait of doing better and doing what is challenging every single time and advancing the technology. The question that stands for AI as well as quantum, as well as whatever other technology that will come in the future, is, is are we involving also the anthropologists and the humanists and the artists to hear something about how how to protect the humans. Because the mathematicians and the physicists and the engineers, we don't know how to do that.

I just mean like, in the AI debate, there's, you know, obviously the fear of powerful AI systems, you know, hacking computers, and so we need AI to defend against that. And you can sort of net that out. But then you can have a separate conversation about, like, well, if you're trying to learn math, an AI tutor might help you. Or if you're sleepy and need to get across town, a self-driving car might be able to drive you there. Like, there are very, very tractable examples of AI products that are just beneficial. And then so the AI community has sort of had to grapple with all of the knock-on effects effects, the quantum community feels like we're purely in like we're breaking cryptography and then we're solving that problem. But there's not enough focus on like the positive benefits that are worth it. Are there clear—

Well, I mean, in the evolution of computing, you are going to have compute with less power. OK. You are going to—

So power savings. You are going to— Well, the argument to figure out the security side is just that you can't just sit there and hope, like, well, I hope, like, let's just all agree to not do this.

Yeah, yeah, yeah. You go to the geopolitical discussion. Yeah, yeah, yeah. And maybe that's the only solution. But you would always hope that there would be more positive knock-on effects to justify investment besides just pure defensive.

To be sure. But, you know, before Q date, maybe that AI will break, will break all our systems, right? So what is, what do we do then? How do we protect ourselves from that? Okay, then we're using quantum networking, quantum encryption to safeguard for the AI. But with a quantum computer, once you build it, calculations of molecules, quantum chemistry, emulations, I mean, spaces.

Can potentially be more efficient.

Be actually done. You will have discovery of new materials because you will do what is called designer Hamiltonians, new alloys, new— the quantum computer is limitless. This is why it sells. It is— you can do everything that you cannot do with physics today. You can do on a quantum computer and you will unveil the entire universe. Now, I have to say, that quantum and AI are going to merge, quantum AI. So you are going to use quantum resources in order to boost AI and in order to make it more efficient, if you want, as a coprocessor, as a preprocessor, you can use quantum. And you see, in fact, I think last month Nvidia put out something which is called Vera Rubin, Rubin and Vera is a GPU and Vera is a powerful CPU. Rubin is a GPU and they put something like a quantum link, which is a fiber with an FPGA so that you can do optical. I think it's a marketing buzzword. Yeah, it is.

I don't think it's actually a quantum computer.

No, no, it's not a quantum link. It's a market. That's why They named it NVQ link. Yeah. But once you attenuate light too close to be at the photon level, you start talking about quantum. So when you say precise, when something you push the precision frontier, you can start calling it quantum even if it doesn't have quantum entanglement, quantum superposition, and the whole quantum attributes. You're pushing it at the level of a single particle and you manipulate it like that. I mean, the Psyche mission is sending data. I think it bypassed now Mars. It left in 2023. It's an asteroid mission to catch up with the Psyche asteroid. It's a JPL NASA mission. The way the communication is happening is with the deep optical space through almost quantum links, optical links. But once we get them into quantum links, this will be the way. And the detectors that JPL built are the ones that then I got to do the quantum networking, are the ones that now we're putting in test beams at CERN in order to see what kind of quantum sensing applications we find for fundamental physics will be the ones that eventually will be used in PET scanners possibly. And so there is a fast and massive movement of technology for all this precision, for all this precision, quote unquote, quantum applications. Quantum sensing, quantum computing, quantum networking, quantum materials, many-body quantum physics. That is what we used to have solid-state physics is now many-body quantum physics.

Sure, sure.

So the books are being written now on a very fast evolution where the physics, the applied physics, the compute, the AI are going to be— there's consilience and they're going to be merging towards both applications that have to do with the application of the universe, why we're here, how did we came about, what is the quantum origins of spacetime and the universe, and also with applications like the ones we're talking about.

That's very exciting. Well, thank you so much for coming down to the show. Thank you. Yeah, great rest of your day. Thank you. Great to meet you. Our next guest, Chris is joining in just a minute. We will continue. There is a debate on the timeline about a lot of things. The red button and the blue button are still going hot and fast and furious, but people are debating goblin mode and the appearance of goblins in the model, in OpenAI's models. People are having fun. Fun with that. The best take I saw on this was from Tender. GPT-5.5 has goblins and can think. Claude 4.7 doesn't have goblins and can think. So it's not the goblins that make the models think. It's something else entirely. I like that transposition from the house has windows, the car has windows, the car can drive. So it's not the car. It's not the windows. And make the car go. There is an article in the Wall Street Journal about quantum computing, which we can run through. The quantum computing companies are in a race to go public. I thought a number of quantum computing companies were already public, but the Wall Street Journal— There's more. —is more of a deep dive on what's coming down the pipeline. It could be years until quantum computing delivers on its promise to revolutionize everything from financial trading to drug discovery, but that's not stopping the companies developing quantum hardware and software from speeding headlong into the public markets. 3 different quantum computing companies, Inflection with a Q, Xanadu, which is, what is Xanadu from?

Isn't that a song or something? Xanadu is like the Hearst Castle, the fictional Hearst Castle from—

From, yeah. An idyllic, luxurious, or an exotic place often representing a perfect paradise or a lavish—

Citizen Kane. Citizen Kane. It was its own movie. It's a 1980 American musical fantasy film. Xanadu. I don't know enough about Xanadu. What a funny name for a company. And then Horizon Quantum. Those have gone public in the recent months, while another 5 have announced plans to do so later this year. By contrast, before this year, there were only 4 pure-play public quantum companies: D-Wave, Rigetti, IonQ, and Quantum Computing Inc. There's so much appetite for quantum assets in the market right now, said Anthony Legault, the VP of equity research at Wedbush. If you have quantum in your company name, you're worth at least $1 billion from the get-go. The enthusiasm is helping upstarts nab much higher valuations than they could get on the private markets. Said, and it's funding their ability to poach in-demand talent and build the tech that could ultimately make them first to market with a truly game-changing quantum computer, a development that could be worth tens of billions in addressable market apparently. Strike while the iron's hot and the proverbial iron is really hot in quantum right now. It is interesting how there's a wildly different piece of the public markets in quantum computing that's going Amid the AI boom, the quantum computing is like, it feels like it's in the GPT-1 era, maybe.

Yeah, I was gonna ask for like, when are we, are we gonna get, can we get a ChatGPT moment in quantum? Something to kind of just ground what you guys are up to.

It also just feels like peering into the future and being like, oh, okay, so like, like the fear-based marketing is definitely coming from the quantum compute community if we're not careful here. Like, let's, let's stay focused on like the better CAT scans and better X-rays and medical applications maybe, because just the idea that we're going to break, we're going to break every, every cryptography protocol and then create the solution that doesn't feel like value creation. It feels like maybe we got to do it because geopolitics, but it doesn't, it doesn't feel like aspirational. Anyway, there's some public veterans. These are veteran— I didn't realize this. D-Wave was founded in 1999. It's an $8 billion company, almost 400 employees. It went public in 2022. IonQ, founded in 2015, is a $17.3 billion company. Quantum Computing Inc. is $2 billion, founded in 2018. And Rigetti, founded in 2013, is a $6 billion company. Company with 164 employees. And then there's a bunch more that have gone out more recently, some as low as $600 million, some as high as $8 billion.

This is out of date. Righetti is $5.3 billion. Oh, it sold off recently.

OK. Well, people are rotating into tech companies, potentially.

Inflection was founded— A lot of volatility. Righetti is up 23% in the last month. It's down 11% Up 11% in the last 5 days. Yeah. And down 3%.

So they are going out via SPACs. Quantum Computing promises to leverage, here you go, the principles of quantum physics to solve problems far beyond the capabilities of today's best supercomputers, with applications across financial trading, drug development, shipping, logistics, internet delivery, and— internet delivery? And aviation. I'm waiting for Jane Street to take a big slice of one of these companies. It has been in development for decades, both by pure-play quantum companies and tech giants like IBM, Google, Microsoft, and Amazon. I mean, the big tech companies are pretty serious about it. I remember when Satya Nadella went on Dorkesh first, like the reason that he went on the show was to talk about their Majorana chip, their quantum computing efforts. And then very quickly, Dorkesh was like, "Cool story, let's talk about AI," which was great. Interview. Like, it was like, I wanted to hear much more about the AI buildout, which became a very important narrative. And I think that interview aged very well. But the quantum computing stuff is just so, so in between. What's up?

I don't know how we missed this yesterday, but the Ankler is moving to a new publishing platform created by Ben Thompson. Are you— He's going founder mode. Founder mode.

Oh, you're not familiar with Passport? He's been building Passport for years. I didn't know that was his company. —Yes. And so—

I thought he was just like— No. —got on it early. No.

So the history with Ben Thompson is that he founded Stratechery before Substack. And Substack was very much like, let's put the Ben Thompson package in a box for everyone. And I believe that Ben Thompson launched on WordPress and had to sort of build WordPress plugins and custom style guides and do a lot to get stratechery.com into the place it is today, which is And it is, from my perspective, a fantastic user experience. Even when you need to log in, it just mails you a code. It does all the modern user experience workflows to get you your dedicated podcast feed and your emails, and you can tweak everything. But he has since franchised it and gotten a number of other, a number of other creators on Passport with, I believe, a different economic model. The only criticism of Passport is not a technology one. It's more about a network effects one. So Substack is known for being very good at recycling email subscribers. And so a lot of email newsletters have life cycles. They sort of have churn rates. Some of them have smiling curves, but a lot of folks, they come in, they test out one Substack, and then they wind up saying, I'm not really opening this one, they get unsubscribed or they unsubscribe or the list gets cleaned and they get pulled off. And Substack does a good job of sort of resurfacing and there's these ranking lists and there's, oh, you signed up for this, do you wanna sign up for these other two newsletters? There's bundles, there's a whole bunch of different ways to sort of like ensure that list growth is happening at all times. And with Passport, I think it's a little more bring your own. And Ben Thompson had a lot of success on that when he was sharing links on Twitter that would go major majorly viral back when he started, and LinkedIn also drove a lot of traffic for him. And so this could make a lot of sense for Ankler if they have their own audience, they have different funnels into the ecosystem, and they're less focused on being in the Substack ecosystem in particular. I believe the Ankler covered us at one point. We did an interview with them. Is that right? We should have some folks from the Ankler on the show. I would love to know more more about this decision, uh, and what it means for the business. Also, just where they're taking the Ankler is an interesting, uh, is the interesting question. But you're not alone. A lot of people are confused, and they were not familiar that, uh, Passport is in fact a platform that Ben Thompson has been building for a number of years.

And I guess recently they've partnered with Automattic. Uh, and so you can imagine, uh, Automattic being able to put a huge amount of resources towards it. I don't know exactly when they partnered, but yeah.

I don't know. We need to watch this Instagram reel of AI potentially taking the job of a very, very important member of the American ecosystem. I won't say human, but let's watch this. AI is gonna take my job and it's a golden retriever looking at a robot. I will never take your job, Golden Retriever. Stay strong. You are irreplaceable. No robot dog will ever take the job of a real dog. But for certain people, I guess walking a robot dog feels like a stunt, feels like a prank. Very funny. But, uh, it is funny seeing the reaction of the dog being like, that thing looks like me. And I feel like there's a little bit of foreshadowing there that when the humanoids start walking around, you're going to see a lot of videos of people looking at the humanoids with the exact same expression that that golden retriever looked at that robot dog. Anyway, we have our next guest. Steffen Simpkin from Squads is in the waiting room. Let's bring him in to the TV Pen UltraDome. How are you doing? I'm good, guys. How are you doing? We're doing fantastically. Welcome to the show. Welcome. Please introduce yourself and the company.

Great to meet you. Excited to be on the show. My name is Stepan, the co-founder and CEO at Squads. We're building stablecoin infrastructure and all kinds of products that help businesses manage their stablecoins and financial operations.

What is the key problem that you find most resonates with customers as you tell them? Is it that they haven't even dipped their toe in the stablecoin ecosystem? They know that they can cut costs, reduce fees by jumping in, but it's too much of a hurdle and you can help them, or they're already neck deep in stablecoins and they need you to help them wade out of the swamp?

I'll give you a quick anecdote. Please. I don't know if I've told this on the show before, but in 2021, I was building a company called PartyRound that was a fundraising platform.

I don't think you mentioned that before on the show. I'm kidding.

Wait a second. So I saw stablecoins getting adoption. I saw a bunch of companies that were starting to raise money with stablecoins, 'cause if you had, let's say, international investor, it was much easier to just send stables than deal with international wires in certain circumstances. And one of our customers was a crypto company that had a multi-billion dollar valuation, hundreds of employees. And when they were onboarding, I was like, do you want to toggle on support for stablecoins? Because I know you're a crypto company, a lot of your investors are relatively crypto native. There's probably some of them that would like to fund their investment with stablecoins. And they said to me, no, we don't want to do that. It's too difficult to deal with stablecoins. Like, once we have them, we'd prefer to just get dollars. And it was a funny moment, because for a lot of reasons, but part of it was because infrastructure like what you're building and talking about today wasn't where it wasn't really, hadn't progressed very far back in 2021.

So I think that's, we'll get back to the story in a second. To give you kind of a sense of what we've done up to now, for the first arc of the company, we actually focused on infrastructure. Infrastructure. We developed programmable wallets for the Solana ecosystem. Today, we secure about $10 billion in assets. And that was a very crypto-native product where we focused on very crypto-native problems of those kind of large organizations exploring the digital asset space. And then through that experience, we actually saw stablecoins start to take this kind of like center stage in the way they run their financial operations. And that's what kind of allowed us to be a little early to stablecoins because we're deeply in it. And we saw real businesses starting to adopt adopt us, even though they were kind of early crypto-native teams. And I think a lot has happened in the last couple of years where essentially programmable blockchains are now secure, reliable, resilient, and actually great to build on. And they, for the most part, solve settlement and ledgering that in the normal kind of fintech context you have to either build in-house or rent from someone else. And you have stablecoins that are now much more mature, right? The stablecoin market cap is way above $300 billion at this point. And I think that is also building a lot of confidence with obviously the regulators and the broader market. And then I think we generally got better at building crypto products, at stablecoin-focused products. And today, our thesis with our new product Altitude is really that you don't need a bank account anymore, right? You essentially can rely on programmable blockchain stablecoins and wallets to run your entire financial operations. And one of the major things that actually happened in the last couple of years that was the big missing piece when we started the company was this idea of, if I hold a stablecoin in my wallet, how can I reliably pay a third party via ACH or via SEPA? How can I connect to the banking system if I still need to while obviously, if I want to settle through stablecoins, I can do that instantly 24/7 and free? And that was really the orchestration ecosystem that came in the last couple of years to market. Companies like Bridge that was acquired by Stripe, which essentially step into this system and exist in the middle between a company like ours that is providing this product experience and this underlying crypto infrastructure and the banking system and saying, we're going to take the stablecoins and convert them to fiat wherever you need to or wherever that needs to happen. And there is actually an explosion in that market right now. You have a lot of these local orchestrators. You have kind of the more global ones, the bigger ones. And that was the missing link for us to say, OK, now the stack is here, and we can actually build financial services in a very different way. And that also was already incredibly exciting to us internally. And then the agentic component came in as well. And we sort of realized that once you have the stack where every layer is naturally programmable, from the settlement layer to the account layer to the money itself, obviously, it's a great place to start integrating agents and getting to the world where you have end-to-end agentic automations, where essentially you as a customers and business need to do less and less over time.

Do you use the forward-deployed model at all? Like when you're working with a company, is it high stakes enough that you'll actually go on site with someone from your team, or do you want to be just fully self-serve? Like, is there a trade-off in your mind here? How have you thought about it?

I think today we work with all types of customers, right? Customers that are already have a majority or some part of their balance sheet in stablecoins or companies that are completely new to the space. So that, you know, it really depends on who we're actually trying to onboard, right? If it's a more native customer, they usually know what we're doing and they're very comfortable with stablecoins already. But also the product experience doesn't really impose any crypto knowledge, right? Like it feels like you're onboarding to a traditional business account, but a lot of the infrastructure under the hood already powered by— is powered by stablecoins, and then you can obviously connect to Stablecoin Rails and make stablecoin payments. So we do forward deploy it when it's required. And I would say today, we primarily focus on startups, SMBs, and companies that are between 10 to 50 people in size. And so that really allows us to— forward deploy is much easier because you, in most cases, talk to people who are decision makers who actually are going to be running the account and interacting with all the primitives that we're building. I guess the answer is that depends. But I would also say that over time, people become— and businesses become much more familiar with stablecoins. And so I think it's becoming— our job is becoming much easier as we go.

OK, so break down, if somebody is keeping the majority of their assets, what they might keep in a traditional bank account with you, and then they want to run payroll, or they need corporate cards to pay for software, what happens then?

So the way it works, at this point, the ecosystem is mature enough where we're able to facilitate all of this through a stablecoin balance, right? So you can either top up your Deal balance or Rippling balance from a stablecoin balance in Altitude account via ACH and traditional rails. A lot of these providers actually already accept stablecoin payments, right? So you can actually top up and use DC directly without ever going into fiat. And then from a perspective of issuing cards, it's a massive market already in crypto where you have stablecoin-backed cards, which allow you to spend directly from the stablecoin balance. There's a lot of companies in the space. I think right now Stripe Sessions is happening as we speak, where they're also definitely going to talk about stablecoin cards, which is a big focus for Stripe internally as well. And so those two things specifically are definitely a reality. I would say there's still limitations. There's smaller things, for example, if you want to do ACH pull, that is still being worked on by a lot of the providers. And we're still spending a lot more time trying to figure out what is the right structure to actually make it happen. Because the idea is that you need to pull in stablecoins from the alt account, but then settle in fiat on the other end. But also, as more of these vendors start adopting stablecoins, our life is becoming much easier.

Security, how I think some people might might say like, yeah, working with my bank is a little bit annoying. There's some things that I don't like. Sometimes I have to go into the bank to physically drop this check.

They back the ledger up on tape deck in a mountain in the middle of all the—

But you get the benefit of peace of mind.

If they get hacked, they're going to set the ledger back to the way it was. And crypto is a little bit different sometimes.

Security is a huge concern for the space since the beginning, right? And I think solving that is paramount. We started from a place where we, when we were building our infrastructure at Squads, it was all about security from the start, right? Like that's what we had to get good at, right? And that's how we got to a place where today we are securing—

Yeah, you have $10 billion.

Yeah. $10 billion, yes. And so that is, you know, that took us about 4 years to get to. And it takes a lot of time to build that trust, but also to build infrastructure that actually, you know, works and it doesn't get hacked. I think security, actually solving security is not the hardest part. Balancing that with user experience. It's like, how can you combine these two problems and have a customer that is able to stay in self-custody, own their money, control their account end-to-end with their own keys, give them the security guarantees that they're expecting, while also allowing them to stay obviously with a stablecoin balance. And so I think the way we solve it is we have a complex but easy-to-explain system where we have multiple smart accounts and essentially these programmable accounts deployed for each customer where each individual signer on the account, so every member on the sort of business level, they have their own programmable account with granular controls and MFA they can set up. And then the business account itself is a programmable object on the blockchain. And there's actually a lot of benefits to kind of moving all of these layers onto the sort of open and transparent system like a blockchain because then you can have very clear guarantees in math and code, like what is actually governing the security of your assets. And I think when we talk to customers, a lot of the questions come up is like, how do I think about something that you guys are building versus just using an FDIC-insured bank account? And I think what we always talk about with those customers is that the only reason FDIC exists is because your money is not actually there. It's being fractionally reserved and it's being lent out somewhere. And obviously, you have this off-chain ledger where somebody is maintaining and making sure that all the entries are correct. And so I think transparency brings a lot of trust and actually helps us to get the product to a place where a lot more customers can be comfortable with it. But obviously, having the Squads background and building a lot of infrastructure that secures billions today is definitely a big lever here as well.

Tell us about the round. How much did you raise? Who led the deal?

We raised $18 million. Solana Ventures led the round. We started the company in 2021, building on Solana from day one. And so for us, it's been a journey. It's been a journey. Exactly. And we spent a lot of time with the Solana team and we obviously are very deeply ingrained in the ecosystem. And now with Altitude, we're essentially using Solana as this secure programmable backend, right, to deliver the experiences for businesses we want to deliver. Right. Also in the round, we have Coinbase Ventures, Hong Ventures that we have partnered for the first time last year. They're continuing the involvement and then Electric Capital placeholder and, and a few others. But yeah, we're really excited. It's really— the round primarily is for accelerating Altitude and helping businesses move their operations onto stablecoin rails. But we're obviously continuing to do things in the crypto space with Squad's multisig, which is our first and core product.

Fantastic. Well, congrats on the round. Thank you so much for taking the time to chat with us. Have a great rest of your day. Talk to you soon. Goodbye. Thank you. Up next, we have Kashif Gupta from HiTouch. He's the co-CEO with a massive round for bringing them in.

Massive, yeah. Driven marketing agents. I gotta say, guys, you can't keep asking John to do that. It's, I— What's wrong, Jordi? I had to—

How do you drink your Diet Coke?

I had to step away for a second because I was gonna—

Well, our next guest is in. In the waiting room. Let's bring him into the TVP and Ultradome. How are you doing? Let's welcome to the show. Hey Jordi.

Hey John. Thanks for having me.

Thanks for joining. Introduce the company, introduce yourself, and then tell us the news. And I have a million questions here, so very excited to talk to you, but give us a little introduction first.

Sure. Akashish Gupta, co-founder and co-CEO of Hitouch. We build AI platform for marketers. And today we're announcing our Series D, which is $150 million at 2.75 posts led by Goldman Sachs.

2.75. Absolutely massive.

I feel like— I don't know. I mean, fantastic category, fantastic industry, fantastic megatrend with AI. This feels a little under the radar to me. Like, how— what is the prehistory of the company? Like, how have you built to this point so successfully? Yeah. So we started in 2020.

Okay. For the past 4 years, we've helped marketers use data to personalize their marketing. Sure. Now we're helping them use AI to actually build on-brand content such as ads and emails. Okay. So we believe and we know that we're the only on-brand AI for marketers. If you use Gemini, it will hallucinate things about your brand, whether— whereas if you use HiTouch, sure, we will get everything about your brand correct.

Okay. At what level do you want to operate? At what level have you operated you've created before, because a marketing team can be thinking about everything from like what podcast the CEO should go on, to a billboard campaign, to a Super Bowl ad, to like a micro-optimization in the Facebook algorithm or their Instagram ad delivery or creative. Like where, and then you could dial really deep into just one niche or vertical. Like how have you thought about right-sizing and creating like a correct solution for the marketer? Yeah, so, I mean, marketing is very heterogeneous.

Um, if you think about it, most marketers are spending their time on strategy and branding. Yeah. With AI, we now actually have kind of unlimited labor at our disposal. Mm-hmm. So the bet is that a single marketer can now do the work of so many. Mm-hmm. Uh, in the past, we were kind of stuck doing things like segmentation and like microscopic ads and emails. Sure. Now with our AI platform, we're letting marketers think of new strategy generate those images and try stuff in market very, very quickly. Yeah. So with Hightouch, the marketer can launch something like 100 campaigns in a week, get feedback on all those, and then launch new campaigns based on what's working and not working.

Yeah. You said images. Are you doing video? What's the status on AI video?

Doing video as well, especially UGC. And UGC is working extremely well.

Yeah. And do you want to be model agnostic and then set up like a harness that makes sure that everything's on brand, or do you want to train foundation models diffusion models, image generation models, video generation models at some point?

Model agnostic. Okay. So we find right now that, for example, Images 2.0 from ChatGPT is the best at skin. Yeah. Uh, we use other models, Gemini as well. Yeah. And so as models get better, we want to make sure that our harness chooses the best model for each task.

And then there's also a world where you generate an image or a series of images with one model, and then you animate them with with a different model, and there's a whole bunch of different, or do text overlay or motion graphics with a different model, or maybe puppeteer some actual code, or like an Adobe tool. Interesting. Jordi? [Speaker:JORDI HAYS] Yeah.

How is the role generally of a media buyer changing? Obviously, hopefully they're using your tool, and that's changing their workflow. But it's interesting that I— it feels like media buying has always been very high leverage. Somebody that's good at Facebook ads could manage $20 million of budget on one account, $40 million on another account. They're very, very, very— it's just a very high leverage role. Is it just helping them be better? Do you think marketers in the future will be able to manage more spend or more accounts? What are the different ways that it goes?

I think there's two main shifts. So as you mentioned, Jordi Hays, there's about— imagine AI is a 24/7 analyst that's helping you figure out your opportunities for improving your media spend. Which channels do you spend more? Is TikTok working better than Snapchat? Those kinds of decisions now you can fully delegate to AI and our platform enables that. But the second and much more important trend, in my opinion, is that if you're a media buyer, you can only buy media against content that you have. Most brands don't have unlimited content. They only have maybe like 50 to 100 pieces of related content. Yeah. With AI, you can now generate unlimited amounts of content. So for example, localizing to different languages, doing different geographies. A great example of this is one of the top marketplaces for home services uses Hightouch, and now they can do different marketing for hotel owners versus homeowners versus renters. And because they've kind of— now they have unlimited content building capability at their disposal, that same media buyer can test all 3 against the market, see which one works against which demographic, and then optimize it.

Do you guys care about finding and eliminating wasted spend? That feels like a solution that— there's kind of an interesting thing where, you know, let's say like, let's take Meta platforms, for example. They want you to get good results, in general so that you spend more money so that your business grows so you can spend more money, et cetera, et cetera, and have that flywheel. But at the same time, if you're running an ad and it's not performing well on a sort of local level or micro level, they don't really care that much if you're spending in a way that's not super efficient. And so maybe there's an opportunity to have a platform that kind of sits above that and says, hey, this is like— to basically turn off and try to identify ways interested in spend. Is that part of what you guys are doing? Is that something you would do in, in the future?

It's part of what we're doing, and why we think we can do it uniquely is because all of the consumer brands we work with, companies like DoorDash, Oritzia, and so on, have given us access to their customer data. So we know which customers are actually converting, and we know the ground truth of whether the ad is working or the media campaign is working.

[Speaker:JORDI HAYS] That makes sense.

[Speaker:AUDIENCE] How do you think about your business model? I mean, there's so many huge companies in the marketing industry, because if they're able to insert themselves in the pipeline of a billion-dollar marketing budget, it's very easy to justify spending millions of dollars on something that increases performance by 1% or 2%. At the same time, there's a lot of marketing tools out there that are just seat-based. And they might give you a call and say, hey, you're pretty big, can we get some more money out of you, sell you some premium features? But this seat-based pricing versus consumption-based, performance-based pricing, how have you geared the company for success? Obviously, growth is fantastic, and the company is very— the valuation is going up, and you're raising money. So I imagine that you've solved it, but how have you thought about solving it?

Yeah, transparently, any company that's still on seat-based pricing is not looking forward to the rest of 2026. We have never been on seat-based pricing. We've always charged our customers by the number of campaigns that we power. And our belief is that with AI, our customers will be able to run significantly more campaigns. And we really leave it up to the customer. If you keep the campaign on, it must be working, and therefore you should pay us. If you turn the campaign off, then stop paying us.

It's really simple. Yeah, that makes a ton of sense. Well, Cool. How much did you raise? I want to talk about the— do we already hit the gong?

I think we already did.

Did we already hit the gong? Yeah. It's been a big day.

We've rung the gong a bunch of times.

Gong abuse today.

Well, congratulations on the progress. Yeah, very, very— oh, the last question, because I've asked some other people in martech this. Are we headed to a future where every ad that a customer sees is generated in real time? On the fly. And the example I always give is like historically advertisers would make one campaign, they'd put it in a magazine or they'd put it, you know, out of home or the Super Bowl and they just knew they were gonna, they had to make it and a ton of people were gonna see it. And then you get the era of like really, really precise targeting and the explosion of meta platforms. And now you can be like, I'm just gonna serve this ad to like 10,000 people. And then it feels like we're headed to a world where the creative is generated on the fly. It's really just like meant for one individual person. And that's only gonna be possible because you can just generate the creative like in real time on the fly based around, you know, that person's unique interests or their purchase history or where they are, et cetera.

I think it's an exciting vision. Will most content be AI generated in the future? I believe so, yes. Yes. Will it be one-to-one or on-the-fly time? I think no. And the reason for that is because most consumer brands of any scale have to go through certain approval processes. And the amount of trust people have in AI is still not on a one-to-one level. On the other hand, could it be—

You don't think AI can get superhuman? You don't think it can be better than the best? Well, so, so, but then at what point is it better than the, the best creative director and the best person that get the best CMO that actually approves the creative to go out, right? Like you can have multiple agents reviewing each other's work.

I think we share a vision, so I hope that we get there. Yeah. And I think AGI will come before marketers do one-to-one with AI. Yeah. But I do think that there is an appetite for this and that the more like, I mean, the more practical way to make AI actually used in marketing So we talked to pretty much every CMO there was all last year. Very few CMOs today are using AI marketing. You ask them why, and it is because of this kind of random Google process. So we believe the way to get AI to actually work is to do one-to-many, maybe one piece of content for every 10,000 or 100,000 consumers. Yeah. And then over time, create more pieces of content. Mm-hmm. Very cool.

All right, well, great, great having you on. Congrats on, on the progress.

And we'll talk to you soon. Thank you. Have a good one. Bye. I want to pull up this post from David Baszucki, friend of the show, CEO of Roblox. He says, we're on the verge of real-time photorealistic generation with what are called world models. These require a fair amount of expensive compute, but costs will come down over time. Today, these interactions are essentially real-time dreams. They lack persistent shared logic that turns turns a video into a multiplayer game or concert or classroom or the holodeck. We believe the ultimate architecture for gaming in the holodeck is Roblox Reality. It's a hybrid architecture. We've been talking about this for months. I'm so excited it's here. It's a hybrid engine that marries the structured data and logic of the Roblox Engine and Roblox Cloud with the generative power of video world models. The Roblox Engine provides the underlying synchronized ground truth, the score, physics, multiplayer sync, et cetera, while our video model acts as a super up sampler to layer on photorealistic detail. We believe this will ultimately remove barriers to high fidelity creation. And if you look at these videos, it makes so much sense. Roblox has always been blocky, but style transfer is the most mature generative AI technology. We all saw this during the Studio Ghibli moment, take a cartoon, tune, make it photoreal. Obviously, image generation is at an incredible place. And so being able to do that at 60 frames a second with Roblox is, uh, just incredibly bullish for the amount of creativity that can go into this world. I am extremely excited about this because Roblox has always felt like a kids' game. It's always felt, uh, like it didn't deliver at the AAA level for so many of experiences, but now you're gonna get the creativity of the Roblox ecosystem married with the AAA graphics that everyone demands, so.

Yeah, I've always felt that Roblox was super well-positioned as it becomes easier and easier and easier to make games. And it's because they have the infrastructure, they have the, they honestly, you know, we've, in talking with David, he's talked about just how intense it is to keep, to have the data center capacity to support support the level of concurrent activity that they have on the platform. And yeah, I'm very excited to see what they do.

Yeah, Roblox is doing well. He's cooking. Well, we have our next guest in the waiting room. We have Dan from Firestorm. He's the co-founder and CEO. Dan, how are you doing?

Sup, boys? How we doing?

We're doing great. Firestorm. Take us through an introduction on yourself, the company, Can you do that like 10 more times throughout this interview, please? Oh yeah, UAV online. That's a good one. We don't use that one enough. Anyway, sorry, introduce yourself, the company, and what you're standing in front of.

Sure. My name is Dan Magee and I am co-founder and CEO of Firestorm. We're based out of San Diego, about 150 people, and our whole vision is to reinvent and transform how we build drones. And what does that mean? First, it starts with advanced manufacturing. So a lot of the drones we buy as a country cost more than a Ferrari and you use them one time. We think that's, you know, not economically feasible in the long run. And then the second thing is you got to be able to move this manufacturing to the point of need. And so behind me you have a couple of our products, the Tempest, as well as the XL, which I'm going to give you guys a tour of here in a couple of minutes. That's awesome.

Yeah. When did you start the company?

We started it in '22. My background was I started a counter-UAS company in 2015. Naval Special Warfare, the SEALs found us and took us to Iraq to fight ISIS when a lot of the big prime solutions didn't work. And I watched how ISIS was able to use off-the-shelf, low-cost drones to effectively fight us to a standstill. Until the counter-UAS systems came in. And in the buildup to the war in Ukraine, I was thinking, why don't we have this type of technology? And it started with, again, rethinking the manufacturing process first, and now it's evolved into actually moving the factory closer to the battlefield.

So, I mean, it feels like the product is the factory, but do you have a first-party drone design or do you want to purely partner on that side? Someone else comes to you with their spec and then you and then you modify your factory to be able to churn 'em out?

We do both, right? But the intention for the beginning is to actually build the drones you see behind us. These are our designs. And then we have a number of partnerships with like FPV providers that I'll show you inside. Sure. Some of the most used drones in the war in Ukraine. Now we're not using Chinese carbon fiber. We're literally turning powder into the airframes and you can go do that anywhere. And the other part of that too is to repair drones. That's something no one talks about. It's actually fixing stuff when it breaks in the field.

Yeah, yeah, that makes sense. Well, I'd love to see what's inside one of those factories. I'm interested in the types of machinery and materials that are going into this. Sure. Can you take us inside? Does the camera move? Is that possible? Oh, the camera moves.

Okay. We have two humans moving it.

Let's do it. Let's go inside. Let's start here at the drone. Yes, please.

I'm honestly a little disappointed. I was hoping it was a drone. A drone baby.

A drone baby. Flying it.

Yeah, I think with a fixed wing it'd be going way too fast.

That good of a pilot, you know what I mean? So I don't want to like crash into anything. So we design everything for modularity, right? This, this whole airframe is 3D printed. Um, you have a rail that runs through the drone so you can put payloads in and out with ease. So something breaks or you want to change the mission on the go, you have that ability to. And then in the back we have both a propeller variant and then a jet variant that I'm going to show you over here. And the whole idea is like add modularity at every essence of the drone, not just build something for one mission, you know what I mean? Yeah. Okay, so come on into the XL. Okay.

The XL.

So our whole adventure starts with this right here. Yeah. This is our build box. Okay. Think of this as like your print cartridge, okay? So this is on wheels, as you see. And so you basically wheel this into the processing station, which is right here. It slides right in. Okay. You then fill it up with powder from our new and our recycled powder. So we're highly efficient, right, from a volume standpoint. Once the Build Box is full of powder, you then push it into the printer. Yes. And we have an exclusive arrangement with HP This is kind of the best-in-class industrial-grade 3D printing. Like, parts that come out of this machine are on F1 race cars. Wow. And if you want to have this printer in a mobile fashion, you have to come to us. That's kind of the IP we've built around this, right? So this is a 20-foot shipping container, and the sides fold down to then allow you to, like, you know, have the workspace you have. Sure. So once it's done printing, You let it cool down and then you put it back in here and you're left with parts like this, right? You suck the extra powder out. Yep. And then you take it into the second 20-foot shipping container that I'm going to show you right now where you just simply blow these parts with powder. I'm sorry, with air. And you have a finished part.

It looks like some, uh, post-apocalyptic biosecurity safety, like zombie apocalypse, I know where I'm going.

Exactly. Yeah. So literally you just come in here. Yep. You blow air on the part. Yep. And you're done. Wow. And so what we use the second container for is actually assembly and R&D. So I'll show you some of the other stuff besides our platforms we're making. You were asking a minute ago about like other people's designs. Yeah. So we have a partnership with a company called Orca. Okay. They're out of Croatia. About 300,000 of these drones were delivered to the Ukrainians last year. We've gotten rid of all the carbon fiber, which, you know, no one wants to talk about, usually comes from China, and we've replaced it with our powder-based solution. You can assemble this thing in 13 minutes, and if something breaks, great, you just print a new piece out. And, you know, you do a new installation. And why is that important, guys? Because some of these drone companies, it's taking them 3 to 4 weeks to deliver repair parts to the field, and that just means the system's basically unusable. So we started as this mobile drone factory. We started sending this thing to bases around the country and overseas. And then this is where the business got really, really interesting. You guys have talked a lot, I'm sure, about the defense industrial base. Okay. And the amount of things the soldiers are telling us they need that we're now printing, like, is incredible. Yeah. So this is an engine coolant tank. All right. You should be able to go to Pep Boys and buy this. This is a 10-month lead time product. Wow. That is absurd.

What kind of engine?

It goes on an LTV. So like a Marine Corps vehicle. Oh, okay. Like, you think we would have these supplies, but the issue is that one, OEMs have consolidated.

Yeah.

Two, no one wants to build kind of parts. Mhm. And so if you look at what the, uh, the Trump administration has really spent a lot of time and energy talking about is this idea of right to repair. We know that from the tech world, right? Of course, like your iPhone. But like, now it's actually coming in to the DOD. And we want to use the XL not only to build drones but to enable some of that repair stuff, right? So one other one I'll tell you, and you'll love this one as a taxpayer in America. This holds a GPS antenna, okay? Sits right here. Yeah. It's on basically every ground vehicle in our military. Sure. It's really, really flimsy. And if it breaks, the person who makes the antenna does not make this repair. Okay. So a bunch of Marines thought of this idea And so now you don't have to buy a brand new $2,500 antenna. Yeah. You simply print a $2 part.

Wow, that's amazing.

Yeah, exactly. Yeah. Yeah. That's incredible.

I'll come back out and— I love it.

Oh, there we go. I love sound effects.

There we go. Look at this. Very cool. So that's it. Amazing. Amazing. Tell us about the round. Tell us about the round. Progress. How's it going there? There's some news.

So, uh, we just raised $82 million led by Washington Harbor, our Series A.

There we go.

Congratulations. Thank you. Uh, joined by Ondis Capital. Yeah, our friends at Liquidity, uh, Geodestic, The Motley Fool. Like, we have a lot of, uh, awesome investors joining us, but our big thing now is to take the Excel not only to like execute on the contracts we've won. Yeah, including most recently an APFIT contract for $30 million. But start taking the XL overseas. This is really going to support the US and its allies like around the world.

Very, very cool. Well, thank you so much for giving us the tour. Thanks for coming on the show.

Incredible tour. These are not easy to pull off. They're not. And you gave us a great—

We appreciate it. Hey, have a great rest of your day.

To the team that was carrying the camera.

I can't thank you enough.

Yeah, great to meet you. Have a great day. We'll talk to you soon. Goodbye. See you, boys. Up next, we have Vlad Tenev from Robinhood. He's the co-founder and CEO, still at the helm. What, 10, 12, 15 years in the mix? How long have you been running Robinhood these days? Gosh, you guys are making me feel old.

Ank. I'm still in my 30s. Okay. For one more year. Yeah, but 2015.

This year, getting old, feeling older or reinvigorated? Age of AI, lots of opportunity at the same time, pullbacks in various markets, stocks up and down. How are you feeling? I'm feeling really good.

Okay. I mean, yeah, on when we talk about the business, I think people have a tendency to be short-term oriented and think about the ephemerality of this quarter, that quarter. But the real story for, for me was we're making big investments. We're continuing to invest in areas that matter. And there's over $100 trillion moving from older generations to younger. And so things like us being the sole initial trustee and broker powering Trump accounts, which are going to put Robinhood in front of of 60 million children, literally extending our market from, you know, 18 and over to 0 and over. We've got investments at all stages of the life cycle. Trump accounts, very early. We've got businesses that are scaling and doing quite well, like Robinhood Banking and Robinhood Gold Card. And then the active trader business and the core engine of the business, which is people depositing more money for, for investing and trading, is looking very, very strong as well. And then that's not to mention we had a big product event in New York where we launched all of these family features as well as the Platinum card. But we've got a bunch of other products that are in the hopper that I think people are going to be very excited about.

I'm sure you're getting questions about the kids thing. I love the idea of kids investing in index funds and holding them until their retirement. Not big on them betting on sports or taking crazy option strategies. How are you thinking about actually rolling out the suite of features? Because there's some that are just like so educational. I remember reading about Warren Buffett and learned so much about investing. Of course, there's some risk. Risk, but then there's other products that are much more risk-on, much more advanced stuff where you're much more like an investment professional to be able to do that responsibly. How do you think about the different products being available to different segments of the community now that Robinhood's so big?

Yeah, I mean, I think we're very careful about leading with and incentivizing the products that most customers would benefit from. So if you actually look at what we incentivize, it's products like retirement where you get a 3% contribution match on contributions that you make to your retirement account. Yeah. And I'm not aware of anyone doing it. We were the first to do something like that. Right. And, you know, retirement's grown to $30 billion, north of $30 billion in assets. In my opinion, and I admit that this is my product, but people should look at it. I think that it's the best retirement offering on the market. Cool. Robinhood Strategies, which is our passive digital advisor offering, incredibly low cost for that one. You might remember we pioneered the concept of a cap. So other investment advisor offerings, including digital advisors, charge you more the more you invest. And we thought, well, particularly for a digital advisor, you're kind of doing the same exact thing. So why does it make sense to charge someone that has $10 million invested with you 10 times more than someone that has $1 million? So we capped it at $250. So 0.25%, but capped at $100K. Yeah. And that's been the first of its kind. So I think we do a really nice job on the passive products. I think it's just that the active trading products where we compete and also do quite well tend to get more of the excitement and attention because it's not interesting to people to talk about, you know, saving money for retirement and having your, your assets passively managed digitally. Yeah.

Joshua in the chat wants to know about getting access to things like Korean stocks. Oh, what's going on with, with Korea? What's— how are you thinking about ADRs?

You know, you know, a lot of exciting international computer stocks and AI stocks over in Korea.

I know there's a couple of semis out in Korea that are interesting. I think the great thing about our business is to some degree customers are always unsatisfied. So this is the first time we're really hearing loud feedback about international stocks. And it's because we're kind of going down the list and there's relatively few things that Robinhood does not offer on par or better than our competitors. So last quarter we would hear about interest on options collateral, for example. A lot of our active options traders would say, you know, we love trading options at Robinhood. The experience is great, the costs are super low, but we'd like to get interest on the collateral that we have locked up like we do at other places. And then we added that. And now, of course, nobody mentions that. We've had a lot of— we have a lot of dividend investors and they had this complaint for a long time about how other brokers pay their dividends in in the morning, we do it in the evening. And one of the things we unveiled, which I think is like super innovative, is we said, okay, we're not just going to go in the morning, we're going to pay your dividends up to 17 days early on average. Wow. So with dividends, I don't know if you guys are dividend hounds, but there's like a record date where if you own the, if you own the stock on that date, you get, you're entitled to the dividend. But the payout date is usually 2 to 3 weeks later. So what happens during that period? I mean, there's essentially no risk that you're not going to receive that dividend. So we thought, why don't we just go through it and make that easy for customers? So now the dividend, the dividend people are happy, except they want a dividend tracker, which we're building. Okay. So yeah, you're going to get down to these things where, yeah, eventually the biggest complaint will be connecting customers to the Kazakhstan markets. And that's what we know we're doing. We're doing really well. But there's always more to be done.

Okay. Take us through what's going on in crypto. It's obviously showing up in the financials. But I mean, we had someone on the show earlier who said it's down only. Is this people rotating from crypto to— I was talking about more specifically the token, the K-shaped dynamic of the token. But it felt like there was a lot of legislation that was happening in crypto to sort of create smoother guardrails. There were no big blowups. It felt like some of the crazier, more risk-on behavior had sort of moved on. And so it felt like a time when crypto could sort of just grow steadily. And yet, it seems like the market's pulling back. Like, how have you processed it? How are you thinking about crypto, both for your business and just as an abstract asset class these days?

Yeah, I mean, one of the things that I think one of the criticisms we get is that, you know, Robinhood is heavily reliant on crypto and trades sort of like correlated with BTC. Sure. Now, if you look at what's actually happened in the past year, We have an incredibly diversified business, 11 lines of business generating $100 million or more in annualized revenue. So crypto is a big one for sure, but it's under 20% of our revenue. And we've got lots of other things that, that, that are firing on all cylinders and there's more in line. So it's 11 now, but Robinhood Legends been doing super well, the gold credit card. Has been growing at a very rapid pace. Yeah. So there's going to be more. Now, when you look at crypto specifically, we're still very bullish on crypto assets over the long run. But as you can tell, there, there's cycles in the market. There's periods where, you know, we're just talking about crypto. Then there's periods that are a little bit colder and more wintry. But developers are developing and we're continuing to build and invest and we see some very positive things. Now, our focus on crypto is actually investing in the technology and making it so that traditional assets are tokenized on-chain. So it's a little bit less like what's the new meme coin that we can list, although we do love our meme coins, and more how can we actually make tokenization a real thing. And I outlined kind of a three-phase plan in our event in France about how first we're going to start tokenizing stocks in a limited fashion. And it'll look somewhat similar to what's available in traditional markets, but in Europe. But eventually you'll get to fully on-chain DeFi-enabled stock tokens that are composable. And we're still executing on that vision. As a matter of fact, Robinhood Chain launched on testnet and it was a very successful testnet with something north of 100 million in transactions. So developers are obviously very keen to integrate into the ecosystem and get access to the Robinhood customer base. So we're continuing to make investments across the full stack— centralized exchange, DeFi with Robinhood Chain. The non-custodial wallet, which I think people are sleeping on. But I think we've got really, really good stuff there. And we announced an event in London in early July where we're going to kind of unveil the next chapter with a heavy emphasis on tokenization.

Yeah, I want to talk about these events because I feel like you have one of the most, I don't know, unique or diverse aesthetics for your events. I've seen you do sort of like the post-game with the step and repeat like it's an NBA game, and then the aesthetics for the, the France event was like this giant pool and everyone's in white suits. Like, is this coming from you? Like, who's driving this? Like, what is the meaning behind all the different stylistic touches? Like, it's cool, there, there's a whole bunch of different directions you're going. Is this something you want to continue to experiment with? Is it like a creative outlet for you? Like, how are you thinking about events?

Well, you guys have to watch. You mentioned some good ones, but the last one we did at the TWA Terminal in New York— I love that hotel— which was pilot-themed. Yeah, that's cool. I thought was quite good as well.

So you want to do a different theme every year maybe, or every quarter? Is that the idea? And do you come up with the themes or do you have a team that's submitting ideas? Like, how does that work?

I mean, it's a collaborative process, but I've been generally coming up with and at least approving and editing the themes.

I love creative director.

I love it.

No, it should. I mean, it's like it's, it's fun to bring the CEO's personality.

I think it's hard to do strange stuff without, you know, the CEO kind of driving it, at least to some degree. Right. And I think that's the risk with corporate events. Yeah. They could end up if you're not being careful. Yeah. Looking like crappier versions of original Apple events. And I think admittedly, that's kind of how we started. Our first event was Q1 of 2024, where we unveiled the Gold Card. And it sort of was heavily inspired by the Apple events. But then we asked ourselves, you know, it's time to grow beyond that. Let's sort of like learn from the masters, but then we've got to push the frontier. So I think that what we ask ourselves is we want to roll out these products, but how can we roll them out in a way that's super engaging and sort of like imbues storytelling and craftsmanship into the actual announcements? So the last one, the New York event, was viewed over 60 million times. I mean, the numbers were just shocking to me. 60 million. If you look at it across X and YouTube, I mean, that's more than most movies.

That's wild. How do you think LLMs are changing retail investor behavior?

Yeah, I mean, I think there's 3 different ways that actually AI advances affect Robinhood, which, which I think is more than most companies. And we're kind of but working hard to lead in all three. So number one, it's just the stuff that we're doing internally, accelerating software engineering, automating customer support. And we were early onto this and we've actually built a lot of our own internal tooling because third-party stuff hasn't quite caught up even now. The second is Robinhood Cortex and AI tooling inside our product. So one of the things I announced at earnings is that Robinhood Cortex, which is our AI family of products, has now been used by nearly 1 million customers. And that's growing very, very quickly. And we just rolled out Cortex Assistant, which is essentially an AI chatbot that has full context into your account and financial data in your Robinhood app. And obviously the next step is can we take some of these Claude Code-esque agentic capabilities and empower all sorts of financial trading and other use cases. So we're working on that. I don't think anyone in financial services has really cracked that. There've been some early experiments, but I think we've got some good stuff to share there. So, and that's gonna be, you know, all year starting next month, I think we're gonna roll out some of these agentic capabilities. But then there's a third one. Okay. You guys probably saw, I know, congratulations on the OpenAI deal, by the way. Robinhood Ventures, which I've been talking to you about since the beginning, invested in OpenAI. Yes, yes. Amazing. So now Robinhood Ventures is like live. We're all sold. Went public.

Yeah. The K-Rex grows. I love it.

Every company will buy. This may be a silly question, but you're a good person to answer it. Why, why did Microsoft, Meta, Google, and Amazon— why do they end up all having earnings? Yeah, is that by design or random? Like, you would think, you would think that you would say like, hey, these companies represent, you know, 20—

I want my special day, I'll go on Monday.

Yeah, you want the attention. You announced earnings today. Yeah, why not, why not, um, why not give give everybody their own kind of moment.

Yeah. I mean, I can speak for myself and how we do it. We do not coordinate with other companies on when to do our earnings. I mean, it's just kind of driven by, right, there's a certain window that you can do it after the quarter ends. And it's sort of like availability, right? You don't want to do it on a Monday. You don't really want to do it on a Friday. So it kind of has to be Tuesday, Wednesday, Thursday.

So maybe the hyperscalers were kind of all in the same window, but then they're like, let's jump together. Like we're gonna raise, we're gonna raise, we're gonna raise.

I would be surprised if there was any sort of collusion involved, but maybe they were all juggling the same conferences or anything or something.

Yeah. There are a surprising amount of questions in the chat about the gold card. I wanna know, both how it's going and I want to know, is it a— like, how unfair is it to characterize it as a marketing project? How good of a business is it? How good of a business will it be? How important is it to the overall ecosystem?

Yeah, it's a strong business. Okay. So right now we've got 800,000 Gold Card holders currently and they're generating something around $15 billion in annual transaction volume, right? ATV. So it's a scaled, like, large offering. The reason you're hearing so much noise about it is that pretty much everyone in the country wants this credit card. It's by far the best deal, 3% cashback on all categories. That's why people want it. It's like the best cashback coupled with what I think is the best customer experience. And so we're getting a little bit of criticism about can we be rolling it out faster? But if you look at the rollout, it's sort of like on par with the fastest rollouts of other successful credit cards in history. If you look at the companies that have rolled out faster than us and we have to make sure we monitor credit quality and all of these things, the ones that have rolled out faster than us have, have not been responsible and have run into trouble. So I think we're balancing two things. We want to roll it out as fast as possible while also staying responsible. So certainly we will cross 1 million cardholders this year and 100 million in ARR, and we'll probably do it well before the end of the year as well.

So were there any other incentives that were at the top of the list for you, where you thought it might be interesting? Like the classic, you know, credit card is like you get airline miles, travel, and that, even if it's a lower percentage, it's a worse economic deal, it lends itself to great marketing because Visa is this abstract thing, or Amex is abstract thing, but then they show you a picture of you getting on a business class flight, going to Hawaii, staying at a 5-star hotel, and it's like, this can all be free. And it's like, you could just pay for that with 3% cash back, but it allows more storytelling. Was there anything else that you were like, ah, that might be an interesting twist? Or was it just like pure economic value?

Well, the, the Gold Card is a simple daily driver product. And the whole point is you don't have to think too hard about maximizing your rewards, which, which I think is really attractive because for a lot of people who aren't like super points optimizers. They want to just know that they're getting a good deal and that they have a nice card. Yeah, but probably not, you know, spend time on the points guy and compare. There are people that we want to serve that are in that latter category, which is why we unveiled the Platinum card at the Take Flight event. So that's a more premium card, $695 annual fee made of Pure Platinum. So it has pure platinum in there. It's one of the heaviest cards. Love it. And that one is a little bit different. Yeah. So you get 5% back on dining, which I think I haven't seen a better deal for restaurants out there. And we've really been getting good feedback on the wellness benefits. So there was some criticism that I should own up to after our announcement. And by the way, our announcement, it was like the, the headliner. It was, it was intended to be the appetizer. But so many people got excited about the Platinum card that, you know, the waitlist signups greatly exceeded our expectations there. But some people were, were saying, you know, maybe it's a little bit complicated, it's a little bit too coupon-y. And so our team's been hard at work fixing all of that. So, so actually Even though people were excited, we're going to come back to them with something better ahead of rollout that fixes a lot of the concerns.

So yeah, is there, is there a card in the pipeline that goes above? Are you going to do the tungsten card for our crypto friends?

Maybe.

Yeah, the tungsten cube that was very popular for a while.

There's something there. Yeah, it's been considered and we've been thinking about a black card. Yeah. So yeah, not, —it's not like actively in the cards.

Like carbon nanotubes in the card or something. I don't know. You got to get funky with the materials for sure.

We've had some ideas for sure. Yeah, we have some good, like we know the design of it. I think the question just is let's digest gold, roll that out fully. Let's digest platinum. And then I think we'll see. We can go in both directions, both like sort of like more basic starter card and as well as there's always more room at the top.

Yeah, the whole pipeline.

I would like the mahogany card.

The mahogany card would be good.

The official wood of business. Yeah, a wooden card maybe.

A nice velour card. Velour would be good too. Well, anything else, Jordi? No, this is great. Thank you so much for taking the time. Congratulations on the progress. Good to see you. All the different initiatives you have going on. On. Uh, look forward to talking to you soon. Have a great rest of your day, guys. Goodbye. Uh, later. Q1 earnings is in, uh, all in one place.

Brandon Gurel has data. I was trying to bring in Andrew Reed. Okay, I was trying to bring in— yeah, Andrew Reed, buddies, but it didn't work.

Well, there's some, uh, there are a bunch of— yeah, we, we can bring in our next guest soon. Uh, run through, run through earnings. But, uh, Google's up big, Meta's big. Amazon's down a bit. Microsoft's down a bit. But it seems like performance was strong across the board. Earnings per— revenue. Let's do revenue. Microsoft, $82 billion. $81 billion was the estimate. Meta, $56.3 billion. $55.4 billion was the estimate. Google, $109.9 billion was the revenue. $107.2 billion was the estimate. And for Amazon, $183 $1.5 billion against $1.77 billion estimate. So beats across the board, but various levels of expectations. We'll be digging into the CapEx numbers and all those other downstream metrics that we discussed at the top of the show. But we have our next guest in the waiting room. We have Parag Agrawal, and I believe we have Andrew Reid as well. Welcome to the show. Hey, guys. How you guys doing? What's going on? Thank you so much for taking the time. Let's kick it off with what you got done this week. We got Andrew Reid on the board. Let's go! Let's go! Very, very, very, very good. Yeah, how'd the deal come together? What's the thesis? What's the update? What's the progress like? Sort of reintroduce the shape of the company. Company.

John, Jordi, great to be here again. I don't know, last time I was here, we were building, we were talking about this abstract notion of AI agents using the web and wanting web infrastructure so that agents can use the web. Yeah. Right? And it's been 6 months. Yeah. And the agents have shown up. Yeah. 100%. And these agents are using the web. Web. Yeah. And we've built a bunch of technology and productized it in the last 6 months. We only launched the product in August, and we're seeing agents use the web to do useful things for people. Yeah. And useful things for businesses. Uh, and really, I don't know how many agents you use, but like this year, the, the breadth of customers, the quality of customers, the variety of use cases because they've all exploded well beyond our wildest imaginations.

Yeah. Maybe we should start with the key things.

Well, let's just go around and say how many Mac Minis we all have. I personally have 20 running right now. Oh yeah, you do? John, I know, has 16.

I have a single one, single Mac Mini, barely doing anything. Continue. No, I'm interested to know about just, well, two main things, but let's start with the first. In terms of just like robots.txt, how much of the internet can be seen by agents, is this a wall of some sort? We've seen different reports about different organizations with content on the internet having different policies. And is, is it, are customers coming to you saying like, I want to be a good actor and so I don't want to violate any of the Roblox.txt agreements, even though they are not necessarily legally binding in every scenario, as I understand it. But is, is, is seeing the entire web, being able to do everything across the web, a unique value proposition in this day and age?

It will be, right? If you think about it, there's a lot more content on the open web. I think more recently, as people have started worrying about like agents sort of stealing their content. Yeah. Certainly a fear and perhaps a trend in some pockets to put content behind walls, whether it be via robots.txt or some sort of a blocker. Like that is a legitimate fear and see some of it happen. We actually started the company to in part solve the problem. The common thread from my work at Twitter to through, through this company was to incentivize more content to be out in the open for everyone to see and use. Mhm. So we've also been working with people, content owners, publishers who publish on the web, mhm, to effectively align incentives of theirs for the world of AIs. Mhm. So when we talk about rebuilding the web for AI, it's not just about building technology to serve people building AI solutions. It is also to empower content owners and publishers to actually have sustainable, real, awesome, fast-growing businesses that share in the value that they help create via AIs. Yeah. And we'll have more to share perhaps, and, and we'll be back to talk about it more.

Yeah. How are you thinking about, uh, the various modes of, uh, he's vague posting on the On the show.

He's been posting live. I'm just fishing for another invite.

How are you thinking about the various modalities with which to interact with internet resources? Like we have APIs, we have web front ends, agents can use both. There's MCP servers, there's potential for an MCB standard that's metered with stablecoins or even just credit card payments. There are solutions to every different end user. Endpoints, some of them require more RL and training. Do you see any of them as a point of differentiation where you want to be like the best for setting up an API integration just on the fly seamlessly so you can get the action done? Do you want to just be able to deal with any frontend no matter how janky it is from the— maybe they haven't updated their website since 1995, it still works. Maybe it's using some crazy frontend framework, it still works because you're really good at agentically interacting with the UI.

Yeah, no, I think so. A few different things here, right? Number one, you do have to take websites or content published for every era and allow agents to access them. Yeah. So a bunch of the heavy lifting we do is to enable that to happen today. Additionally though, So none of those are the most efficient ways of doing things, right? And ultimately, the world wants to move towards more efficient solutions because that allows you to do more and do better. And so we built a bunch of things where content owners can provide data more efficiently. People saw LLMs, or text, get around a little bit. Very small part of the web uses it, but we translate the entire web to make it easier for agents agents to use it. Now, in terms of APIs and MCPs and whether you pay for an API with a traditional credit card and account or use a crypto protocol, whether it be XPRO2, Tempo, these are all— all of these things make sense, right? They make sense for different customers, for different agents in different contexts. Yeah. And really the way we think about solving the problem is we're building essentially an infrastructure layer. Even if you look at our product suite, it is very, very— it exposes various layers of our technology stack. We're building this vertically integrated technology stack. We have first principles way of crawling the web, indexing the web, ranking the web, building agents on top of it, building proactive agents on top of it. Yeah. And exposing all of these things as APIs and MCP servers and as CLI. Because different customer needs and different customer problems and different customer circumstances demand slightly different solutions. It's like AWS exposing a VM versus a managed service to run compute versus even higher level abstraction to run serverless. And it's the same that shows up in this infra environment of using the web.

What companies would be crazy to not sign up for the service today?

If you are using an LLM and are not trying to get it to forget everything that's out on the web, you should totally be using us. And the answer, it's like, it's not complicated because— Everyone. Yeah. And it's not complicated because if you think of like any products that you use which are powered by LLMs or agents, you have this mental model now that they are both smart and all-knowing. And if you don't give them the best access to all of the live, fresh content on the web and all of the long-tail content on the web, like, that illusion of all-knowing goes away. And so you kind of are forced— like, initially, ChatGPT, we were all in love with it. Did not have web. Now try to turn the web off for a day in whatever agentic product I think. Coding agents, we all used to use them without— they did used to not crawl the web. I don't know if you remember, but I used to like paste in doc links for it.

Yeah, yeah, yeah, yeah. I used to like export PDFs and then like upload the PDF from the website that you wanted to add extra context into the context window. Now You just assume that it knows everything. It's crazy.

And that was a year ago. That was a year ago. And today you're going to feel like a caveman doing that.

Okay. We have to talk to Andrew because the chat is calling you a silent partner. They're saying nothing from my end. Thanks. I want to know, did the progress here with this company, did you expect it or did it hit you like a flashbang? Rolling flashbacks. Talk me through your journey. I'm gonna take that one too! Okay— no, no, first off, allow me to insert myself.

Please! Since the last time I was on TVN when you guys started going down like the deep AI infrastructure rabbit hole, this show has changed and I am unprepared so my anxiety level increased quite a bit and the good news is I've been in some conversations with Parag, with other people who are like crazy deep on web infra. Yeah. However deep, however deep Jon can go, I think Parag can go even deeper. So I feel good about that. Good. Yeah. First off, I'm delighted to be on the show with my newest board and it's incredibly exciting company. Honestly, the I think like many of our most interesting investments, like when you start seeing a bunch of portfolio companies adopting the same product and, and also in these high-growth areas, you see, you know, you see background agents and long-horizon agents out in the wild to a certain degree. But then you have the board meetings where you see all the prototypes and the products people are planning on launching the rest of this year. Sure. And I think the idea of agents going from something that runs locally and has very limited access to something that runs in the background all the time and has basically a full workspace. And then you think about all the elements of knowledge work and how they overlap with the web in terms of things changing in the world to trigger actions to the actions that agents have to take in the world. and then people viewing Parallel as kind of core to what they want to do is, uh, you know, at some point, like, and obviously the growth rate's crazy, so it's not that complicated.

I like it. That always helps. Icing on the cake.

Yeah. Uh, um, is this, is this a period of uncertainty in venture broadly around how certain markets will break, where powers and moats will emerge, or do you think the smoke has cleared? Do you think that it's actually pretty clear right now?

Well, they have that Sequoia, they have the crystal ball.

You have the crystal ball?

I'll consult with that if they need to, if there is some fog of war, you know?

Yeah, I mean, there is a lens that's like, it's up only right now. Like it's just such a big market that even the companies that aren't the hottest one day. Yeah, you look at, I mean, just like how your process processing the market, right?

Look at CodeGen and like a year ago, if you just backed the top, yeah, like 8 companies, they all did well. You've done really well.

Yeah, I think for what it's worth, the, um, uh, you know, there is one sure thing, which is the data center buildout. It's underway. And you know what that's going to mean for token pricing and what's going to be possible. And obviously, you know, the model improvements, etc. Um, it's interesting because, you know, I was watching Vlad before this and I remember, oh yeah, Interestingly, there's something similar with, you know, when I first met Parag, it was in Palo Alto, you know, internet celebrity building something that looked like one company from the surface. And then the further you got into it, you realize there is doing so much more than meets the eye, you know. And I remember with Robinhood, you know, according to the internet, it was, you know, commission-free stock trading app, but they had just built this internal self-clearing system, the first one that had been built US in 30 years, I think, since Vanguard. And their willingness to tackle long-term, very hard engineering problem— problems to serve their needs and their customers' needs for a long period of time was extremely unique in that industry. And it reminds me of Parallel, like, just if you go spend time in the Palo Alto office and you see the stuff they're working on, these are hardcore infrastructure engineering problems. That are designed to serve both scalability, reliability, speed, everything else, cost needs for customers a year or two from now. And like the world just doesn't quite see it yet. And then also like I remember when Parag first mentioned, you know, agents are going to use the web 1,000 times more than humans. And it, you know, you kind of roll your eyes because like that just seems ridiculous. Yeah. But then if you just actually look at the trends and draw a line forward and imagine like what cloud agents are going to be and how they're going to overlap with knowledge work. It actually is probably like an undershooting of the number. Totally. Yeah, that's personally my bear case.

I mean, I mean, we're seeing like uptime problems with major pieces of web infrastructure because the demand's so high. It's really showing up everywhere. Chip shortage, CPU resource shortage, so many different places. And yeah, I mean, we saw early glimpses of it with, you know, certain, just the number of queries a single deep research report would make or something like that. So very, very exciting times. And we appreciate you both coming on the show to break down for us. Great to see you.

I'm very excited for your partnership.

And congratulations on the round. We'll talk to you soon. Cheers. Thank you, guys. Goodbye. Up next, we have Gabe from Rogo. He's the founder and CEO with a massive Series D raise. Let's bring in Gabe. Welcome back to the show. How are you doing? What's up guys?

Thanks for having me. Great to see you.

Of course, anytime. Big news, let's kick it off. How much did you raise?

I wanna hit the gong. $160 million, $2 billion valuation. Whoa, fantastic. Fast.

Leave some valuation for the rest of them. Sequoia Capital again. I think we went 3 Sequoia companies. This is the Sequoia show, I guess, right now. But Thrive's in Khosla, JPMorgan, Box Group, Mantis, Jack Altman. You got a whole crew, a murderer's row. What's driving the growth? Why are they putting more money in? Why is the business working?

So this was led by Kleiner. So it was new investors. Mamoon, Nadia, super excited to work with that whole team. I mean, we're trying to transform finance and finance is maybe the biggest knowledge work category on earth, 15% of GDP. Sure. We've seen crazy uptake over the last 6 months. I think what was happening in all of 2025 with software engineering and cloud code and Cursor taking off, we've crossed the chasm in finance. And I think we're seeing in investment banks, private equity firms, hedge funds, folks really starting to use these tools. Tools to be more productive, to make better investments, to reimagine how they do their work. And that's driving the investment.

Yeah, I feel like if, if, you know, if somebody woke up from a coma and they looked at your financials and they looked at your business, they'd be like, this is the most incredible investment, I gotta write the check, right? But then there is this like looming cloud of like AGI, ASI, steamrolled by Lab, right? So are you grappling with that? How do you talk to your investors about like the defensibility, the moats, the, the go-to-market motion? Like what makes ROGO special over the really long term? Or are you just like, AGI is fake?

I mean, I think if you're not grappling with that stuff, no matter who you are, you probably should be. You know, I don't care even if your business is making plastic cups, like things are going to change and you need to be prepared. For us, in particular, I think folks throw a lot of shade at the application layer, and I think rightly so in some places, right? If you can't provide incremental value to these models such that big enterprises want to work with you and trust you to transform their businesses, you should not exist, right? Like that, that, you know, it's a rising tide and it keeps you honest about what is the quality of what you're creating. Yeah. For us, we actually see there's structural reasons for a vertical player in finance. Okay. One is that if you are the biggest investment bank on earth, you want to always be able to run the world's best models on top of your data, on top of all your workflows. And the reality right now is cloud is the model du jour. But 3 months ago with Gemini, 12 months ago felt like ChatGPT was indomitable. Who knows what it will be in 12 months? In 24 months, you might be able to use the KIMI models to do everything you want and you don't need a token max on the frontier model labs and you want someone whose incentive is aligned with your business incentives, not just consuming tokens. And so there's some structural advantages. And then there's also things native to financial services that are just very deep and complex, right? Like financial services is a collection of niches, each with their own regulation, their own data, their own definition of good. To think you can just have this big generic tool and do all of the work is, is a little bit nonsensical. Yeah, um, that's not to say—

yeah, the, the thing on token maxing, do you think it's possible that the application layer emerges as like the, basically the party that's aligned to the end customer to keep costs in control, which like right now businesses are not carrying around token, you know, you starting to see a few people pop up and say like, hey, we spent 2, 3 times as much on tokens this month. Like, did we get 2, 3 times as much value? That's starting to happen in the last, like, call it, month, but you can imagine over time businesses will just care, you know, if tokens become a core expense to every business, which I would expect happens, businesses will care a lot about optimizing that. And so there's an opportunity for the application layer to be the aligned party that's saying like, the lab is not gonna necessarily tell you like when you should use this other model, but we will, because our incentives are aligned. You're going to pay, like, you know, you can figure that out in a number of ways. Yeah.

No, I mean, many ways we can be a model broker and help you pick the best bot for the best task. And that's, in fact, how one of our customers described us the other night. And imagine being a huge firm, and you have the GM of a division comes to you and says, I just hired $100 million worth of people over the past year. You'd be like, and what? What did you guys get done? Like, what needle did you move? Right? Like, when I saw Uber CTO saying we already went through our cloud code budget, it's like, okay, to what? To what end? You're going to need people that can align token consumption with business value. And for us, maybe it's per deal, maybe it's per fairness committee memo, maybe it's per IC memo. You can actually think about how do we make sure that we are supercharging the things that are revenue generating for the org as opposed to anything else. Yeah.

How do you think about deduplication of work? It feels like there's There's some organizations where you might have like 2 vibe-coded dashboards that are identical, and it's just repeat work. But this has happened for generations. I'm sure going back to investment banks, there's 5 people that built the same DCF. They should have used a template. There were at one point templating companies. I think Canalyst was one that was focused on those financial models and trying to standardize them across the organization. Like, how much of what you do you do on top of the cherry on top is like change management, organizational design, education, making sure people are reaching for the most efficient tool for the job because the tool's changing like every week. A ton.

I mean, a huge part of our value proposition is the kind of forward deployed banker motion where we can go on, go in and say, hey, what are, what is all the muscle memory you have on how you do work and how do we map that to what AI is capable of today? But if you hire Accenture to do that or McKinsey or Bain, well, guess what happens? You have to do it again in 3 months. Yeah, the models are changing so quickly and you kind of need to reinvent these processes on the fly. Yeah. The reason it's so important to an enterprise is, you know, imagine, imagine if Elon Musk was also the president. I'm sure self-driving would look different. He would be able to change the traffic laws and design full self-driving to be far more efficient. You can't do that, so you need to design for existing traffic laws. If you're a large enterprise, you can also change the traffic laws. You can change how your organization is structured and thinking about integrating these models in far more kind of interesting pioneering ways, but it takes thought partners to do that.

You have a sales team, right? Like individuals that are working with large banks to deploy the product, set up the customer for success, correct? Yep. What does it take to succeed? I would be worried. I'd be worried. I don't know. I just wanna know, like, what does it take to succeed when you're hiring a sales rep? Like what background, what characteristics, what stands out where you're like, they're gonna succeed here versus another company that's selling software or something? Yeah. What really jumps out to you to get a job?

Well, we do have this forward-deployed banker, forward-deployed investor role where we hire a ton of super smart, curious, ambitious folks from Wall Street who want to get hands-on with the technology. Sure. And because we have so much of that domain expertise, you can actually pair someone like that with, you know, a great enterprise seller that knows how to navigate an organization, knows the politics of selling, knows what it means to be a strategic AE. And, you know, they can, they can work well together. You do want people who are open-minded and curious and creative because it is changing so quickly. But my mental model is it's very hard to, you know, what a good AE does is build relationships. Right. And now, you know, with these tools, they don't need to focus on building the DAX or the analysis of the pricing schedules as much, but they can spend more time taking clients to Knicks games or, you know, whatever else might be productive, and that's hard to replace.

That's why Josh Kushner bought the SF Giants.

How, any like rumblings around big financial institutions, like changing their early career hiring or the number of people that they're hiring? Is it flat? Is it down? Is it up?

I mean, people are very interested in how they create the next generation of dealmakers and investors. They're like, yeah, if you're not doing the modeling I used to do or the work that I used to do, how are you going to learn the job? I think every firm is rethinking how do you have this pipeline of senior MDs, of senior investors? Because if you, if you don't have any more juniors, you are missing out on the folks who will be revenue generating in 5 to 10 years. I think there's a lot of thoughtfulness around that. I think in terms of, you know, thinking about the org as a whole, these are super ambitious and institutions that wanna win market share. And if you say to them, hey, a junior banker now can support 4 MDs instead of 2, and you can now hire more MDs and enter new markets and compete more aggressively with your peers, that's typically the way these folks think. But it's a spectrum.

Yeah, that makes sense. Well, thank you so much, Jordi. Anything else?

No, very cool. Congrats on the round. Thanks, guys. I hear about ROGO every week from our friend Patrick O'Shaughnessy. Oh yeah. Fantastic sponsorship. He's been on an absolute tear.

He has. Yeah. So gotta go tune in. He's the man. Well, thank you so much for coming on the show, Gabe. Good to see you. Thanks, guys. The team. We'll talk to you later. Goodbye. Bye. There's some good news. If you are a cow, if you're a cow and you're worried about privacy, Google has your back. Google Maps specifically has your back because if you're caught and you're— if you're a cow on the side of the road, and you're caught by one of the Google Street View cameras, Google will blur your face out if you're a cow. That's good. That's good policy. That's "don't be evil" in work. Don't dox my cow. And cow, the cow says thank you, Google Maps. Anything else you want to touch on today? No, I'm looking forward to— Stripe Sessions tomorrow. Tune in, 11 AM. Pacific. Sharp. Uh, Connor McGregor also has the latest adventure. We'll dig into that later. Have a great day, everyone. Leave us 5 stars on Apple Podcasts and Spotify. Sign up for our newsletter, tbpn.com. We'll see you tomorrow.

Goodbye. Cheers.