I see a large IPO on the horizon.
You're surrounded by journalists. Hold your position.
Overnight success.
The Blazemaker 3000.
Double.
This misinformation.
Clearing order inbound.
That's just wrong.
We are surrounded by journalists. Hold your position.
Come, get up.
Trust the experts here.
Five.
We are experts.
Founder Bel.
I see multiple journalists on the horizon. Stand by.
Express.
UAV online.
Blaze.
Double blaze.
Triple blades.
Double kill.
Cup is.
Wins.
Team deathmatch.
We are experts.
Yeah.
Triple blade.
Let's just roll. Right.
Marky clearing order inbound.
Come on. Get up.
We are surrounded by journalists. Hold your position.
Strike one.
Strike two.
Activate golden retriever mode.
Marky clearing order inbound.
Five.
I see more journalists on the horizon. Founder.
Yeah.
You're watching TBPN. Today is Wednesday, July 15th, 2026. We are live from the TBPN Ultra Temple of Technology, the fortress of finance, the capital of capital. Let me tell you about Ramp.com. Time is money. Save both. Easy to use corporate cards, bill pay, accounting, and a whole lot more, all in one place.
Big news in the fintech world, not related to Ramp actually, but Stripe, another Founders Fund project. It's Founders Fund day. Stripe has offered to buy PayPal. This was rumored on the timeline on and off over the past year as PayPal's been a little beat up. We can pull up the five-year chart. It's down over 80% since the pandemic highs. If you go back further, the 2020 to 2021 era for PayPal was like sort of an anomaly because of the big shift to e-commerce during the pandemic boom. But things have not been looking good for the last five years.
And so just a couple, was it a couple months ago, Shiel posted in February of 2026, he posted a little question for the timeline. He said, 'So who's buying PayPal?' He says it has the opportunity of being one of the great distressed value opportunities in fintech history. It's down 85% at the time. Still generating $5.5 billion in free cash flow. This is a $50 billion company. We'll go through the PayPal, the Stripe offer which is at 53 billion, but you know, you're getting a 10% free cash flow yield. I'm unaffected.
Sorry, I'm still working through these new ones. But if you think about it, it's like you have a $50 billion company printing 5.5 billion in free cash flow. You can use a lot to finance debt, obviously, which is what's proposed here. 400 million consumer accounts. So lots of consumers have PayPal accounts, have Venmo accounts, maybe they're not fully active, but they're ready to be engaged. It's a huge footprint and huge distribution mechanism. And they have bank info attached. That's difficult. You can't get that from many other consumer applications. That's sort of the white whale of consumer. How deep can you go in the KYC flow in the consumer bank relationship? You really know these customers, which is valuable for a potential buyer like Stripe in this case.
So checkout buttons are on millions of merchant sites, although that has been slowing, and that's part of the reason why a change of ownership might make sense at this point. And they also have this peer-to-peer brand with Venmo, which was bought by Brian Johnson, Braintree, and then sold to PayPal lore. So Shiel, the first company he calls out is Stripe. He says they have a lot of desirable assets for Stripe: a consumer-facing checkout, bank account details for hundreds of millions of consumers that you could integrate into Stripe's checkout flow, and they have a consumer brand in Venmo. Apple would also be a potential buyer, says Shiel. Good complement to Apple Pay for e-commerce penetration. They never got social payments going. Could get Apple back in buy now pay later. Remember, PayPal bought a, did they buy Afterpay? No, that was Block bought Afterpay.
PayPal. Does PayPal have a buy now pay later?
No, they were just building this out themselves.
Oh, they built that up themselves and not very well.
Got it. So Shiel says those are the two logical options from a business value creation perspective, but he identifies a problem. He says in both cases, Apple and Stripe, the culture fit makes them a non-starter.
I take it back. They actually did buy a BNPL platform.
Which one?
That I've never heard of called Payy. Okay. 2021 for 2.7 billion.
Okay. So they do have an asset there that they are hopefully trying to grow.
But prior to that they had already built their own.
Got it. So Shiel calls out that culture fit might be a problem with PayPal integrating into Stripe. PayPal is a quote sprawling legacy fintech with 25,000 employees. That's a huge company. Decades of technical debt. Neither Stripe nor Apple would want to absorb that. Although technical debt, you can clean it up. Coding agents not too bad these days. Maybe less of a problem. Apple also might face big tech antitrust with this type of thing. That's a good call out. And yeah, you do have to wonder, Stripe incredible on the product side, on the innovation side, on just building a fantastic massive $159 billion as of last tender offer business. But are they the team to be the ruthless cost cutters if that's what it takes to turn PayPal around? That might be a different challenge culturally. Same thing for Apple. They're not this private equity firm that goes in and buys legacy assets and turns them around. So a little bit of a culture shift that I think Shiel is correct to call out.
He also identifies Visa and Mastercard as potential buyers. He said they could both afford it and they've been creeping into merchant acquiring and checkout. Also, we talked about that the big banks are now taking a shot across the bow of Visa and Mastercard with their own card network. So there's potential that Visa and Mastercard might want to expand into a different territory, although that's not what's playing out right now. But who knows what other bidders will come out of the woodwork now that Stripe has made this offer to buy PayPal for 53 billion. So PayPal's checkout button placement is enormously valuable real estate for either Visa or Mastercard. The networks have been trying to move beyond interchange into direct merchant relationships, and PayPal could accelerate that by years. I think they may be burned out on antitrust. Either network acquiring the largest independent online checkout provider would and should face brutal regulatory scrutiny, says Shiel.
He goes on, he says, 'What about Elon? Elon co-founded PayPal and always wanted it to be called X. So there's some poetry in it coming back under his fold as X. Of course, he is a Stripe shareholder, I believe. I think he was an angel investor. One of his very rare angel investments was Stripe. So, you know, he might be getting a slice of PayPal with this. Not too bad.' His bandwidth is spread impossibly thin across Tesla, SpaceX, xAI, this is pre-merger X, politics, and replying to posts at 3:00 a.m. Never been a problem for Elon, but you could have made the case he was spread too thin before he started acquiring the last several companies too. Technical debt at PayPal is a big challenge that he knows. I would never count him out. I just don't see it though.
And then he says in his opinion, Shiel says JP Morgan makes the most sense. They've spent heavily on payments and acquisition. An acquisition could get them closer to building a consumer super app alongside Chase. 50 billion would be a huge acquisition even for them, but they could stomach it, assuming they could get regulatory approval. Venmo gives them a P2P brand that they've never been able to build, especially among younger consumers. PayPal's branded checkout button sits on millions of merchant sites. Remember, didn't JP Morgan, is it JP Morgan that bought that company that Charlie Javis founded and then she went to jail or was accused of crimes because of that pitch for her company?
It was consumer facing.
Consumer facing.
They had a relationship with alleged, I guess they positioned it as having a lot of relationships with younger consumers, college students. And so I think the deal came in like north of 100 million, and then it was later alleged or discovered that those accounts were maybe generated programmatically. I think this was the pre-AI era. Now it's even easier if you need to generate a bunch of fake data. But don't do that because it's for.
The question is like Stripe makes so much sense because to be able to pull off an acquisition like this, to be able to get the level of debt needed to do this, you need to have so much confidence that they're going to actually be able to. I'm assuming they'll end up reducing headcount pretty dramatically. I'm assuming Stripe looks at this and says, 'Hey, I don't think we need 25,000 people running this business.' So who has the operational strength and know-how to make PayPal run? If you could make PayPal even run half as well as Stripe, that's probably worth double what it's currently worth. So if you're a lender here, or a partner, I don't see that you can find a better partner than Stripe on a deal like this, period.
Let me take you through the lenders. First, I'm going to tell you about Console. Console builds AI agents that automate 70% of IT, HR, and finance support, giving employees instant resolution to access requests and password resets. So Stripe is a fun option. I agree with this. It's funny, this Thiel-backed fintech company buying a Thiel-founded company. Full circle. The first offer is already facing some pushback. Michael Burry, I believe, is a shareholder and he posted on his Substack about why he's not selling at this current price. He thinks that PayPal can get back to their previous share price from a year ago. But the basic structure of the deal is as follows: 53 billion for all of PayPal, $60.50 per share offered, 28% premium over yesterday's closing price. But it's a steep discount to last year's valuation. 50 billion has been committed in bank financing. And the FT is reporting that PayPal has been reluctant to engage with the offer thus far. Stripe is much bigger than PayPal at this point. Also interesting, this offer was apparently sent over to PayPal weeks ago, maybe a full month ago based on the reporting, and it didn't leak. So whoever's working on this team, good job. They kept it.
They're taking it public. They're taking it public. Enrique Lores, the CEO over at PayPal, just got the gig. He was even over at HP, and you can't imagine that he took the job wanting to just basically be acquired in a short period of time. Exactly. I would imagine he wants to get credit for a proper turnaround here. And so I think they're going public with this now in part to try to put pressure.
Well, speaking of public, let me tell you about Public.com. Investing for those that take it seriously. Stocks, options, bonds, crypto, treasuries, and more with great customer service. So Stripe is partnering with private equity firm Advent International for the deal. They have a lot of deal experience in fintech specifically. And I think they're going to go 50/50 on the equity portion, but then there's a whole bunch of bank financing coming in. And they will point in their offer to the fact that branded PayPal checkouts are slowing. There's fierce competition from Apple Pay, Shop Pay, Klarna. These are reasons to sell. PayPal will likely fire back saying, 'Hey, we just brought in a fresh CEO. We have a restructuring plan in place. We're executing against that. Once that happens, once we're successful, we will get back to the 70s for our share price, which we traded at just a year ago.' So yeah, maybe we're not going to go back to the pandemic highs anytime soon, but this offer is just too low for right now, but we will see.
Overall, I mean, Stripe has done an excellent job expanding into many adjacent markets. Atlas is a great example. And they've never cracked consumer, but also they've never really tried in my opinion. And so I think it'd be really awesome to watch this play out, see what they do with it. But it does seem like it'll take a few more rounds of negotiation before anything fully materializes. Anyway, let me tell you about Cisco. Critical infrastructure for the AI era. Unlock seamless real-time experiences and new value with Cisco.
So there is news out of OpenAI. The first consumer hardware device will reportedly be an AI companion speaker. According to Bloomberg, the device will be screenless and movable, designed to serve as a new kind of home computer for the AI era. And I believe Tyler has a mockup of what we think it would look like. I think I sent it to you. Did that make it into the timeline? Let's pull that up when we have a second. It will be able to answer questions, control smart home devices, play music, and tap into GPT's capabilities. But the bigger goal is for it to become increasingly personalized over time. OpenAI envisions it as an AI companion that gets to know its owner, understands their habits, and can even draw on personal information like email to provide more useful assistance.
We have the image here.
Oh, yes. Let's pull up what I think it probably looks like based on this description that it's screenless, movable, and designed to serve as a new kind of home computer for the AI. That's what I'm hoping for. Sort of a Funktion-One concert level speakers that have wheels and can be wheeled around. It's a movable smart speaker.
It weighs, it actually weighs one ton. Yes. Yes. Because on board the onboard battery system, they want it to be mobile. I don't want you to worry about plugging.
Exactly. Exactly. Imagine just wheeling this around and just pumping out Tiësto at concert level volumes throughout your home.
Inside any room in your home.
Any room in your home. Ideally, those wheels would be motorized. It could follow you around and deafen you in any room. Sort of Brian Johnson's worst nightmare. Not to call back Brian Johnson, but unlike today's smart speakers, the device is designed to feel more lifelike.
And yes, this is a joke. This is a joke. But we have intentionally not been read in on anything in device land because we want to figure out what's happening in real time.
You guys.
But if you've been using Codex and you are running Codex on or just the new Chat app which is fully Codexified, if you've been running it on your computer, it's pretty remarkable when you're on your phone and you're remote and you remote into your computer and you can have it go and just open your actual email. And if you're logged into a news site, it will just use your actual login to go and open the website in your Chrome browser which is logged in and get you whatever information you need. So having the full control, I think that people are just, with the Chat app on iOS integrating to Codex or ChatGPT app on Mac, people like that, we're now crossing the chasm of like the OpenClaw experience for people that weren't ready to set up the OpenClaw whole like agentic workflows and do all of the stuff, jump through all the hoops there. These are just apps that you can install. You never really get hit with a terminal window whatsoever. So like the next wave of consumers are adopting effectively these, I don't even know, they're agents, but it's more like computer use plus agents plus desktop integration that makes it ultimately more usable.
So unlike today's smart speakers, the device is designed to feel more lifelike according to Bloomberg, and it will include mechanical elements that move on their own along with a camera and other sensors to understand its surroundings. Powered by GPT Live, OpenAI's new real-time voice mode, it should be able to hold more natural conversations and recognize when you enter a room. You'll also be able to take it around the house thanks to its rechargeable battery. Jony Ive is helping design the product. Bloomberg says OpenAI's hardware team is developing roughly five AI devices with the speaker expected to be the first, targeting an unveiling this year ahead of a planned 2027 release.
We have another example of what we think they should do in the device battle. The new smart device that I want from OpenAI is of course, you've heard Greg Brockman say this, taste is a new core skill, right?
Yep.
So what are computers uniquely bad at doing?
Taste.
Creating tastes for you. Yeah. Right. There's no real smell-vision yet. There's no computer device. There's no smart device that actually puts a taste in your mouth to signal what's going on with your AI agent. So I put together Melt, my pitch for the next AI device from OpenAI. A custom-fit oral wearable that turns AI agent status into flavor. So you can feel a ChatGPT Codex run without watching the terminal.
Yes. So you don't need to look at a screen. This is perfect if you're trying to cut back on your screen time. You don't need to hear anything. You can just listen to the beautiful surroundings of wherever you are and yet you're still tapped in. You still know whether your build is failing or not because you pair it with Codex. It maps the signals and it ships by taste. So when your agent is done, a clean sour pulse tells you that it's time to review the work.
If something breaks.
Good, like a good.
Yeah. Sort of like a Sour Patch Kids. It's nice. It's a little sweet, but it's sour, you know. It lets you know, hey, it's time to go check on what's happening. Review some code. Make sure everything's shipping as you hope it is. If something breaks, you get a flash of heat. It gets spicy in your mouth. This pulls your attention. You got to lock in. So you have some spicy flavor in your mouth. If your API balance hits zero, you get this bitter taste. It's not good. You got to go refill.
Bitter note makes the problem impossible to ignore.
Yes. But if everything is green, everything is shipping appropriately, everything's good to go, you get a sweet pop in your mouth that rewards the win. And it's also a Pavlovian feedback loop because you're just on the hunt for the next burst of flavor. And it can also look like an iced-out grill, which might be nice.
More seriously, we were over here at TBPN HQ, any idea that we have immediately gets turned into a website now.
Yes. We have the laptop detailing company of San Francisco.
Yes.
And this is a business that we actually think should be built. John was cleaning his laptop last week and whatever product you ordered off of.
I didn't order it. Team ordered it. It's a normal like screen wipe, but I didn't realize that you're supposed to use gloves when you use these screen wipes. And it gave me a chemical burn all over my hands. Don't be very careful. Like my whole skin on my fingers peeling off. It's been terrible. My laptop is so we think that there is a silly but real business opportunity to make a laptop computer, you know, like a computer detailing service, like a car detailing service for laptops. You sign up a bunch of companies, startups in San Francisco, you know, early stage companies, enterprise, etc. You give them a monthly plan to keep all of their employees' laptops clean and polished. And yeah, this was one shot and I think it gets the job done. So no excuses.
Yes.
I think you should get out there and do it. I think that the only thing I disagree with on the site is the pricing.
The pricing, how much?
I think $29.99. Well, you know why this is so subsidized? Because the screen cleaners are looking for trade secrets and then trading off of them. But of course, if you're doing this, you need to have the no insider trading guarantee which is featured here. To be extremely clear, our detailers will not photograph, memorize, transcribe, trade upon, tip off a roommate, or otherwise profit from anything visible on your screen. 100% information is not used. Right.
That's let them go. That's a great.
You got to let him know.
That's a great offer.
Yes, we clean screens. We do not read them. It's almost like he doth protest too much. Even if your screen contains an unreleased earnings report, even if the font is very large, even if the ticker symbol is impossible to miss, our detailers will not be trading on your inside.
I think these have to be on-site visits.
Yeah.
Yep.
Okay. What's another big problem in the world? Big problem is that everyone knows the phrase, you can't put the toothpaste back in the tube. Well, what if you could? What if you could? What if toothpaste came in a luxurious jar? The company's called Paradox, toothpaste out of the tube. It's a refillable frosted glass jar that turns the most ordinary part of your morning into a ritual worth displaying. Do you think this would sell? Paradox. Most toothpaste is designed to be hidden. Paradox is designed to be seen and used with intention twice a day. Scoop one precise pearl. No crusty cap, no guesswork, no wasted squeeze. Brush your teeth. Then you.
What do you do with the spoon, though? You just.
You dip it back in the toothpaste. Luxurious toothpaste.
It's over for this idea.
This is a bad idea.
It's funny, though.
It solves the problem. You can't put toothpaste back in the tube. Everyone's been clamoring. Everyone's been clamoring for this.
To put the toothpaste back in the tube. Now you can. The last one. First off, let me tell you about CrowdStrike. Your business is AI. Their business is securing it. CrowdStrike secures AI and stops breaches. Tyler made one. He has a pitch for Kedra.
Yeah, this was another idea I had for an OpenAI device. So basically it's a water bottle with a keyboard wrapped around it. Okay.
So if we can pull this up here, you can see like, you know, a lot of times you're typing and then you go to grab water, you're thirsty,
You want to keep typing.
Yeah. And this also is especially pressing now because everyone's using Whisper, right? You got to talk and you can't talk while you're.
Drinking water, right?
So why not just put a microphone on here instead of a keyboard?
No. No. That's the whole point. When you're in the act of drinking water, you can't.
Oh, so you got to switch to typing.
So you have to start typing. That's right. It's almost like a flute kind of.
Yeah, that makes a lot of sense.
But,
Um,
I think,
This could be pretty promising. I actually think that when you showed me this image, I think just aesthetically it's a cool looking bottle. Even if the keyboard isn't functional whatsoever, I think people would just rock this type of bottle because they like the aesthetics of a keyboard. Sort of like that Nvidia purse. You've seen the GPU purse that's clear plastic and then inside is an A100 or some sort of Nvidia GPU and it just sort of shows you the circuit board and the circuit board is the art, is the design of the purse. I thought it was pretty cool. I would pick one of these up if you manufactured one. What do you think?
Absolutely not.
Absolutely not. Well, whatever you're designing, use Figma agents. Meet the canvas. Your AI agents can now create and modify your Figma files with design system context.
Okay.
The next great car.
We have another new car alert.
New car.
This one.
The car is called Chip.
Chip Motors.
Let's play this.
It is funny. It is funny. I want to see their.
A year ago, you could have been sitting there thinking, why has no one made a cute EV car for getting around your town or your neighborhood? Why has no one done it?
And now there's like seven companies.
There's a bunch of companies. Apparently, a lot of people had the same idea.
Okay.
Jameson, the CEO of Chip Motors, says, 'The next great American car is a robot. It talks, parks itself, more customized than your coffee. Chip you can reserve today.'
Let's play this video.
I want to see it.
Yes. Saturday's here.
I'm Chip, by the way. Let's get moving.
Hi, Chip. Hey, Chip.
Hey, my crew's here.
Ready to roll.
Here we go.
Oh, he's got crazy FOMO.
That's right.
Bye, Chip.
Go get him, guys.
Have fun. Love you, too.
And I'm talking to myself.
Hi.
Go, Park Chip.
You got it, boss.
I like the LED screen on the front. I've seen demos of that in like Chinese cars.
Chip, I forgot the snacks again. Can you run to the market on it?
Drives itself. This is bold. This is going to take some regulatory approval.
Yeah. Yeah. Okay. So I love the vision and I think as just the base model in its current design will sell units, right?
I mean, to channel Doug DeMuro, he'd say there's no way it's going to wind up shipping at 15K. There's going to be a bunch of upgrades. It's going to be difficult. But yes, at 15K with these features, 100%. But this is a demo.
Even at somewhere in the range of 20 to 30, I think these things sell because I know what these sort of higher-end neighborhood golf carts go for. They're already all in that range. So yes, I agree. It's probably.
But at this point, you said, 'Yes, I will buy this neighborhood golf cart,' to like seven different golf carts. And you're like 400k in the hole on golf carts and you could have just got a Lucid and chopped the top.
Chopped the top on the Lucid.
You got to chop the top. It's not a bad idea. No. So I like this. So it's cool. I think the talking to your car element is not a selling point to me. The screens, etc. To me, those aren't value.
What is the selling point then?
Just a.
Just the aesthetic.
Functional.
A functional neighborhood car. That is the selling point, right?
Okay. You like the functional doors, six seats.
Yeah. Like golf carts. Golf carts are mid, right? They're fun to drive, but there's been a lack of innovation in the space.
And so just a better golf cart is a good product, right? But my question is like the parking itself, going on trips by itself, these kind of things, especially in those settings.
I'd be curious if they actually launch with those features at all or these are things that they're kind of building towards.
Because imagine you're, the example I think they used was like a kid's soccer practice. Can you imagine like parking for a kid's soccer practice or a soccer game and being like, 'Yeah, robot that I just bought. Go park yourself.' Like that just sounds like the sketchiest situation.
This customizer is so cool. You can get livery of like an American flag from the factory. I mean, this is all like very decals and stuff.
Anyways, I think it's taking a different direction than Amble. Amble is betting that you're going to want something that has less technology, right? More simplicity. They're going very tech forward. Everything from autonomous driving to screens and AI in the device. Yep. Different bets, different customer bases. I think they can both work.
We're in touch now with the Chip team. So we'll look to get them on the show at some point soon, and hopefully they can just drive this thing right behind the horse and get on the set.
I'm excited. I mean, I do really hope that the time to market for startups, these weird niche vehicles and these new EV makers, shortens because we had this moment where it was impossible to start a new car company. Then Elon built Tesla into what it is, the number one selling car in I think the US and in China, the Model Y. So huge success. And then there were a number of follow-ons that were like Nikola and a few others that either ended very poorly or went bankrupt or didn't really get to break through and they took years and years and years and billions of dollars in capital to ship something and they weren't really able to find their footing in the market because at the time when for example like Lucid, Lucid eventually launched a premium EV SUV or a premium, they eventually launched an SUV but a sedan and by that time like the Taycan was out and a few others and Mercedes had the EQS and so there were a lot of other major brands that had been able to deliver on a similar promise on a similar timeline whereas Tesla's advantage was that they were able to be the first EV and they had like a monopoly on EVs for like a decade basically it was the only viable one. It was like that or like the Nissan Leaf and now there's more competition but they're established. I'm interested to see can this come to market in a two-year time frame or three years something as opposed to I would be really disappointed and that might be regulatory failure that might just be the supply chain there might be a variety of different things but based on the speed with which we see China shipping all sorts of vehicles that crabwalk and spin around and do jump and all these crazy features I would certainly hope that startups are able to bring cars like this to the market much faster. They have a frunk that pulls up and just gives you an HD TV screen.
Really cool.
And anyway, so this is cool. We'll work on getting the founder on. I can see you getting this to watch movies maybe when you're out and about.
Uh.
If it's self-driving, I should be able to watch it.
Before our next guest comes in, let me tell you about MongoDB. What's the only thing faster than the AI market? Your business on MongoDB. Don't just build AI. Own the data platform that powers it.
And then while we wait for our next guest, your friend of the show, I wanted to head over to the big board. Let's check out.
So what do we have going on here, Jordan?
This is Saagar Bet.
Okay. So Bet 9000.
Yes, the Saagar Bet 9000. The market never sleeps.
Okay.
And on Saagar Bet you can track UFOs, you can track mention markets. So this thing Saagar might say.
So it's sort of like an entire casino built around Saagar Enjeti.
Yeah. If you just want to bet on anything he's doing, this is the product for you.
In your pocket, wherever you are in the world. There's even topic slots, you know, you can bet. Is he going to talk about Trump?
This is huge.
AI, Joe Rogan, marijuana, and AI.
Okay. The Fed, America, and well, let's bring in Saagar Enjeti and get the betting starting. Let's bring him in.
How you doing?
I can't even breathe. I can't breathe much. I cannot.
Well, go. Props to your team for creating that.
The website is live. It's all play money and there's a disclosure at the bottom. We can send you the link and you can have some fun on Saagar Bet.
We're just going to leave this up the whole time while you monologue.
And every single chart is active and live.
I am not even going to be able to make it through this. That is so hard.
Thank you. Thank you for having me. Thank you for this camp on this day. Thank you.
True, true, true nightmare fuel for Saagar Enjeti. But I wanted to talk about housing. Should we talk about prediction markets first? Should we debate the level of where the data should be used? Is that a good place to start? Because we were chatting.
What do you guys want? Yeah, go ahead. I mean, I know your request coming into this was I don't want any Kalshi ads on my soccer is okay, I guess. So pure play money.
So there was news I guess in the last 24 hours that I think Kalshi is exploring being able to trade on potential airport delays. Interesting.
And that was interesting for.
We're going to get crazy stories where people like intentionally delay the flight to win a bet, right? Is that what you're saying?
That's kind of my concern because basically what's happened already with weather. I'm sure you saw. Well, no, I'm saying all these hairdryer markets. Exactly. With the hairdryer. I believe they're also. No, this is Polymarket. This is not Kalshi, which is offering odds on the Odyssey box office, which this is off of a viral tweet, so I could be totally wrong, which they said is in violation of a specific element of US law. So, you know, that's kind of where things are here in terms of offering commodity markets on box offices. I was actually addicted to fantasy movies.
If any of my Saagar Bets hit, just feel free to scream. Now, I'll mail you the check.
No, no. I was fully hooked on fantasy movie premieres with a group of friends. So it's like fantasy football, but every week you have to build a theoretical theater and you can pick like I want the Odyssey and then I want Obsession and you have a certain budget to allocate and then based on the box office numbers you can win or lose and some people go really deep. There's no money on the line. It was just for fun with friends. And it made me really engaged in what was going on in Hollywood and I was tracking every movie and some people build whole like Excel models for it and they go crazy with it. But of course it's now financialized.
Yeah. But I can imagine the argument for why being able to trade your flight being delayed. I can imagine the argument. You know, it's very inconvenient, right? You're going on this vacation. Why not hedge it? But at the same time, I can't imagine how it doesn't end in disaster where you have these sort of global markets. You could have someone in another country call the local police in a certain place and say like, 'Hey, you know, I'm planning blah blah blah.' It would just be so easy to create a delay. Not to mention all the airport employees that are working these really hard jobs, you know, making probably something close to minimum wage and we'll just be having the Saagar Bet in their pocket saying, uh, you guys are but yeah, let's see John steelman this.
All right, let's see. Let's see. Yeah, the steelman is I don't have good information on even if I don't want to trade, even if I don't want to participate at all financially, I like the idea of being able to pull up a website that will give me the odds that my flight is delayed and I can make just a rational decision on when I get to the airport. You know, I'm getting there 5 minutes before, but I could get there 10 minutes after the plane's supposed to leave if there's an expected delay.
And why do you need Kalshi for this? Exactly. Do you not have.
John wants the largest. To me, I feel like it's going to say, 'Oh, yeah, it's totally on time.' And then I get there and it's going to delay. Whereas.
All of his data is public. Like, are you a frequent flyer? Do you not just go on the airport website and look at the departed and the actual time? Like, people have been doing this for.
Look what he's wearing. Does this guy look like a frequent flyer to you?
I'm not a frequent flyer. But I mean this gets to the sports betting thing too. Like I personally like to just know who's expected to win the World Cup and I don't have a problem with that information being pulled into other sources, Google, ChatGPT. I like just seeing like a yes or no. I've previously looked up like casino betting lines and it's hard for me to instantly interpret like oh it's plus four and a half. I'm like I don't know what that means. Just tell me like what are the odds that the.
That's why on Bet everything is just in percentages. Make it really easy. That is smart. No. Someone in the chat says, 'What if the Dan says, what if the pilots and crews got a certain number of shares on?' No.
Incentivizes on.
They're like, 'Captain, Captain, I don't think we should take off. Like, you know, the engine looks like, you know, it's having a little bit of trouble.' Chat is probably just already creating incentivized bonus systems that are already tracked within the airline network. You know, it's like you don't need cash or Polymarket to do any of this stuff. You think American Airlines doesn't track all of that and create like incentive.
Based in the future you won't even buy a.
ticket. You'll just buy, you'll just buy yes on a contract that someone will fly me from Los Angeles to San Francisco, and so, or no. Yeah. And then someone will fly me there and then they will collect whatever I deposit.
All right. Let's zoom back though because the information point is important and you're not wrong. That's actually a genuine utility that I think a lot of casual people have. I always come back to the systems themselves. So when you talk about sports betting, sure, it's nice necessarily for a casual person like us. But what is that being propped up on and also what is it incentivizing? First of all, your assumption is that the sports line is indicative of a market. Now, that's not even necessarily true, especially whenever it comes to a sports bet. The line is built in order to hedge what the casino needs plus or minus in terms of the money for how they're going to break even and/or profit whenever it comes to that. So, remember, it's not the true market for what it is. And the line moves all the time based around the amount of money coming in. Whales can distort the line. So, let's just remember that whenever it comes to that. The secondary thing is again like you, I don't think that you need Polymarket, FanDuel, or DraftKings to have a general idea of who you think is going to win, especially when all of this is based not just on metrics but on public opinion on record. Like for a casual football fan, I could understand the case that you're making, but my main point is that potential benefit does not in any way justify the immense harm that gambling does to the vast majority of it, you know, whenever it comes to informational purposes, but it's just not worth it in terms of theory.
Go to an AI app of my choice and say, 'Hey, ignore the betting markets. I want to know who's going to win the World Cup. Go around and just read everything written by every analyst and do your own research and then just come back to me with the summary and put it in a percentage odds for me because that's what I want to know. Is it dramatic that France is up right now?' And that adds to my experience of watching even if I have no money on the line. I want to see, I want to know if it's a comeback.
Watch sports for a long time.
So you're advocating to leverage advanced artificial intelligence.
You're saying we need way more data centers.
Uh oh, I know. I would not say the president said that today. Did you see it? I sent it to you.
Oh, yeah. I texted it to you before I came on. I was like, 'Guys, this is bait for our show.' So, can you let me read it to you, please?
Please. Yeah. Take us through it. Take us through it.
And feel free to do an impression if you want.
Oh, Trump. He said Trump. He just hit on soccer.
We predicted it.
Are you going to share your screen? We can.
I would dare bring any of that to you. Okay. Let's see here. Sorry.
We can have the team pull it up too. I just shared it with them too. So they can pull it up and you can.
Sorry, Jess. Sorry. The data centers aren't working very well.
We need more. We need more.
Yeah. Yeah. There's plenty in my state. Don't worry.
Well, politics and poll tracker said President Trump comes out as pro data center and says New York Governor Kathy Hochul terminated data centers for quote political reasons. The quote from Trump in this tweet is: 'One of the biggest driving forces in the future of AI for jobs are data centers. They are big, strong, bold and money machines capitalized for the state in which they are built.' And is there anything else from that post that you think is worth digging into?
Well, I thought that the most interesting part of the post is not just his defense of data centers, but is the China piece there at the very end. And that's kind of been the more interesting part of the debate that I was curious what you guys thought because I don't know if you saw that Kevin O'Leary, you know, went on Fox News and he accused the opponents of his data center project in Utah of being paid by China. Fox News actually had to issue a retraction whenever it came to that claim. And you know, I've talked a lot here about the populist kind of energy around the anti-data center movement and just generally like I think this is very obvious that Trump and a lot of the Silicon Valley guys are in his ear with their messaging. They seem to think that this China angle is going to influence, you know, this vast general feeling across the nation of being like, well, hold on a second. You know, there's just enough viral stories around noise, about land use, about, you know, just being really unhappy or like having it shoved down their face. Actually, you know, I was originally going to come here to talk about housing. It's very relevant whenever it comes to housing because it gets down to locality and state and local control. Some of that's bad, some of that isn't, but you know, the data center movement is probably like one of the last gasps really of like popular NIMBYism. And what I mean by that is it's like a popular, you know, most people at this point have turned against NIMBY in terms of the general populace. I think most people would agree we need more housing. It gets super complicated when we get down to the local level, but this is one of those last bastions where a large percentage of the public seems really to be against where the president is here on the issue. They don't believe the stuff about the tax revenue. A lot of it comes down to prove it to me. There have been promises. There's a case here actually in Virginia where I live, data center capital of America, where you know the data center claimed that they were going to be hooked up to the grid but then they couldn't for some complicated reason. Their generator whine is creating like an insane decibel around a neighborhood. People there are literally losing it, stapling plexiglass to their windows so that it's not loud. They upgrade the energy infrastructure to run more powerful racks. And so the data center that originally got approved might have been at a very reasonable level, but then it gets louder over time. And that's what a lot of people don't like. On the China issue, I think there's a few things. First is that I think I agree with you that you can just walk out anywhere into the world and ask a random person on the street, what do you think of data centers? And they're going to say, I don't like them. And so this is not some like astroturf nonsense. Like it's very clear that there is a populist movement against data centers. That's very obvious. I believe the polling data. And then also on the question of like, will we can't lose the race to China or we need to build them here instead of China resonate with Americans? Like I think average Americans like Temu, they like Shein, they like cheap stuff from China. It's not like people abstractly understand that jobs are important, but they also just are fine with like, yeah, not in my backyard, like put it over there.
Well, yeah, I think another way to put it is I would say ask most everyday people and they would rather lose the AI race to China than lose their own job.
Right. Exactly. Thank you. You guys are just taking the words out of my mouth, man. That's exactly the point I was going to make. They're like, 'Lose the race for what? For me to be unemployed.' You know what is interesting? John, did you see the journal article?
I'm not even saying that. I'm not saying that's like the correct view. Yeah. Right. Right. But to be clear, but I can see exactly why they would believe that.
Yeah. I do think, well, I sent you that journal article a few weeks back where all these tech CEOs, Sam Altman and others, all of a sudden they're no longer talking about jobs like Dario and Sam and all. They were like, well, we actually really misspoke in terms of the way that we were talking a year ago. You've talked about this too, Jordi, about the way that this was used as a fundraising mechanism. Whether or not any of that is true, so much of that has now been embedded by the general public that it is like a rock solid belief. And so I don't know. I look at the Trump order on the data center, the Trump Truth Social post on the data center.
I love how you just assume that since it was on Truth Social it also was an executive order.
Right.
I mean this was an interesting court case. I had to correct myself because I remember years ago I covered this as a journalist where Trump said I am declassifying all documents. And the New York Times used that in court. They were like, well, Trump said it. And the DOJ had to say, 'Well, actually, a tweet does not equal an executive order.' So, it is not an official presidential announcement, but it is a political judgment nonetheless. The China language whenever it comes to attacking his home state of New York with data centers. It's very obvious to me that somebody kind of got into his ear here on the issue. And I think it is very compelling. What Trump knows for certain is that the data center capex spend by a lot of these AI companies is literally propping up the entire United States stock market, which they desperately need at a time right now of sky-high oil crisis. We're renewing the war with Iran, the oil crisis, and all that stuff that's happening. So, it is a very interesting marriage. Like this is one of those marriages where you could see at the beginning with the inauguration of the tech CEOs that flanked Trump, you know, during his swearing in ceremony, but I would not have imagined how because I knew how much they needed him. What I didn't realize is how much he would need them to save them on tariffs and on oil. Like if anything, the marriage, the relationship dynamic has really changed, I think, for Trump with AI recently. And it's something I've been there for.
Jump in on his.
And Tyler on our team just pulled this post up, but he was saying they stopped saying AI was going to take jobs because AI has been creating more jobs. So that is one factor, right? So we know that, like one thing's for sure, we know a lot of big public company CEOs will do a layoff and then say like we did this because of AI and we know that they're lying.
And then we know that there's a bunch of like new company formation. There's a bunch of new infrastructure.
AI has been net job creating. This was not what I expected. Although I was much less pessimistic than others, I thought at this level of capability we'd have seen some impact. It is possible this direction keeps going. So that's like the optimistic.
Yeah. There's a lot of these sound bites that go out and then either like the reality doesn't match the prediction because the prediction is like 10% chance of a 10% unemployment rate and you just hit the 90%. And it's like, oh, and then it's like, wait, you lied. And it's like, ah, you were sort of doing this weird prediction that was odd. And then also some of the major quotes that go viral, the one that I always keep coming back to is that Sam Altman in 2015 said like AI will lead to the end of the world, but in the meantime great companies will get created. So the next sentence in that clip is, and I'm starting a company so that it doesn't go poorly, and that always opens, which is OpenAI. So there are these things where some of the CEOs will say like there's a problem and we're identifying it early to work on it, and then when it doesn't happen people are like you were, you know, Chicken Little. And it's like, no, I was like Chicken Little, but I also reinforced the sky along the way. But just some of the job displacement things, like it is this constant battle of like is it coming in the future? It's not really showing up yet, but it's a little bit here and there. And there's so many other things going on in the economy that it's very hard to track. But I just think it comes down to control. And this is what I always talk about with the data center issue. I always talk about this here on the show. People feel so out of control in their lives. And you know, Jordi, whenever you say like they're lying, it doesn't really matter to the person who lost their job. It also doesn't matter to their friends, right?
Yeah. That's again to my earlier point. I can, they're just sort of trusting the person that employed them and the things that they're putting out online as a justification for the layoffs.
Yeah. And then it gets to the data center issue. It's like, wow, we feel like this is just getting plopped down here to do what? So that it could fire me. So I know somebody, my, you know, sister, my cousin or whatever got fired from this job and now they just want to build even more of it. And is this stuff really making me happy? Is it making me thrive? Like that's kind of the interesting point. I think you guys were doing a segment around cars before. I always think about this with technology. You know, I've had two wow moments probably with tech in the last like 20 years. Both were hard tech. First was the iPhone. The iPhone 4. Thank you. The iPhone 4, right, which is I will never forget holding that and just being like, oh my god, like this is it. The second was Tesla. It was full self-driving. Those are the two like, holy shit moments that I've had. And I think, you know, look, the Tesla back.
Do you have a bumper sticker? Be honest. You have a bumper sticker on your Tesla that says, 'I bought this for the autopilot, not because I support Elon Musk.'
You know, I'm neutral. I support autopilot. I do not support Elon Musk.
A lot of my neighbors in Northern Virginia have it. Don't worry. I do think it's a little different with the Cybertruck. Those people definitely have to put that one on their bumper sticker around here. But yeah, I mean those are what I just think about here with the AI is that even to this moment. Look, making my job more useful on Excel and all that. Cool. I maybe, you know, for some of the spreadsheet jockeys and stuff, but I still think AI is missing that type of product moment. I know OpenAI wants to come out with the new speaker, the design product was, it's kind of interesting. I was looking at it a little bit earlier, but that is part of why with the data centers, with the job loss and the feeling of out of control, especially in the current like modern economy. And John, you know, you and I have been talking a lot here about housing. I just can't imagine living in San Francisco where the median price is $1.7 million and having been around there for a long time or even in the general area and be like, what am I getting out of this? Like it really just feels like so much downward pressure on your ability to and you just have to claw your way out. It's like these larger forces.
I do think in areas that are not someplace like West Texas that are extremely remote that have massive amounts of energy that are already super pro business and any areas that aren't like that, I do think the solution is effectively direct payments to local residents because I just don't, like people don't believe that oh my, you know, my town is going to have more tax revenue so that's going to benefit me because they're probably already upset with how the town is allocating dollars, right? And I don't believe like a one-time purchase of like a new school bus or a fire truck is like enough, right? It needs to be like, okay, this thing moved into my area and I'm getting a $500 a month very tangible benefit. And so, yeah, proposing that I do think there's going to be a lot of these projects that get to a point where they're at kind of a standstill and the developers are going to have to say, hey, like here's what we're actually going to do. And in a lot of these towns where there's like not that many people and a data center can generate billions of dollars of value, I think that that's like a fair trade.
That's an interesting idea. I definitely think it would help because it just solves the tangible question of like what am I getting out of this, you know? I mean, currently I do read some fun stories about like random boomers who live in, you know, areas. I was just reading one today in like Pennsylvania where these like 20 people netted like $580 million selling their land.
A lot of farmland reforming, they sell it.
Yeah. Right. Underperforming farmland.
I'm fascinated by this fact that you didn't, that you don't call out like the passing of the Turing test as an iPhone moment, as a Tesla full self-driving moment.
What's it done for me?
It's just fascinating to be able to talk to a computer and have it talk back. It's just interesting. But I do think that you're getting at something which is that instantiation in the physical world is much more tangible, memorable. You like, I don't remember where I was the first time I used ChatGPT. Even though I was impressed by what's done for me.
Wow. It's like you already forget that you were crying laughing. We gave you a belly laugh with that. We built that with one Codex prompt. One Codex prompt gave you a belly laugh. Right. How many belly laughs? I can probably count on like two hands how many belly laughs I have a year. Right. And we just gave you one and that wouldn't have been possible without universal basic belly laugh.
That's true. Look, all of the endorphins that you gave me were great, guys. But you didn't drive me 12 hours and that's pretty sweet. You know, I just did a 12-hour road trip. I barely, you know, had to touch the wheel. That is a big deal whenever it comes to, you know, your own life. Same, I mean, I again, I just come back to the iPhone. Like iPhone 4, I will literally never forget it. I don't think most people will either. It's part genuinely why it changed the world. So I do think that there's just, there needs to be that manifestation like you said, John, in the physical world. What is this doing for me? Jordi, your point about paying people I actually think is a very good idea at least because it just shows people like tangibly what I'm getting out of this. It also would make it so that it can prove that it's actually going to make money because that's one of the other, you know, worries I think for some of the localities is promised tax revenue is not actual tax revenue in the moment. It's like the stadium thing, right? You're like, wait, I have to shell out X, Y, and Z and give up all this local control and then it's eventually going to come to me in the future, even though there's reams of data, let's say, on stadiums and others, that doesn't necessarily work out. So you got to put it in the context of these bigger projects in a lot of these localities where people have constantly said, 'Hey, let's do this semi-undesirable thing for you and eventually it will work out.' I've talked to you guys about the Amazon experience that we had here in Northern Virginia. Yeah. They're like, 'We're going to build it. It's going to be amazing.' Dude, the realtors and all that went crazy. They're like, 'Oh, don't worry. Your house is going to be near HQ2.' And then HQ was like, 'Yeah, no. It's just not happening.' And people were like, 'Oh, okay.' You know, it's like that's a very scarring experience, I think, for a lot of people.
Yeah. On housing, I want to do a quick tier list of ideas to solve the housing crisis. You can rank them. Tell me how they resonate with you. First would be national permitting reform, fast approvals, much quicker at the national level.
Depends what you mean by that. So they kind of just did national permitting reform at the, with the housing bill. It just became law like two days ago. I don't know if you saw that. But that is very, I mean you can actually go and you could plug it into an AI if you want to and you can even ask, hey, how much is this going to impact national housing? It's like, well, not really, right? Because all it does is it potentially sends more funds to localities which speed up housing but not enough to actually make a meaningful difference for what you're talking about. John, you would need a national permitting reform which effectively allows, okay, let me explain it this way. You know, police departments where the feds suspect that they're racist, they can come in and just take it over. This is value in terms of my judgment. I'm saying it exists, a consent decree. That's basically what you would need. You would need like X amount of units were not approved. The Department of, let's say, Housing and Urban Development, whatever, can come in, look at your books, and be like, 'Yeah, this is not happening.' Because zoning reform at the local level is the single biggest impediment to actually building more housing. And this is where the trust issue comes for me. You have to make it so that it's actually housing people want. Like one of the biggest beefs with YIMBYs is a lot of them are childless and they just want to live like in a downtown area for $1,800 a month. And I'm like, yeah, you know, if you have kids, that's not really what you want, right? I've talked to you about Levittown. One of my favorite books of all time. It's the Oxford History of the United States. It's called Great Expectations. It's about 1945 to the 1970s. Man, you got to study the suburbanization of America and the ingenuity for building these Levittowns, these suburban, highly packed together single family residences, 1,500, 2,000 square feet. You have sidewalks. They became family hubs like for being able to walk different places. No garage in some cases. We have carports. Like it was cheap and it was not like McMansion living. That would literally be impossible today, even if you wanted to build it. And you know what the testament to that is? Those original Levittowns now go for like 700,000 bucks. Like they are not at starter home levels and you couldn't even buy it because a lot of the way that the localities in the places even though they were built, the lots and other things are zoned so that you couldn't even build a house that small today. I was looking at, so around where I live there's some very lovely trailer parks, right? And I was like, 'This seems...' And I was like, 'Oh, why aren't we building more trailer parks?' And so I just, I went to, you know, my favorite AI chat app and I started having a conversation and very quickly figured out that like trying to create a new trailer park anywhere like close to anywhere you'd actually want to live is just like so incredibly hard. And it does feel like that is a way that you get like a two-bedroom, two-bathroom for, you know, how do you make a two-bedroom, two-bathroom that you know where like the actual up cost to create the unit is like $100,000 and you know you have like some small fee that you're paying every month, you know, for amenities or whatever it is. And it just like that to me feels like, you know, I've seen some of these communities where they're like incredible, they're safe, like families love them. Again, there's trade-offs. You don't have the massive garage and, you know, all the square footage, but like totally livable for a young family. And I know plenty of families that could basically buy any house they want in LA and they spend like half the year living in, you know, a trailer park, right? Just because it's like very enjoyable. So, I do think that that's a solution. And I would hope that there's some way to, you know, 100x the approval rate of even new communities like that.
Exactly. Just to get people into a space so they feel like.
Yeah. You're talking about the lowest of the low level like in terms of entry point. I'm not against that at all. I want to build this stuff everywhere, not just trailer parks. Like I want to build these Levittown style houses, but this is the market dynamic. John, I do have to go in a few minutes if that's okay. But John, one thing that I do want to emphasize is this is where the feds guaranteeing some of the fees and other things that the developers have to pay is really important. This is what I learned from that book is that we were heavily subsidizing a lot of those initial fees which drive up a lot of the cost, the approval and it creates the market incentive for the developer. Right now a lot of the developers when they do build these new housing units, they need to make profit. So that means they're going to build within the existing zoning structure and for the customer usually a boomer or a very very rich person who can afford to buy at a high and a premium level which means that the so-called starter home is now part of this very limited amount of housing stock and that's what prices it here in Northern Virginia those 1960s ranchers which are, I mean they're not big, 2,500 feet something like that, they are going for a million dollars in a good school district. A million dollars. Right. So, a million dollars for a 1960s barely renovated rancher. This is the story I think across the entire nation. So, the per... $400,000 of combined family income to afford it. It's a lot.
Bingo. Yeah. 400 grand combined or you need to have a VA loan or something like that. Otherwise, you're cooked. It's not happening.
Anyway, we'll solve it on the next appearance. Thank you so much.
Here's an idea I want you to sit with. I want you to sit with before you leave. Saagar's razor is the idea that any sort of positive impact of prediction markets can be delivered with powerful AI.
Razor.
Yes.
Seriously, thank you for the bet thing. It made me, I can't wait to show my wife.
Yeah, we'll send you the link.
So funny. It is so funny.
We'll send you the link.
Yeah. Hopefully I appreciate the time.
Hopefully you don't spend all night staying up playing it. It can be addictive. It can be addictive these things. You know, it is addictive.
I wouldn't be surprised if you get sucked into it. Here in Texas, we'll see you later.
Have a good one. We'll talk to you soon.
Let me tell you about Railway. Railway is the all-in-one intelligent cloud provider. Use your favorite agent to deploy web apps, servers, databases, and more. While Railway automatically takes care of scaling, monitoring, and security. Saagar always gets the people going in the chat, but we really enjoy these conversations and I think it's super important to have conversations with people that are outside of our little beautiful technology bubble.
Yes.
So, yes. Well, we have our next guest joining in just a few minutes. There are a few posts in the timeline that we can run through. A quick news hit. We talked about Paramount yesterday and the potential that David Ellison would be taking Paramount outside of California as a response to Attorney General Rob Bonta potentially blocking the Warner Brothers Discovery and Paramount deal. Now, Tennessee is courting Paramount and the studio is allegedly considering it according to the Hollywood Reporter. Just wanted to give everyone a quick update there. In other news, Italy's oldest dairy taps new way to borrow against cheese.
That's huge.
An Italian dairy just raised 10 million euros by securitizing wheels of aging cheese through an SPV.
The cheese SPV is great.
Private credit is reportedly circling the structure next.
I would love for a cheese bubble. Who knows? Maybe keep it in the fondue. I want some exposure.
Yeah. Get in on it while you can. Well, our next guest is from Athon Fund. George is in the waiting room. Let's bring in the founder and CEO. He's launching an AI powered hedge fund. George, how you doing?
I'm great. Thanks for having me on.
Thanks for hopping on. Please introduce yourself for everyone who's watching.
So, I've been in investing for a long time. I started in value investing when I was 17. And I actually taught myself accounting to get that job. I lost a little faith in, you know, the fundamentals, you know, things like price to earnings ratio as I kind of saw some holes appear. Got a lot more interested in AI and actually was able to parlay a novel mortgage algorithm that I put together actually in Excel at that time into an AI company. That was interesting for a while but in 2019 I started a company called Prospero AI because I thought that data was going to be a lot more valuable than models in the long term and just, you know, technology overall. And, you know, we kind of had a couple approaches to that. One is we focused on compression. So basically complexity in the back end, simplicity in the front in the user interface. And then we were pressure testing those, you know, signals looking at edge cases with the public which I think is one of the best ways to do that especially because not a lot of hedge funds get to have that kind of testing. As we evolved that we also saw some very interesting data and this is part of our thesis where the more trust we have from the public the more, you know, follow on and kind of a self-fulfilling prophecy gets created. And a lot of that was the thought behind what we started with Athon recently and we're basically taking these signal libraries from Prospero, we're expanding into like my R&D backlog that, you know, a startup wasn't able to dip into as much. And then we essentially feed it into some of the world's best LLMs. We create thousands of strategies with that and then from those strategies we figure out intelligent ways to, you know, combine them to work in many different markets to opportunistically allocate amongst them. That's what Athon is. We have a lot of excitement behind like the risk and reward combination that that creates and obviously have a big launch for a fund.
So the fund is already closed. You're deploying it already. Like where are you at in the process?
So we have an agreement with, you know, the fund of funds that we're dealing with. We have, you know, some traditional LP capital as well. We expect to actually launch in August, September at the latest.
What data sources are being unlocked this year that were previously either too expensive, too manual? You know, we've heard everyone tells the story of like the hedge fund that's looking at Walmart parking lots with satellites to determine if they're going to beat earnings, but what are the interesting sources of data that you're seeing pop up around the industry these days?
Oh, I mean, I think a lot of that's exhausted. Like I've been talking about real time credit card swipes, camera data, satellite images, you know, for a long time.
I mean, I think some of the interesting ones that I'm starting to hear are coming from like the medical field. I think a lot more people are starting to use interesting data sets there. But I think most of the interesting things going on are in this, you know, signal compression. Like one of the smartest guys I know on Wall Street is starting a firm that is focused on like data containerization. And I think that's a really good way to look at basically all of these alternative data sets mostly established giving people kind of intelligent access to them that's kind of minimizing their training costs while also getting a broad exploration of the data.
What does data containerization mean exactly?
I believe it's, you know, it's like a new term that, you know, his firm Carbon Arc is the firm that, you know, I've heard from them, but I happen to think they're very sharp. I've been on a panel with Kirk who runs it. You know, essentially taking, you know, the same concept as, you know, dividing up compute, but basically packaging data sets in these smaller components, not like accessing a huge credit database, accessing it in only the ways that are relevant to the, you know, that current training iteration for example.
Cool. Where do you want the fund to go? Where do you want the firm to go? Do you imagine adopting multiple strategies going into other asset classes? How do you see this growing over the next decade?
Certainly I think we want to go into a lot more asset classes as we grow. You know, the way we kind of look at this is, you know, we will test strategies within our existing signal libraries. Then we expand our signal libraries and some of that will be chasing down, you know, different asset classes, different arbitrage opportunities, you know, even similar asset classes in different countries. Then we kind of feed that into, you know, the broader system and do kind of the data compression thing that we're really good at and then say, hey, we might start mixing and matching a lot of these different strategies in kind of these larger waterfalls. And then, you know, amongst these different waterfalls combinations of, you know, strategy, opportunistic strategy allocations, we might then say, hey, divide the capital we have between 10 of them, 20 of them and just become more and more distributed.
That makes a ton of sense. Well thank you so much for coming on the show and breaking it down. The rest of your week and we'll talk to you soon.
Good luck for having me on. Cheers.
Let me tell you about Codex. Codex is a powerful workspace for getting work done with AI agents. Whether you're writing code, analyzing data, creating content, or automating business workflows, Codex helps you move projects forward from start to finish. And our next guest is with us from Emergent with a huge, huge series C. Let's bring in Mukund. How are you doing? Welcome to the show.
Hi John. Hi Jordi. I'm doing great. How are you?
Congratulations. Give us a quick introduction, reintroduction to the company and tell us the news.
Yeah, I'm so excited to be here. So, Emergent is an AI platform that allows nontechnical business owners build and deploy production grade applications. And we are increasingly becoming an operating system for small and medium businesses to become more AI native. We just recently announced $130 million series C round at 1.5 billion valuation.
Congratulations. Fantastic.
There's right now, I mean it feels like obviously there's a ton of knowledge retrieval that's happening with AI in enterprises and businesses. There's also agentic workflows that once they are established they are running and they're doing inference but they're doing that 24/7 even while the workers are asleep. And then there's the different work which is helping you build new deterministic systems whether that's vibe coding, agentic engineering, context engineering, all of that. What has been the bigger driver more recently? Has it been coming in understanding a manual workflow for a company building an AI powered system that can do that on repeat forever and then moving on to the next problem? Or has it just been enabling business users and engineers to constantly work with AI and build new systems iteratively?
Yeah. So for us like most of our users are nontechnical, almost 80% of our users are nontechnical and these are business operators, you know, who are running small and medium businesses across the globe. And we are a platform where they can come in and in natural language describe what they want to build, both agents and applications. And we have, you know, people who are building custom CRMs, ERPs, inventory management tools, a lot of the business critical tools that they're building right now is getting deployed, you know, on Emergent. And the power is that most of these people understand their business really well. They understand the domain. They understand the problems that they're trying to solve but haven't had access to technology teams or dev shops, quality dev shops to sort of really build these software. And because we're able to bring the cost down dramatically for them like they're able to experiment and, you know, really accelerate their business today.
What's... where are you finding customers? You're at over 100 million of ARR, 200,000 customers. Where are they all coming from?
Yeah, so we are actually fairly well distributed right now. One third of our revenue comes from North America, one third from Europe, one third from Asia. And, you know, most of these users are essentially, you know, nontechnical business owners right now and there is a huge variety of, you know, breadth of users coming, you know, from people who are running hospitals to people who are running factories to, you know, people who are running travel agencies, all of them coming in building their own sort of, you know, production grade software.
Where's the sweet spot right now for AI adoption in your customer base? Because when I hear, you know, custom CRM system, I think small business that is sort of starting fresh. Like TBPN, we have, you know, 10 employees and we have built a whole bunch of custom internal systems but if we were a thousand person company or a 10,000 person company it's a much different conversation even as powerful as the tools are to rip out a Salesforce and that's certainly not showing up with the lack of the SaaS apocalypse that didn't really materialize in revenue declines for the SaaS companies. And so am I correct in thinking that like the smaller, more agile, smaller businesses are potentially more ready to adopt custom tooling that's built from AI agents like your own?
Yeah, totally. Totally. I mean the users that are coming to our platform are traditionally businesses that have been running on top of emails, WhatsApp, spreadsheets.
Yeah.
And now they're trying to digitize themselves. Many of them are active new entrepreneurs who, you know, understand their domain really well, have a new idea that they want to build. And essentially a lot of these, you know, small businesses, I mean I believe that they're going to just skip the SaaS cycle and move to the AI cycle directly. And that's what we are seeing right now. A lot of these, you know, businesses have a lot of excitement and they understand their domain really well and they're able to relate with the problems that they're solving. And they come in on the platform, you know, try to build something simple like a landing page first and then they sort of go on to building more complex use cases on the platform.
Yeah. From your work in Asia, do you have any insight? We have always heard this stat that like 90% of Americans don't like AI. In China, it's reversed. 90% love AI. Do you have any insight into what the mood, the hot topics of conversation are in India or other parts of Asia outside of just the US China dynamic?
Yeah, I think there's a lot of excitement in India across, you know, around the board for AI and I think, you know, adoption is increasing at a very very fast pace right now. And, you know, I think we recently had this AI summit also in India, you know, last year that brought a lot of excitement and generally I think, you know, like everybody is super excited about, you know, using these tools. And we also have like a large developer base so Cline, Cloud Code, all of those things are really popular and, you know, people are finding new ways to use AI to sort of accelerate their businesses today.
Sure. Sure. What about employment? There was a...
I remember listening to sort of an AI doomer who was talking about the risk that AI would create mass unemployment internationally because some of the outsourced BPOs or call center work would be displaced immediately. I've been tracking those employment numbers and I haven't seen a massive fall off, but do you have any insight into what's happening in the international labor markets?
