Le Chonk, Buffett Discovers YouTube, The Case Against AI Job Doom | Greg Fleming, TJ Parker, Joowon Kim, Rick Dungey, Dylan Field, Hardik Kabaria

You're watching TVPN.

Today is Tuesday, October 6th, 2026. We are live from the TVPN UltraDome, the temple of technology, the fortress of finance, the capital of capital. Let me tell you about ramp.com. Time is money. Save both. Easy to use corporate cards, bill pay, accounting, and a whole lot more all in one place. Also, Christmas. How many days until Christmas? That's the question you've been asking.

That's the question on everyone's mind.

We have the answer today on the show. I believe it's 79 days until Christmas. Oh, Merry Christmas! Is this new? 79 days till Christmas!

There we go.

They're ready. There we go.

Thank you to the graphics team.

And the great thing about this whole bit—

Yeah.

Is it will never get old for us.

It really doesn't.

It's probably getting old for some of you.

No.

And we're sorry, but we just can't help but feel the Christmas spirit.

Also, You asked.

You asked.

You asked for more Christmas.

We delivered.

You asked for more Christmas content. We delivered. Today on the show, we have Rick Dungey from the National Christmas Tree Association. Let's give it up for Rick. He's the executive director, and he's gonna break down the real Christmas tree industry, the official White House Christmas tree, and Trees for Troops, all the different things that they're doing over at the National Christmas Tree Association.

And how he's thinking about the road to Christmas broadly. And we're also gonna ask, Christmas really is his Super Bowl.

Yeah, it is. No, great show. Great show for you there, folks. I was leafing through GQ this morning. Tom Cruise is on the COVID It's a good edition. There's a funny anecdote in here that I wanted to share. I got to find it. But they got the whole Tom Cruise spread, lots of photos. But they did a little back-and-forth interview with him.

We recognize that we are having some audio difficulties.

Brutal.

Team is working on fixing it.

Let's test it with some soundboard. Did that work?

Did that fix it?

What about this? Well, we'll talk through this and then we'll move on to other stories and hopefully the audio catches up by then. He's a course bro. He dropped a course, a private course, and we gotta get access to this. The question from the GQ interviewer, you have mentored a lot of younger actors. Ben Powell has talked about watching a 6-hour instructional video about screen acting that you made, right? And Tom Cruise says, it was very basic. It was just something to have people understand, just to have a different concept of lenses and what the lens is and what the frame is, just the same way as an actor—

Being a global superstar for dummies.

Yeah, basically. He made a 6-hour— that's a course. He made a 6-hour video, just the same way as an actor on stage or in the development of cinema, where it started and what that frame means and how it—

Great success, AudioSynthetics.

And how to understand that frame.

Woo!

Thank you, team.

There's just different layers of storytelling, of structure, but also to understand certain things so that you can forget about it. What else do you tell younger people, other people in your business? I listen first. What do they want? It's not me. I'm more interested in what are their goals? What do they want? And it's not just younger people. This is me when I was a kid. I'm going, what are you interested in? They say, really? I said, yes. What is this movie about? What is this? There's a bunch of other fun things in here, but I thought it was funny that there's a secret 6-hour training video on how to be an actor from Tom Cruise. We gotta get access to it. I feel like there's a lot of journalists that they get documents, they scoop things, they release things, but I've never seen anyone just be like, I'm putting down 10 racks for whoever can bring me the 6-hour Tom Cruise video, you know?

Yeah.

I'm putting it behind a paywall. I'm gonna monetize it, but there's a bounty out. Like, Kevin Roose has the Dario Amodei big blob of compute thesis from 2017, like this leaked PDF, sort of on the scaling law and the bitter lesson, basically. And I've never seen a journalist just be like, I know this document's out there. Like, Eric Newcomer should just be like, Bring me Founders Fund returns. I'll give you $5,000 if you got them, you know? Because it probably maths out at some point.

$5 million.

Maybe more, maybe more, yeah.

To risk your allocation in the next—

Yeah, I suppose that that would violate journalistic ethics to just put a bounty out for a document that you know exists.

I will bribe you for the truth.

And it probably really— like, I was thinking about the The various leaks that go out, like some of them violate NDAs, but if they're secretive enough and they're squirreled away via Signal, like they probably don't trace back, they probably don't really trigger anything. Also, some of them just don't really matter. They will, you know, a company might not be that frustrated about a particular leak if it makes the company look good.

But—

Well, speaking of movies, the team saw a movie yesterday. Ben, could we get a review? Oh, you didn't go.

Ben didn't.

Wow. Didn't come.

Yeah. Wow. Tyler, review.

I saw it.

Who saw it?

So we saw it bigger.

Nick said that he was starting to look at his phone after like 40 minutes.

Wait, what?

Nick did not enjoy it. I thought it was pretty good. And so basically the reviews are really bad, right?

Yeah.

And famously, well, not famously, but like the box office is really bad.

Not all the reviews are bad. I saw Marc Andreessen gave it a positive review.

He did. Yeah. And that was really one of the only like very positive reviews I saw. But yeah, I think the budget was like $160 million. First weekend, it was like eight million.

Eight million, yeah.

Pretty brutal. I thought it was it was good. It was very cool. Like the like the framing of the movie. It's it's it's hard to like reveal too much.

I thought it was very cool.

Film school? Like this this critique is so nuanced. It's remarkable.

Thought it was.

I don't want to spoil anything.

I don't want to spoil anything.

What was cool about it? The costumes?

The music? Give us like a— just give us like 30 seconds of spoilers. I'll bleep it out.

Okay. So basically—

The bleep working.

We're still working on that.

Let me tell you about the New York Stock Exchange. Want to change the world? Raise capital at the New York Stock Exchange. Just do it. Wait. So can he talk over the bleep basically and you can hear both?

Okay.

We're working on a new feature.

The game decompilates The compilation thing is going strong. Ben Thompson was writing about it in Sirtecri today. There were some more knock-on effects. He took it in a very interesting, like, legal direction about the implications of taking a video game, decompiling it, doing whatever you want with it. Obviously, IP licensors don't like that, but there's not a lot that you can— that you can do to control that. And obviously, open source models are coming. And there's a bunch of interesting knock-on effects. I saw a really funny one. They're getting funnier, these mashups. I saw one where someone was playing NHL. Hockey in Resident Evil. Did you see this one? It was really, it was really such a good mashup because the guy's just like on ice skates and like flicking pucks at zombies. It was, it was pretty good. I was, I was into it. Again, probably not really something I want to play for 5 hours, but will I watch a 30-second video of it? Absolutely. But yeah, Polygon has been breaking this down. If you've been on X over the last few days, you probably noticed that things have gotten very weird when it comes to modded video games. There are clips of Call of Duty operators doing kick flips and other tricks from Skate, people playing Black Ops 1 Zombie Mode and Skyrim, Miles Morales swinging through Gotham. Here's another to add to the list. Someone has gotten the Xbox version of Halo: Combat Evolved running in a web browser with multiplayer and split-screen support. And the other funny thing is that it's not— they didn't— apparently the Halo community doesn't like the PC release of Halo, so they ported the Xbox version to the cloud browser. It's like you can really do anything you want now. And so, Ben Thompson goes on to describe a little bit of what's going on. Games are written in human-readable languages, C, C++. Obviously, there's scripting languages on top of that, but then they're compiled into machine code that's executable by the processor that they're intended for. For Halo, that's x86. You've always been able to disassemble machine code into assembly because assembly maps almost perfectly to machine code. The difficult part which takes humans years is decompiling assembly to C or some human-readable readable language that then you can actually modify. But LLMs are really good at that. And so is it legal? Is it legal? The story conflates 2 different AI capabilities. The Fallout project appears to be a new game. There's no Fallout New York, although using the name Fallout is likely a trademark violation. I think we were joking about the lawyers being really excited seeing all these going out because it means full employment. Decompiling is not in and of itself illegal. Sega versus Accolade held that disassembling code is fair use. If it is the only way to reach unprotected functional elements and there was a legitimate purpose. And Sony v. Connectix extended the same reasoning to reverse engineering the PlayStation's BIOS. However, that is likely where protections end. Code that is derived from decompiling is by definition very similar, if not identical, to the original code which is covered by copyright. So even if you go and decompile a game, and then you ship something that's like a mod, you're going to get a DMCA notice. And this is already happening. Like, something will go up, it'll get— it'll go viral and then they'll get hit with a takedown notice. This already happened. 2 hours later. Exactly. Is that me or is that—

Getting hit with a copyright.

2 hours later, you'll get hit by— yeah, copyright. So you have to take it down. There's a couple of interesting things. So the actual value of these, like, decompiled games mashups Pretty limited because you basically just get a viral video and you can't really monetize it 'cause you get the takedown notice. But the training rollout for the AI labs, like the training data that comes from like having this fleet of like all these different game modders go and spend a bunch of tokens that they're basically paying for, maybe it's a little subsidized, but like they're basically paying for all this decompilation. And then also the labs probably have RL environments for this type of work. And then there's tools for this as well. means that I would expect like the next version of models that is trained on all the inference that's happening right now, because everyone right now in the game modding community is like, let me go point GPT-Astra and Opus 5.5 at whatever game binaries I can get. So it's like, oh, I like Skyrim, I like Minecraft, let's go decompile those. But then they're also doing it with all the long tail. All of that's going into the next set of training data. So you would imagine that the models get much better at video game development, which is cool.

Yeah.

He also just talks about like how things will change over time similar to— just similar to what's happened in video. AI seems to exacerbate a lot of trends including the shift to service-based games. So he talks about like in a world where if you send someone a disk or you send someone an executable binary, like they're just going to decompile it and be able to do whatever they want with it. Maybe you want to stream that game. So you— this was— Google had an effort for this. I forget what that was called. It never really took off. But Google had the ability to basically stream you a 1080p—

Google Stadia.

Stadia, that's right. So they would stream you the video output and you would stream them the controller input and the game would run on the server. So you never saw any code. You just saw video coming back.

This is also how the Meta Quest Xbox Edition worked.

Oh, yeah.

You were streaming the game in.

That's right. That's right. And I think Xbox has stayed with the cloud gaming a little bit longer. But gamers have always sort of pushed back on that. I don't like the latency. What if my Wi-Fi goes out? I'm just trying to play a single-player game, blah, blah, blah. And so it's possible that this is something where the game industry comes together and they're like, we really got to fight these hackers. We got to hide our code more behind more of a firewall, like that cloud streaming firewall. Or it could be something that they just like embrace and go the IP protection route. But yeah, to the extent it dooms AAA standalone games, that is only because it accelerates what was already happening. There's no reason to be sad, however. The real intrigue about AI is not the fact that it can recreate what was already made, is that it can create entirely new worlds and do so for the individual. The real analogy might be video. but not Netflix, rather TikTok and Reels, and the fact that everyone gets to choose their own adventure. So you get to make your own video games in the future. And this is definitely happening. There's— whoa, they even actually even took down the video of this one, GTA 5 in the browser for free. Bbomb posted this, playgta5.com, got 8,600 likes. And I have a screenshot of it running, but when we pull it up live, that's the actual tweet today, it says media not displayed. Not only did they take down the website, but they took down the images and the videos that were shared on X. That's pretty crazy.

Hardcore.

Uh, Tenebrous gives some more context here. Uh, this got taken down within hours, but it's starting now. Within a few months, instead of sketchy sites with popup ads hosting torrents of the latest AAA games a few days post-release, we're going to have sites hosting fully playable free WASM. I forget what WebAssembly? WebAssembly versions, that's right, of AAA games, whole libraries of them. They're just a click, zero download required, zero risk of viruses, zero effort or thought. I would call this an existential threat to the gaming industry if it weren't for the fact that every input to previously $100 million to $200 million AAA games is effectively dropping by orders of magnitude in cost and time. Instead, it's more of a Cambrian explosion. Yeah, it's hard to think that— This doesn't go the way of generated text where, like, people still care about the posts and where they come from, effectively intellectual property. The instantiation of the actual product is less relevant than the story around it and the people behind it. And the Schelling point of, like, we're playing the same game. Of course, people will do their individual stuff as well. But anyway, let me tell you about— Okay.

Codex.

Codex is a powerful workspace for getting work done with AI agents, whether you're writing code, analyzing data, creating content, automating business workflows. Codex helps you move projects forward from start to finish.

Stretch this out.

What else we have going on, John? Well, I am so glad to be past that subject because you just don't like video games. Uh, I had enough. I had enough.

You had enough?

I had enough. What about racing series?

2 days in a row.

That's the video game.

That's the final frontier.

Yeah, you're going to be able to race anywhere you want.

Mustafa is back.

You're going to be able to put your morning commute in a video game and try and get your lap time down to 6 minutes. Do you think you could make it from Malibu to the UltraDome in 6 minutes?

6 minutes?

In a hypercar, an LMP2.

With the road cleared?

Yes.

I think it's— What is it? No, it would actually be—

15?

It would probably—

I assume you're going 200 miles an hour on this thing, right?

Yeah, yeah, yeah, yeah.

Okay.

And you could find a pretty straight line. Like, there wouldn't be that many that you'd be able to be hauling. Yeah.

We got to make it happen.

Yeah, we should.

We'll do this.

Put it to the test.

This is what super intelligence is for.

Because your Nürburgring time is already in the impressive category. I think you got the raw skills. to take it to the next level.

The 10-minute commute.

The 10-minute commute.

Turning your— turn your hour commute into 10 minutes.

It'd be pretty—

yeah, that's a fun game.

Anyway, Mustafa says, coming in with a major white pill, AI won't take your job anytime soon. In 10 years, only 5% of what humans do will be replaced by AI. Let's see what he has to say about that.

They're doing new media. over at MAI, the Humanist Review of AI. They made a magazine and they have a number of articles and they tapped MIT economist and Nobel laureate, Nobel Prize winner in economics, Daron Acemoglu, to write a piece about how AI will diffuse. And it's interesting because in this one, it's total white pill, but then he was on 60 Minutes and he's a little bit more blackpilled. So, we'll have to go back and forth. I think it's the same.

Wait, that's the same guy?

Yeah. Yeah, it's sort of an odd— we'll have to reconcile this, okay? So, let's see what Deren Acemoglu says in the Humanist Review. It's a longer piece, but Mustafa Suleyman summarized it here. He said, he argues that we need to stop building AI to replace people and start building it to make them better at their jobs. 52% of Americans are worried about AI's impact on their jobs. The fear is overblown and it's steering how we build AI. I like the 48% of just confident people. Like, good luck.

Good luck. I'd like to see you try.

Yeah, I'm not in the training data. I'm not. There's no RL environment for what I do. AI isn't in the productivity statistics yet. This was the famous line about the internet. It shows up everywhere except The productivity statistics. Most firms aren't— or most firms using AI aren't seeing real gains yet. Expect roughly 1.5% added to GDP over 10 years. Not a revolution. Electricity took decades to spread. There's some problems.

The problem is like— so, so the problem is like going back to our conversation with Sean Frank yesterday from Ridge. This is a company that's over 10 years old right now, growing 50% this year. Using AI across the entire business, from demand planning to creative to a number of other things, and the business is performing pretty much as well as it ever has. The thing is, you AI is impacting the whole business in different ways, in big ways and small ways, but you can't really pin it on. You can't give AI credit for the growth because there's so many factors contributing, right? New products, push into retail, all these different things. Yeah. But it's— but, but— and I think you can still say that maybe most firms aren't using it to see real gains. But when you say that electricity took decades to spread, I just don't view AI as It's not the kind of technology that takes a long time to spread when it works.

I agree.

And that's because we can use the internet, which pretty much everybody that's productive has access to, to spread AI. So already just a few bullet points in and—

But the Nobel Prize-winning economist has another example to push back on you. He says, if you don't like the electricity example, here's another one, speech recognition. He said, turning speech into text has been an aspiration since the 1930s. And this is a harder one to grapple with than electricity because a lot of what was in place for AI is also in place— a lot of the things that you're saying about why diffusion should be faster for AI was true for speech recognition. So people have been trying to do speech recognition, you talk, the computer types. Since the 1930s, going on 100 years of pursuing this. It became a huge step closer to reality thanks to the work by American— by the American husband and wife team, James and Janet Baker, starting in the 1980s when they founded the iconic company Dragon Systems. Dragon truly revolutionized voice recognition, and it would not be an exaggeration to say that everything else that has followed in voice recognition since then is thanks to the Bakers. Their breakthroughs— their breakthrough was to eschew previous attempts to build speech recognition systems based on syntax and meaning, and instead deploy machine learning techniques in order to build a pure prediction system which would map speech onto the best match text. Already in 1997, the year Dragon launched Dragon NaturallySpeaking, The first program that was capable of voice recognition from natural continuous speech. The momentousness of this was obvious. So in previous software, including the previous product from Dragon, Dragon Dictate, you had to pause in between each word to let it process the gap between words. You no longer needed to do that with Dragon NaturallySpeaking. This was 1997. I can vouch for all of this because Darren Ismoglu, writer. After developing repetitive strain injury in both arms in '97, he was an early user. Even with the very slow processors of the time, Dragon NaturallySpeaking was like magic. It would score close to 95% accuracy. I didn't know at the time that I would be a user of their products for the next 30 years. But what I immediately knew was that the processing power of computers improved. I could have access to much more powerful voice software enabling me to carry out my work despite a continued inability to type. By 2000, everything was going as I and others had predicted. The software had gotten better, and with better computer chips, accuracy and functionality were improving at a fast pace. What nobody could predict was that in the next 2 decades, despite advances in the underlying technology, consumers would see little of the gains thanks to a series of corporate disasters. Dragon was acquired in 2000 by a Belgian company, LearnOut and Housepee, that went bankrupt within months of the purchase. With its co-founders later being found guilty of corporate fraud. You hate to see that.

Let's hit him with the naughty naughty.

So Dragon was then purchased by ScanSoft, rebranded as Nuance. I actually worked for a startup in college that got acquired by Nuance. Very, very talented team over there, which progressively deprioritized the consumer in favor of medical and enterprise markets. If you walk into a hospital today, you will see a lot of doctors and nurses dictating notes even before the Whisper revolution that just recently happened. Microsoft acquired Nuance in 2022 and continued the same model. The result, voice recognition software on PCs today is only marginally better than what was available in 2000, far less progress than the underlying technology would've led me to expect. This disappointing history lesson is one of the reasons why even greater curation is warranted when making predictions about the fast diffusion of AI models today. Breakthroughs in the infrastructure of a new technology need to go hand in hand with the right kinds of applications built on this infrastructure. And then, of course, you need the right marketing structure to reach potential customer base, as well as inputs from consumers to act as the impetus for modifications and improvements of the applications. This is what we are not seeing for AI. This is what was completely bungled for voice recognition.

What do you think?

A whole lot of mumbo-jumbo, John.

Okay.

Let's move over to 60 Minutes because Darren Acemoglu takes the other side of his own argument, arguing against himself with Brendan Foote from Merkur.

Job category on LinkedIn.

And just as machinists once replaced artisans, Brendan Foote insists he's creating a career of the future.

Out of work?

Fear not.

Come crane AI.

Literally all of the top 20 economists in America, they would say the exact same thing, which is that there's going to be more jobs in 10 years than there are today.

You're a top economist in America.

What say you?

No.

One of the greatest edits ever.

What we are living through with AI is unprecedented. The first phase of the Industrial Revolution.

You're positive this is the same guy?

That—

Yes.

In this article that we just read.

Yeah. Wait, I don't think these articles, like, these 2 things are incompatible.

Break it down for us.

So, the article was—

Yeah, because he is saying 5%, 5% of the things humans do are at risk.

No, he's saying that right now, if we keep building AI the same way, it will replace workers, but we need to build it a different way that, like, empowers workers. That's what the headline of the article is, Will AI Replace Workers? Not If We Build It Right.

Okay.

He's saying like right now, on the course that we're doing right now, it's going to replace workers.

Got it.

So, we need to shift courses.

Okay.

Right?

That's what I—

And his argument from an economic perspective is like, that will actually increase productivity. That will be better, not just because higher employment is good, but also higher economic growth is good. So, we should be trying to solve problems that cannot be solved by people right now with AI as opposed to just—

Just displacing. Yeah, just moving stuff around versus doing net new things.

Okay.

I like that.

I believe that's what he's saying.

That's a pretty good position. But that is such a funny edit. You're a top economist. What do you say? No. One of the greatest. Anyway, let me tell you about Cisco. Critical infrastructure for the AI era. Unlock seamless real-time experiences and new value with Cisco.

And here's the thing.

Yeah.

Top economists have never been wrong on anything.

So, you know.

Yeah.

Literally 100% hit rate.

Yeah. You can find a take for anything.

Yeah.

Find the fast takeoff, the slow takeoff. Never better summarized than by— I believe it was the Financial Times, not The Economist, that had the chart, which was either things are going to go completely asymptotic upwards Completely asymptotic downwards or completely the same, which is the best analysis possible. I don't know. It does feel like it's a critical moment. There's been so many people building up to like 2027 is going to be the crazy year. 2026 is going to be the last normal year. Like, like something has to happen in 2027 that will answer a lot of the questions that have been being asked over the last 5 years, you know? Since GPT-3, basically, which is like RSI, superintelligence, serious move on GDP, serious like showing up in the statistics somewhere other than just debt. But I mean, at the same time, you see like weird things happening in financial markets, like gold is spiking and like it's hard to put that on AI, but there's still like, there's pockets of like— Okay, there's something odd happening that you might, in retrospect, attribute to the AI boom. Anyway, what do you think of Jeremy's take? What agriculture did to our bodies, AI will do to our brains. Make us stronger and taller. I think he's being negative.

I think he's being negative. Didn't it in some way make us weaker because we're consuming more grains versus—

Yeah, I mean, some agriculture made us a lot taller and stronger, but then some made us fatter, for sure.

Yeah.

Because we overconsumed. And so, yeah, if you're overconsuming information, it's brain rot, yeah. But a good encapsulation of the thesis anyway. Let me tell you about Railway. Railway is the all-in-one intelligent cloud provider. Use your favorite agents to deploy web apps, servers, databases, and more while Railway automatically takes care of scaling, monitoring, and security. Where do you want to go next, Jordi?

We got to— we got to talk about Warren Buffett because he's just like you. He's just like you folks. A viewer's guide to Warren Buffett's nightly trips down the YouTube rabbit hole. Glenn Close clips, 1950s commercials, Uzbekistan's version of America's Got Talent. The legendary investor can't resist the algorithm either. Feeling guilty about spending hours each day watching YouTube? Don't be. Warren Buffett is right there with you. Glenn Close clips, commercials from the 1950s, Paul Anka videos. The famed investor can't get enough. Buffett enjoys speeches by complete unknowns. He's a fan of Berkshire Hathaway's own content, too proud of how it performs on the video platform. He watches it all nearly every evening, usually from a recliner in his Omaha home. Warren's range is broad, says Sharon. Osberg, a close friend and frequent viewing companion. It can be the silliest things. Buffett also goes down random rabbit holes, just like us. The next day, he'll come into his office gushing over a moving amateur rendition of My Way. In August, Osberg visited Buffett's home in Omaha. They spent hours on YouTube grinding. We watched a really good documentary on Roy Cohn, she says, and the Uzbekistan version of America's Got Talent. That would be Uzbekistan Got Talent. It's unclear. Easy to remember. Easy to remember. Somewhat predictable. It's unclear if Buffett is one of the channel's 34,500 subscribers. Buffett, who recently turned 96 and stepped down as Berkshire's chairman, declined to comment on his YouTube habits. What's up with Sharon just blowing up, blowing up his whole routine? Like, if I invite you into my home to watch you— like, if I invite a friend to my home—

You're an algo doc, son.

If I invite a friend to my home to watch YouTube videos with me for hours and hours and hours and go down the YouTube rabbit hole, and then a week later you're calling up the Journal spilling, spilling the truth about my, my, my video consump— my rabbit hole habits.

Yeah.

I feel like, I feel like Buffett had to be like, I want you to tell the truth.

You think it's astroturf? Yeah.

I think I want you to take this. You think he's paying people to read these charts? I think he's trying to— I think he's as locked in as ever. I don't think he's doing anything for entertainment. He wants the other allocators to get comfortable. I can spend a few hours on YouTube every night. Meanwhile, he's studying the blade every single night. Yeah, just giving him the benefit of 4D chess.

Yes.

Occam's 40 razor.

Yeah, that is Occam's 40 razor. Yeah, this is what he's doing. Makes sense. Uh, um, tier list time. Should we do it? Do you want to rate it or do you want me to rate it?

Uh, I want, I want you to rate it.

Okay, we got a tier list. First, Glenn Close clips. I'm kicking it off with C tier. I'm not that— it's a little old for me, not into it. Commercials from the 1950s though, I'm going A tier.

Yep.

Commercials from the 1950s, you're going to learn a lot. It's going to be an interesting time capsule, and they lend themselves very well to YouTube videos. Paul Anka videos, I'm not a huge Paul Anka fan. I would go D-tier with that. Berkshire Hathaway videos, studying— he's watching game footage. So, I go S-tier with Berkshire Hathaway videos. Also, deeply emotional. His business partner. It's a way to remember him. Makes a ton of sense. Amateur musical performances, including— see, I liked Frank Sinatra's My Way. An amateur cover needs to have a twist. It needs to be a heavy metal rendition of My Way to get A tier for me. So I would go B tier.

Or some, like, if he's watching early Justin Bieber performances, somebody that's about to go on a generational run.

That's good too, yeah. That's good too. Roy Cohn documentary, fascinating. If you haven't seen The Apprentice, it's a biopic about Roy Cohn. I would definitely watch that documentary. I'd go A tier with the Roy Cohn documentary. International talent shows, never really dipped my toe in that pool.

Don't knock it till you try it, John.

I just, I need to put it in C tier because I don't, I don't have an educated, uh, an educated opinion here. Railroad videos, on the other hand, fascinating. Tons of lessons to be learned about the modern era.

Alpha.

I'd go S-tier. He also owns a railroad, so he's again studying game film effectively. Big band era content, a little dated. I would put it C-tier. Let's do Charlie Munger speeches. Of course, that's That's in the Berkshire video category. That goes S-tier.

That's right.

12 Angry Men. I like that he was watching this on Turner Classic Movies. He was watching the whole film. So I'd put that A-tier for TCM movies generally. But I noticed that when they mentioned Bill Murray movies and clips, Lost in Translation, I'm not a big clip movie watcher. I'd rather just watch the trailer and then the whole film. I don't really like the whole clip, although I do get sucked into them every once in a while. For movie clips, I would go B tier.

Yeah.

Do you agree with that roughly?

I think you nailed it.

Okay.

You one-shot it.

That's our tier list for today.

Yeah.

At the end of this article, it goes into— I'll just read it. Lately, Buffett has been turning to YouTube to remember Munger, who died in 2023. He's been viewing Munger's speeches, including a 2018 talk at the University of Michigan, marveling at his wit and wisdom, colleagues and friends say. You feel the bond between the two of them as he watches Charlie. It's really quite something. And when we were reading this earlier, we almost started shedding a tear.

Yeah, it's emotional. Truly one of the greatest partnerships in history.

The greatest partnership, hands down, the greatest partnership.

It's hard to think of anything else that even comes close. But fun that we're getting more color commentary on the Warren Buffett life. Everyone knows the McDonald's habit and the house and the indefensible plane and stuff.

Yeah.

But this is a fun different spin in the modern era. It shows you that he's not entirely a Philistine. There's a whole version of the lore that could have been, oh, he still only reads the newspaper. He doesn't use the computer at all. And that would have been somewhat predictable. So, I like that this is like a left turn for his brand.

And good 4D chess, too.

Yes. Good 4D chess. Yes. Fooling— well, yeah, we'll have to ask Greg.

All right. Scott Stevenson.

What'd he say?

Been on the guest before— been a guest on the show before.

Yes.

He's running Spellbook.

Yes.

He is saying, speaking some truth that's gonna be very controversial.

I'm gonna read it.

He said, if you do hardcore knowledge work, you are likely severely underestimating how much sugar you need to sustain peak performance. Protein alone cannot sustain a full day of System 2 thinking. He follows it up and says, allowing myself to drink sweet drinks while working has been the most profound performance enhancement in the past 5 years for me.

Hmm.

For the past 2 weeks, I've been outputting at 50%, higher focus longer, 1 Powerade at my desk per day. I'm able to do 12-hour days the way I used to be able to when I was younger. I don't have any crash. I crashed much worse end of day when I was avoiding sugar. I'm not sure why this is the case. It seems contrary to most of what I've read. Hypothesis 1, I burn a lot of energy working at a computer and easy fuel gets burnt off quickly like a runner. I was in a calorie deficit that was hurting my performance previously, same reason chess players consume chocolate milk while playing. Hypothesis two: I have diabetes. Any other ideas? Sugar is a superfood.

Okay.

Get it through your head, John. Get it through your thick skull.

I like sugar.

I know you like sugar, but not enough.

Not enough.

Not enough. Sugar is you know people have been told that it's evil and bad for you and all these different things. But for the listeners at home, if you haven't already. It's time to study Dr. Ray Peat.

Mm-hmm.

That's really all I'll say. Go down the Ray Peat rabbit hole tonight on YouTube, Warren Buffett style. Spend a few hours listening to Ray Peat talk about sugar and metabolism and a bunch of other topics in that whole orbit. I started— I've probably talked about it on the show before, but I just started drinking a A classic Coca-Cola every day years ago.

You still do?

Still almost a daily habit.

Interesting.

I try to fit it in every afternoon.

Every afternoon?

Every afternoon. It has some caffeine in it, so I don't want to do it too late, but it's tasty. You feel good before, during, and after, and I can't recommend it enough. And a prediction, I'll say it on the show. I've said it off air many times. I think in the same way that salt became one of the hot sort of like supplements of the last few years, sugar is the next one. Every single new drink brand—

Imagine getting the stick packs that are like LMNT, but it's just sugar in the raw.

No, that's— I mean, it's basically— I'm not gonna go that far, but all we've been through 10 years, 15 years now of like all these fake What is in Liquid IV?

Is that sugar?

That was pretty high sugar and they always got flamed for it.

Oh, interesting.

But the truth is that proper hydration requires some amount of sugar. Yeah. So anyways, I'm glad that we are about to exit the slog that we've been through of all these different artificial sugars that one tastes inferior, don't give you energy and often are bad for your gut or have other negative long-term impacts. So sugar is coming back. You heard it here first.

There's a really good source of sugar that literally comes from Mother Earth. It comes from corn. It's like high fructose corn syrup. If you go to 7-Eleven, you get a Big Gulp. It has literally like Mother Nature created the corn They extract the syrup. Syrup's like as American as it gets. You put it on pancakes and you just chug that down and you're good to go.

Okay, well, I want to see you just with some high fructose, just a bottle of high fructose corn syrup.

Challenge accepted.

Chugging that. I'll have—

Superfood. You said it, not me.

Not quite.

Not quite.

Stick to good old-fashioned cane sugar, folks.

Let me tell you about console.com. Console builds AI agents that automate 70% of IT, HR, and Binance support, giving employees instant resolution for access requests and password resets.

All right, John, do you think— do you think what's the right amount to spend on your office per employee? Because CastleHook Partners, an $11 billion hedge fund, is paying around $600,000 a year in rent per employee.

That seems pretty cool.

And that feels like—

Whoa, that's really a lot. OK. That's really, really high. Okay, Castle Oak Partners, the $11 billion hedge fund founded by former Soros Fund Management investor David Rogers, signed a New York City record of paying $350,000 to $400,000 a square foot. That's— yeah, for in the top 2 floors of the new 625 Madison Tower. Castle Oak, which is backed by Stanley Druckenmiller, will pay $21.2 million for 53,000 square feet of penthouse space overlooking Central Park. Castle Hook has just 35 employees. Are they expecting to scale up a lot and change that?

Hopefully not. I mean, you don't think this setup is going to deliver alpha? You're like, beautiful space. Yeah. They need to scale AUM. They need to deliver returns. Otherwise, they're going to be thrown out of the building.

I mean, it's $21 mil a year.

Yeah.

Okay.

No, I don't think it's unreasonable given their scale.

Yeah. I mean, it also probably makes sense for clients that want to come by.

There's just not that—

A lot of hedge fund stuff is just over the phone.

Well, given their fee base.

Yeah.

No, no, no.

They can clearly afford it. But the question is, like, what are the knock-on effects other than just, like, When you go to work, you feel great. There are certain businesses where it's like, oh, well, like if you're a VC and entrepreneurs come and pitch you and they see a nice building, they're like, oh, this is a serious firm. I should work with them. But hedge funds oftentimes, like the CEO isn't going by the hedge fund's office all that much. And same thing for the clients. They might not actually be going and doing a pitch in person, although I'm sure that happens. So it just depends. I don't know. Pretty cool. I want to see a video. I have a question.

tour.

It looks like these windows are really, really big, but it feels like the top 2 floors, it looks a little lower than some of the renders that I've seen from taller skyscrapers in Manhattan. Can you be too high in Manhattan?

I haven't been yet.

No?

No fear of heights or anything? I feel like if you have a clear view, like, I mean, I guess you always leave yourself open to a taller building popping up and blocking your view, and that's one of the advantages of being extra high. But I feel like a view like this looks a little bit lower, but maybe I just have the perspective all wrong.

I don't know.

Let me know. What do you think of the new passport? People were going back and forth on it. Paul McGregor was saying, if I was redesigning the COVID of what is arguably the country's most important public document, I'd first hire both the best typography typographer and crest designer money can buy, then fiddle about with the layouts, probably. And Joe Gebbia fires back, hi, when you're done finding— when you're done fiddling, please send me resume of top crest designer. Make sure they can do better than the 1782 official national anthem. It looks like a very—

National emblem.

National emblem. Yeah, the redesign looks like a very tasteful and moderate iteration. Like, I think people are reacting to this like it's like now it's all gold and it has like a bear and an eagle fighting on it and it features like the UFC octagon, which like was sort of where people's expectations are for these types of things. But this feels like—

Yeah, it's surprising there's no UFC branding on the passport.

Yeah, no ads is a big step backwards, but Yeah, seems like a pretty moderate iteration here.

I don't know.

Yeah, they made the eagle a little bit bolder.

And also, I think the eagle is like the exact same design, right? It's just bigger. I think this is cool.

And I like that the United States of America is bigger than the passport itself. They just said, make the eagle bold. Bold the eagle.

And I mean, just— I'm no typography expert, but in the previous passport, passport was the biggest word. But everyone knows what a passport looks like, but they don't know what country it's from. So, if you pull out a book at the airport, it's very obvious that you're pulling out a passport, but it's not obvious what country it's from. So, putting the name of the country bigger makes more sense, right? Am I crazy or is this like—

I think you might be onto something.

Design 101.

I think you might be onto something.

Okay. So, I think the new iteration makes a lot of sense. It's just like put the name and put the logo bigger. So that you can see that and everyone knows that it's Passport, so put Passport smaller.

Yeah.

I will say that the Passport text where it's kind of sans-serif, all caps, and there's like spacing, that does look like AI slop kind of.

It does.

This is like very common when you just have too much—

Too wide of kerning is your problem. Okay. The kerning needs some work.

Yeah.

It does read a little odd.

That's pretty high-stakes graphic It is. Most stuff, you're like, yeah, we're gonna make this, you know, pamphlet for a conference and nobody's gonna read it. And if they do, they'll look at it for a second. The passport's like, we're gonna make 100, you know, 100 million of these and then they're gonna carry them for a decade.

Yeah.

Representing the country globally.

Yeah.

They will be inspected many times.

I'm not, I'm not in the I don't want to leave America. It's, like, not a good thing. They can just get rid of them. Just stay here.

I don't need to go. You would love that.

Just wind down the whole program.

If you could hit the button that just, like, ended the US passport.

Yeah, just wind down the program. Just go to Alaska if you want it to be cold. Go to Arizona if you want it to be hot.

Problem solved. Go to Hawaii if you want to go to the tropics.

Why do you need to go to another country? We got everything you want right here. Mike Isaac, he's out of the New York Times.

That's right.

Went to the Atlantic.

Trade deal.

Do you think there's anything to that rumor that he had amassed an immense amount of soft power and was considering a CEO coup?

I think he was making a run for it.

He might have, and then he was ousted, defenestrated for becoming too powerful.

Yes.

That seems like Occam's razor, Occam's 40 razor in this scenario. It also could be that he just wanted to go to The Atlantic and write longer, different pieces in a different style, have a fresh look.

Yeah.

But I do like that we caught him via the paparazzi. It was very good. There's some drama at The Atlantic, though. We had the CEO of The Atlantic on our show. He has a podcast. This is Nicholas Thompson, Atlantic CEO. He was talking to the media about the media business on Vox Channels podcast last month when host Peter Kafka— I did Peter Kafka's show, but I didn't know it was— I didn't think it was called Channels. Maybe he has a second show. Peter Kafka asked him, how was the magazine's editor-in-chief Jeffrey Goldberg feeling about the growing success of Thompson's personal personal podcast. So he's the CEO, but his podcast is blowing up, blowing up because he's getting really great guests. And then there's also a— there's a tension here. He says, we make a lot of money on the stuff I do. Okay. This doesn't seem controversial anymore. I don't know why they're writing about this. Just giving him a victory lap. What is this? This is such a puff piece. This is insane. Insane puff piece in Semaphore about Nicholas Thompson, just printing on his podcast. Okay, we get it. That said, he says Jeff likes it when we make money. Thompson's response ricocheted around The Atlantic's newsroom. Thompson's podcast, The Most Interesting Thing in AI, has emerged as a source of tension between the magazine's business and editorial teams. Journalists have complained that the lucrative and high-profile show is not subject to The Atlantic's editorial rules input or oversight, yet a casual listener could easily take it for an Atlantic editorial product. I did see some people going back and forth when he did the Boz interview. People were like, is this a sponsored podcast? Like, this was like— because he opens with like a pretty soft question, just being like, what are you launching today? And then he does ask a bunch of really hard questions later in the interview. But I think people just keyed off of the fact that Right. Like, it was the moment when Meta was in the depths of like, they're screen recording everyone. And so, that was the question you gotta ask in that interview. They— he did ask it, but just later in the interview. And people were like, oh, he didn't ask it up front. That's not enough. Anyway, Thompson's show is made by the magazine's business side. And like any number of contemporary podcasts, yes, features interviews with technologists and business leaders interspersed with sponsored content. When the, when the podcast launched in 2024, the newsroom raised no objections. But now that it's printing, they got an issue. The Atlantic has featured a robust paid content arm for years through its creative marketing studio and custom events business. They got to go further. You know, if they're doing sponsored content, I think that they should be auctioning off the guest slots, but they should also be auctioning off the host So it should just be like, okay, you can just post whatever you want on our channel if you pay enough.

Just like a pipe.

But also sell ad slots. So it's like, okay, Boz wants to do some SpawnCon, but Alex Wang also wants to. They're gonna talk to each other on the Atlantic podcast, but then there's gonna be ads in the feed too.

But it's presented by Anthropic.

Yes, exactly. Exactly. But maximal, maximal financial things. Anyway, they're going back and forth on that. We'll see where that goes. What else is going on?

What else we got? What else we got?

Oh, le chaton fat.

It's time.

Mistral.

Le chonk.

They're calling it le chonk.

Le chonk. Yes. They're leaning into the jokes. Mistral AI, the French NeoLab. Is it fair to call them a NeoLab or are they just a lab?

I think they're just a lab.

They're just a lab, right? They're pre-NeoLab.

Yeah, I would say so.

They're kind of like— you would call them pre-NeoLab specifically?

Let me actually consult our nomenclature.

Okay. Oh, I think they were like National Lab or something like that, up there with Lawrence Livermore. But they launched Mistral Large 4, a.k.a. LaChonk, 1 trillion parameters, Natively made it multimodal, 49 billion active parameters. It is the best open weight model from US or Europe on aggregated benchmarks, state-of-the-art on critical workloads, including cyber defense, manufacturing, and finance. And it surpasses closed frontier models on visual grounding, forged in Europe.

Forged.

End-to-end deployable from Europe via our own Mistral cloud infrastructure. Available to all via API today, working with cybersecurity partners privately. Open weights release end of October. So, presumably they'll release the open weights for like the cyber nerfed version, I would imagine, something like that. I don't know. Anyway, what is your review, Tyler? Do you have any other data to share about Mistral's progress?

Yeah, I mean, it looks very good. I think it's very much on par with the frontier of the open source. So the Chinese labs, it's very much a contender. It's not necessarily destroying all of them on every benchmark, but it's certainly, if people are worried about using Chinese open source models because maybe downstream there's going to be some legal ramification or whatever, this is a very good choice. It really does change the open source, closed source debate because Yeah, I think Mistral and Reflection, which is kind of the American open source lab, like, we're now starting to see, like, okay, these models are actually, like, quite competent.

And Reflection came out with— it's called Beam yesterday. Whoa, not tapped in. They're like, sorry, sorry, I can't hear you. I can't hear you.

It's called Beam, yes.

Anyways, we can come back to that. But great to see them getting a new model out.

And Leo has the funniest benchmark of them. Says, it's pretty funny that it seems the one thing Mistral Large 4 is actually leading on is a benchmark about regulation. Another EU banger. Harvey's Legal Agent benchmark is up to 15%, whereas other models are down at between 5% and 13%. Interesting. I wonder if that's intentional. I wonder if there's actually anything to read into that. I really have no idea, but it's certainly valuable. I'm sure Harvey and all the other legal companies will be excited, because we read the whole thing about Harvey, their margins being in a really rough spot for a minute. when they were overly reliant on like the most expensive frontier models. Now there's a ton of price competition between OpenAI and Anthropic. And then obviously there's a bunch of pressure from open source models. So they like— it's the true— it's truly the create the benchmark that you want the labs to hill climb. And they just created the benchmark for like their exact workload and then they're able to do that.

Sushi Commander says, what are you talking about? Both Beam and Mistral Large are a generation behind the latest Chinese So, putting you in the truth zone, Tyler.

I mean, you can look at the benchmarks. It's like, yeah, it's not leading them. It's like among one of the contenders. It's not the frontier of open source model for sure. Yeah.

It's like in line, but a lot of the—

a lot of the—

like the most recent round of Chinese open source happened like a couple of months ago, which is always rough because like you catch up to the frontier and then it's like, well, you're going to take another 6 months to get something out where they're going to only take another 2 months to get something out. So that window— but there's so many other geopolitical implications that lead companies to go one way or another. So we'll see. We'll see. Olivia Moore has some data on what's happening in the battle for consumer AI dominance between ChatGPT, Claude, and Gemini. A16Z pulled the data across web and mobile plus consumer revenue. ChatGPT is at 2x Gemini scale 6x scale on web and 2.5 Gemini scale and 14 times Claude scale on mobile. Interesting. I mean, I guess that makes sense. More consumer-focused. You're pulling there.

The main thing here is just look at, look at the run from Claude.

Yeah. I mean, we saw this in the App Store, right? Like you would open up the App Store and Claude was at the top for a while and then, and then fell down and now ChatGPT is back on top. And they're just going back and forth. The war for the frontier is just trading shots back and forth, back and forth.

Right now, Muse still at number one. Vinted, who we're working on, is number two.

Vinted is still number two.

So dominant.

The folks over at Vinted are really doing a fantastic job. Well, we should bring in our first guest. First, let me tell you about Shopify. Shopify is the commerce platform that grows with your business and lets you sell in seconds. Online, in-store, on mobile, on social, on marketplaces, and now with AI agents. We have Greg Fleming from Rockefeller Capital Management. He's the—

Here we go.

CEO.

What's going on?

Welcome to the show. How are you doing?

Well, good to see you, gentlemen.

Thank you so much for taking the time to come chat with us. Since it's your first time on the show, I would love to— for you to introduce yourself for the audience and give us a little bit about your background and then your business today.

Great. I'm the CEO of Rockefeller Capital Management, which we started in 2018. I've spent most of my career in the financial services business. I was a management consultant at a company called Booz Allen and Hamilton in the late '80s, early '90s, which was actually great training for learning about lots of different industries. I went from there to Merrill Lynch, and I was an investment banker mostly in the financial services space. A noteworthy deal that I did, I was the lead banker on the IPO of BlackRock in 1999 at $14 a share. So it's done quite well since then. And I actually just rejoined the board for the second time last year.

Congratulations.

I was the president of Merrill Lynch during the credit crisis, and I negotiated the sale to Bank of America, which was— and I have said this before publicly— the most disappointing moment in my career. I was happy to keep the company out of bankruptcy, but it never should have had to be sold. And having to sell it was very difficult. I then went to Morgan Stanley, and I ran wealth and asset management. And I left there in 2016, and I wanted to start something. And I found out that I could partner with the Rockefeller family, buy the family office that was started by John D. Rockefeller way back in 1882. So, the guy who made all the money.

The guy.

His wealth, by the way, gentlemen, at one time was 4% of GDP. So, maybe it's still The wealthiest guy in history, that family. So we partnered with them. We started a business. I was backed by an investor, Viking Global Investors. They've been a great partner. We started on March 1st, 2018, and we built a firm that is, I think, among the best serving wealthy American families. We call them high net worth and ultra high net worth, and we provide comprehensive advice to these families.

Yeah, I have so many questions. Let's start with the process of actually building the firm. Wealth managers have lots of different products, lots of different relationships. How did you think about building out a portfolio of offerings quickly as an upstart firm to actually give a client who's coming along for the ride at the early stages a full-service experience? Was this like a hiring spree? Was this partnering with other banks and firms to try and fill gaps. What was the secret to actually delivering a top-tier level of service quickly?

That is a great first question because the reality was we bought Rock& Co., which was originally the family office. There wasn't that much there. We had to build it. It was the classic J-curve in the first couple of years. We were investing, Because you couldn't really onboard those families that have a complex set of needs until you had a broader set of capabilities. So the first few years was a significant investment time for us.

Yeah.

We started hiring the private advisor teams to take the lead working with these families in late '18, early '19. But we started building the capabilities right out of the gate. And it's really taken us— we have our 9th anniversary next March. We've been building, building during this time, and only in the last few years did we have all the capabilities in place so we could provide comprehensive advice, generational tax, financial education. If the family owns a business, we give them advice on that. We have a boutique investment bank.

Sure.

It was only in the last few years that we really had all those capabilities come together so that we could really start to grow the firm.

So obviously the brand is, is incredibly storied. But— and you had an incredible journey in your career, obviously tons of connections. But how did you think about the top of funnel? How did you think about onboarding clients? And then once you'd onboarded that initial set of tapping your Rolodex, I imagine, how did you think about growing that client base? What actually works in this particular market? I imagine it's not Super Bowl ads, but— But what does move the needle in the modern era?

The thing that we think is the secret sauce is providing advice across the full range of the family's needs, because wealthy families have a complex set of needs. They're trying to plan generational passage of wealth. We do that. We give them a lot of tax advice. We put them in the best investment solutions. And those include the best alternative managers, also direct investments. You know, we put clients in a lot of the companies that you guys talk about that are earlier on. We provide advice on how to talk to the next generation about money. We have a Next Gen Advisory Council with some of the prominent wealthy American families represented on that Next Gen Advisory Council. We do philanthropic advisory. We find that most of our families— one of the greatest things about the United States, in addition to the whole entrepreneurial machine that you guys talk about all the time, is the philanthropy. $600 billion given away every year. Almost every single family that we work with wants us to help them put in place a philanthropy plan, including young wealthy people, wealthy, you know, new, you know, new people making new wealth at some of these AI startup companies. Even if they're on the younger side, young people want comprehensive advice right up front. They want to obviously, you know, spend some of the money, save some of the money, but they also want to start giving it away. So, we spend a lot of time on the philanthropy side. That comprehensive advice, nobody really did that in the way that we do it when we started building our company. That was what we really set out to do.

How is the upper echelon of family office and multifamily office business changing what you do? Is there just a demand for A wider breadth of services for a longer period of time? Is there something where you need to react to the fact that some ultra-high-net-worth individuals are building whole trading desks and real estate teams, and they're doing such complex management that you still want to partner with them in some meaningful way or advise them on their journey?

You know, John, for most of them, until you get to, you know, really significant wealth, you know, north of $10 billion. Yeah, that's enough wealth to really have a fully staffed family office.

Yeah.

But, you know, in the first certainly hundreds of millions into the early billions, you know, we encourage families to think about using us because we'll do everything that they can build.

Sure.

We do it for lots of families, and it's easier for us to deliver that for them than it is for them to build it and manage it. There are headaches involved in building your own family office so that the very high end will, will, you know, do everything in-house. But really, everybody else is a potential client for us. And even the very wealthy families will sometimes do some aspects of things for them. They'll have a lot of it internal, but they might be turning to us for advice on different kind of investments or, you know, for for help on some of the sophisticated tax and estate planning, they might turn to us for that. But, you know, right up to a couple of billion dollars, that's a sweet spot for us.

Got it.

How, how has your, you know, private market strategy evolved over the last few years? So many of the world's most exciting companies have obviously stayed private for a long time. They're also quite high profile. So I imagine some of that strategy ends up being led by the client base coming to you and various people on your team saying, get me into Anduril or get me into XYZ name. And it feels like it's probably coming, you know, the ultimate strategy is being led in some way by both sides.

It is, Jordi. And in fact, Rockefeller Capital Management, we're in over 30 wealth centers, over 50 cities across the country. And one of the reasons we do that is we want to be embedded in local communities with the clients, with the families, because wealth is created in this country. You know, when I was younger, you had a certain set of cities and they're still there— New York and Boston and Houston and Dallas. Now you've got Nashville and Charlotte, obviously many places in Florida and in California, obviously Silicon Valley. So we're in these communities. So we'll often have people working at, you know, the companies that are getting built and built quickly. Come to our advisors and look for help. So we get a lot of introductions that way because we're embedded in the community and we do a lot of work with entrepreneurs across all different industries. And, and frankly, they benefit from our advice because they're really spending 24/7 building these companies and they're not— they don't have the time. They don't really have the training to do the types of things that we do in terms of helping them start to think about, you know, more of a diversified approach to long-term investing, philanthropy, things like that. So it's been a great— you know, this whole growth in entrepreneurs. I mean, I think we have 6.2 million new companies being created in the United States this year, which will be a record. This entrepreneurial surge has been great for us in terms of creating clients all over the country.

Yeah, totally. I want to talk about the research the research function at the firm. I imagine that clients want your house view on interest rates and inflation and whether AI is a bubble or not. How do you think about answering those questions? What team members do you have in place? How do you grapple with like the hot questions of the day to keep like a consistent view from a firm, but also tailor the advice to the specific client?

Yeah, we have a chief investment officer, Jimmy Chang, who takes the lead on pulling together a firm view. I have another colleague, Ruchir Sharma, who writes every other week in the FT. He's our chairman of international. So we've got some great minds. You know, and I spend a lot of time on this myself because clients ask me all the time and I'll give them my own, my own views. You know, I've been pretty clear that this technology is revolutionary. I mean, I'm in my 39th year in in my professional career, I've seen a lot, but there's nothing like this technology in terms of its impact on every part of society. You know, you have the CapEx investment and all the data centers and everything that's happening there, which is driving the economy. But then you have the other side of it, which is every private and public company in the country trying to figure out what to do with these tools, how to use them. How to drive productivity in your firm, how to enhance the client experience. So, you know, we think this is a revolutionary technology and is beyond anything that's happened to this point, frankly, in human history.

At the same time— so I completely agree with you, obviously, but at the same time, there are a number of Cassandras out there right now. That is offset by a lot of investors, a lot of people in finance who have lived through the Great Financial Crisis or dot-com or both. And I'm wondering, from your experience, what are the canaries in the coal mine that you do look towards as useful signals for something maybe getting too far? Because even when the technology is real, there's always a risk of overinvesting, right?

Yeah.

Look, I mean, you know, and I lived through both of those times, dot-com and the financial crisis. Not every dollar is going to get the return that you're— that it's looking for when you have a time like this with this kind of massive investment. That's clear. The question is, is the overall technology going to pull the economy and productivity forward in a way where it's going to have the kind of impact that we all hope for? I believe the latter is going to take place. I think that you're already starting to see some signs of productivity improvement. Now, you know, you've got different people arguing about that. But, you know, output per hour seems to be going up in some industries without employment going up. So, you know, as long as all of the investment leads to the kind of growth in the economy, the growth in productivity, that's obviously what the government's hoping for because You know, one canary in the coal mine that we could talk about is the fiscal debt situation, which is unique in my history and is a problem and may be driving some of the longer-term rates up. But from an AI standpoint, if the overall investment that's being made, trillions of dollars, leads to the effect on the economy and on productivity across thousands of companies and every part of society, then the, the return on the capital will be there. But it'll be differentiated. Not everybody will be a winner. There will be winners and losers. And, you know, even though we saw— and I lived through this— the dot-com explode, you know, in 2000 and 2001, obviously there were tremendous companies that ended up coming out of that. And the same thing coming out of the credit crisis. So not everybody's going to be a winner when you're looking at investments and, you know, and the capital that's chasing different companies now. Some of them will justify the the capital investment, some won't. But from a macro standpoint, I'm optimistic that all of this capital that's behind this whole AI investment will have the impact on the economy, productivity, you know, people's incomes. You know, a lot of the positives that are in the, in the dialogue, I think that that can come to pass.

Yeah.

How, how are you processing the I think there's a pretty obvious bubble in the early-stage technology private markets, where you have a company that goes from being worth, let's say, $0 to $2 billion a couple of months later.

You're talking about pre-revenue?

Yeah, yeah.

I mean, right now, there's all this NAV appreciation that you're seeing across the top early-stage funds because there's effectively multiple rounds happening in a very short period of time. There's so much excitement around these companies and these categories, and the TAMs feel so big that it's easy to get around to justifying, hey, if we give this company $500 million and they're going after this hard problem and there's meaningful technology change, they're going to be able to figure something out. But at the same time, we go through this every week where we'll have Multiple times a week, we'll have a company on the show we've never heard of, and somehow they have $500 million of fresh capital and a pretty small revenue base, at least, and a lot of potential. But it's a little bit nerve-wracking having just gone through a similar investment cycle in 2020 and 2021, although we didn't have this this sort of same technology shift.

Yeah, we didn't have the underlying technology shift, but yeah, there was that cycle. And remember, again, I did live the internet cycle. And John's saying, uh, uh, while you were, uh, asking the question, Jordi, pre-revenue, that was— remember clicks and different things that were looked at in '98 and '99?

And now it's just like talent. It's talent density. It's like if you get— and so, and, and so the investment thesis is just like, hey, These— we got 5 really smart people here. Time to give them half a billion. You know?

Yeah. Look, I mean, you know, in '99, you get in a taxi and the taxi driver would ask you for 3 recommendations on, you know, stocks you could buy or companies you could invest in. So there's some of that going on now, no question. The way we deal with it, first of all, we do a lot of diligence on the private investment side as well. We try to only put things in front of our advisors and clients that we think, you know, look like they could pass the test of time here. But that's still a challenge given the number and quantity that are coming out. Jordi, when you guys are seeing companies that you've never heard of and you live in this ecosystem, we're going to see companies that we've not heard of as well. Diversification is the way that we do it with our clients and also not just diversification within the space, but only a certain amount of the capital goes into this, this part of their capital allocation plan. So, you know, we have these families is broadly diversified across alternatives. Obviously some public market exposure on the equity and even on the fixed income side with rates backing up. And then a certain amount of the capital into things that fit the bill with what you're describing where some of them might work and some of them won't work. But the client is not exposed to more risk than they should be taking given the fact that we're in a we're in a time like that. And we're definitely in a time like that. What's different about this time— and that's always a dangerous phrase, and I've been around long enough to know that— but what's different about this time, and John was saying this before, the underlying technology and the underlying business shift is real. But a lot of these companies are not going to get the valuations that they're being awarded now.

Yeah.

One more.

Yeah, one more. How do you— how have you processed how success ends up clouding the judgment of otherwise really smart people? That's a good question. And you even talked about a low point in your career with Merrill Lynch, which avoided, you know, the worst-case scenario. But when you have absolutely knocked it out of the park for long enough, you start— you know, otherwise smart people start to not even be able to imagine the possibility of failure. And I'm seeing this, you know, just in the private markets where we're spending most of our time. Some of these managers have put up such unbelievable results for the past few years that— and at a certain point, you start feeling like you're playing with like house money to some degree too, which is another dynamic. But having seen all sorts of managers across your career, when do you feel like— what are the signs that you look for to think like, okay, maybe this person's overplaying their hand?

Well, part of it is what you describe. It's even generational. Have they been through the cycles? Is this the only cycle that they're seeing? Part of it is back to the diversification I was talking about before. But, you know, what you're describing is an incredible— is spot on in terms of what's going on right now. I mean, you know, you have clients that everybody's kind of expecting. You know, anybody who got into SpaceX early, you know, that's kind of the vision they have for every private investment. You know, am I going to be up 30, 40, 50 times?

Yeah.

It's like, look, this is not something that is, you know, recurs on a regular basis. You know, I was looking at the scale of companies now. Apple was the first trillion-dollar company. I think it was 2018 or 2019. There are now 15 or 16. I mean, it is— there's a momentum here. And, you know, look, this is what our advisors do. They've been around a long time. They've been working with wealthy families. They say, you know, let's stick with the plan. We can put this much capital in these types of investments. You know, we like these vehicles here for these— for this reason. We think this vehicle might be too speculative. We think these people They've done well, but they've really only seen one cycle. So we want to be careful there. Our advisors are providing that kind of advice, and we're constantly reminding, you know, people throughout the firm because I've lived these cycles and, and this one, you know, is taking on some of those characteristics. Again, having said that, the underlying driver is real, but there will be winners and losers. We look at diversification. We say to clients, just a Use a portion of the capital, be careful, and let's look for managers that have actually seen something go wrong because things go right and they go wrong.

Last question from me. How have you grappled with or implemented AI at the firm level, not as an investment strategy, but just you can imagine, you know, every email getting passed through AI, but wealth management is a luxury product. Some clients might want a handcrafted email, even if it takes a little bit longer. Where have you seen it integrate usefully versus, okay, maybe we're going to sit that one out?

Well, one of the great things about our business, because the clients— there's so much complexity around taking care of these families, the interface stays human beings. So, the advisor is going to stay out front, and we think that's the case for decades and decades. The family will come in, And they will meet with the advisor and the team. We use AI as a growth enabler. We're— and, you know, we've got, you know, partnerships with some of the top firms. We're building things ourselves. We're trying to make the advisor teams more efficient in dealing with clients. So a lot of things that used to be more manual or semi-manual will be done by agents. Yeah, and we're working on that. So I'll give you an example of that. Advisors spend a lot of time getting ready for the meeting with the family. They have the meeting, then they have the follow-up and, you know, making sure the follow-up occurs and the investments occur. We're trying to automate that meeting prep, you know, and have an agent really help pull that string all the way through.

Yeah.

So that's the type of thing that we're investing in so that the advisor teams become more productive, gives them more time with clients, they can work with more clients. And we think the tools will be very efficacious there. Yeah.

Removing the drudgery. From just piecing together portfolio, understanding what's happening. It makes so much sense. Thank you for coming on the show.

This was fantastic. Yeah, very, very cool. Great to meet you, Greg.

Great to talk about more soon. Have a great rest of your day.

Yeah, let's do it again soon.

You too.

Thank you. Goodbye. Let me tell you about Figma agents. Meet the canvas. Your AI agents can now create and modify your Figma files with design system context. We got TJ Parker back on the show four. He's joining the 4-timer club. This is his 4th time on the show. TJ Parker, welcome to TBPN. How are you doing?

Good, how are you?

So you really, you really said, you really said take this job and shove it. Is that what you said?

You know, I think I'd focus more on what I'm excited to go do now.

Okay.

Which is general medicine.

Yeah, which is the Nikita beer of healthcare. Is that kind of fair? I mean, in some ways, it feels like obviously going back to your roots and taking another crack at a similar problem in some ways.

Yeah. And you probably know this, but ironically, Nikita and I shared desks at Techstars in Boston in 2013. So it's particularly apropos.

We got to find that desk. We got to put it in the Museum of Business.

Yeah.

That's a billion-dollar desk right there.

That's a great desk. Yeah.

I actually had no idea.

That's crazy. But yes, restate the mission, then we'll go into the decision to step in, what you're planning to do, where the company is now, but walk us through it.

Yeah. I mean, I think our vision is super simple. We want to make it just as easy to shop in healthcare as it is to shop for anything else in your life. You know, I think today it's super convoluted. You have no idea what the price is for things. You don't really know what makes sense for you. And so we're trying to follow all the modalities in retail that we know work super well and apply those to healthcare. And so you can come to Gentlemedicine for anything. Could be as simple as just need a renewal of a prescription or as complex as you've got this thing you can't figure out what's going on and you want advice, you want to go see an offline specialist. We just want to be the one-stop shop for healthcare.

Yeah, the front door. Is there a Yeah. Like a secular trend here around consumers taking more initiative with their healthcare and actually knowing what paths they want to go down versus an annual checkup or just being responsive to there's a problem and my doctor handled it from start to finish? Has like the pattern of engaging with the medical system changed or is it changing?

Yeah, I think there's 2 trends there that make sense. The first is that obviously in certain pockets of healthcare, consumers have gotten very accustomed to a much more seamless CX, right? So it could be GLP-1s, it could be your annual wellness tests. Like, there are these kind of very specific categories where consumers have gotten used to this and they've started to expect it across the spectrum of healthcare. And then the other obvious trend is that with the advent of LLMs, consumers are just far more— they far better understand what they might need at any given time. And so it's gotten them much farther down that curve. And the challenge right now is that the output of an LLM is still just giving them advice. It's saying like, you probably should call the doctor and ask them this question, or you should get someone to write you this prescription, or you should go get this X-ray or this MRI or these labs.

Yeah.

And you're back to dealing with the health system as it's always been, which is that's a super complicated process. And so I think that trend is really important here where we're trying to close that gap from I know what I want to do to actually being able to immediately execute that.

What is your view on the AI models, the agents, like just cloning your voice and setting up a Twilio account and calling the doctor for you and basically just interfacing the old system just as an AI posing as you.

You're describing creating a deepfake of yourself.

Basically. I mean, that's kind of what's happening in some pockets of commerce. We're talking about a few million people testing this stuff out right now. It's pretty niche. But is that something where you think it's like that's going to be slower to take off, more early adopter niche, or you still have a play and that will actually be acceleratory for you?

Yeah, I think AI and LLMs and agents are like very good at replicating things humans already do. I don't think there's a single human that would tell you this is a particularly good experience and it's just administratively annoying, right? And so, I think even if you take the AI doctor side of this or the agent booking side of this, like I think that output results in replicating the current system, which I don't think anyone thinks Mm-hmm. Is particularly good. And so, and I think if you compare this category to all the other retail categories, like we've had 2 decades of building marketplaces and infrastructure and booking experiences, none of that exists in healthcare. And so, to actually make this work as elegantly as other categories, that's the thing you actually have to go build is what is this interface between digital and the traditional healthcare system.

Is it Is it going to be harder to scale the top of funnel without a hero killer product? I'm just thinking about the previous companies that were known as the hair loss company, and then they did a couple other things, or the ED company, and then they did a bunch of— or the peptide company, and then they did a bunch of other things. Do you need a beachhead offering to onboard people, or Or is there a message that's resonating that's more broad?

Yeah, I think if you asked me a decade ago, and I think if you asked most people in the ecosystem, everyone assumed that these very vertical-focused companies would eventually just chip away at category by category and ultimately go horizontal. It's like Amazon starting with books and then just like increasing selection.

Yeah, it has to be in every of the early-stage decks of like Hims, Roman, all those companies. They all probably We'll eventually be the Amazon of healthcare, right? I would imagine that was the—

Yeah, that obviously hasn't happened, right? And so part of the theory here is that you actually have to be horizontal from day one so customers can come to you for anything.

Interesting.

And I think there's a couple reasons for that. Like, if you just take the current version of offline care with a GP, like a primary care provider, like, they serve everything, right? Like, you don't have to go to 12 different places to get everything you need. Like, you have— like, you're used to going to one place. for whatever you need. And I think the other vector that's relevant is that oftentimes customers don't actually know what's going on. And if you take ED drugs in particular, probably half the customers on those meds should be on something else treating the underlying issue. And so, you end up with kind of a hammer-nail problem versus being able to actually get customers to exactly what's most effective for them at that moment.

And what What level of abstraction can you interface with the healthcare system as a relatively new company? Like, if you're building a fintech company, like, you'll just use Stripe for a lot of the card processing and whatnot, and then you do something fancy on top, right? If you're building like an e-commerce site. And with healthcare, there are companies that help with booking doctors, and there's healthcare record systems. There has been a first generation, if not a first and second generation, of medtech, healthcare tech companies. And I'm wondering, are those natural integration points? Or is the technology at a point where you can go deeper and interface directly with doctors or directly with insurance companies, and you don't need to aggregate aggregators?

Yeah, I think probably an unpopular opinion is that the data is actually out there now. And healthcare in many ways has the most open data because it's been federally mandated. And so as an example, when a customer signs up for general medicine, on average we're getting 11 years worth of medical records and they don't have to go log into 12 different health systems and go find everything. We just, we get it automatically.

Yeah.

And the other thing that's happened is the federal government mandated open pricing files a handful of years ago. So providers have to contribute what their reimbursement rates are to this central repository. And so when you combine these things, we can effectively give customers a clear cash and insurance price for any procedure or service with any provider. And that doesn't require us having gone and done a bunch of deals with a bunch of offline systems. We can do that on our own.

Mm-hmm.

I think then if you take the, like, the pharmacy side of this is a good example. We just launched our pharmacy experience. You can check out on General Medicine like you're on Instacart or Uber Eats at 50,000 pharmacies across the country. So, you can see, like, what is exactly my price going to be and just click buy like you do everywhere else and get the prescription at whatever pharmacy you want. And so, we've, you know, we've really accelerated our ability to allow you to interact with anybody in the system. And I do think a lot of that is a byproduct of regulatory work that's been going on for a decade, coupled with the capabilities of LLMs. And so I just don't think this would have been possible to build 5 years ago.

What are your strengths and weaknesses as a CEO?

Job interview.

I'm really good at tweeting. I had a really good day on the timeline today. So that's my biggest strength for sure. Important these days. Yeah. Weaknesses. I don't know. I try not to talk about those too much. I mean, I think like for me, like I was so excited to get back to building partially because it's a bunch of people I've worked with for 10, 15 years. And like we've built an amazing team across the org. And so yeah, in many ways, building an amazing team and getting to work with that team to build stuff.

Where do you see yourself in 5 years?

No, I'm curious, like what, you know, From PillPack, there's certainly like 10 massive mistakes that you made, and then you did probably like 1,000 things right, and that worked out to an amazing outcome.

Yeah, totally.

Are there like— this time around, is it really you're just sharper, more knowledgeable? You probably know that you can actually go way bigger this time than the last time, because why? What's the point of of running it back with the team if you're not going to reach even higher highs. But is there anything that stands out as like, we're definitely not going to do that, even from a hiring standpoint, mistakes that you made building the team the first time around?

Yeah, I mean, the thing that I definitely wouldn't replicate that we did at Propack is we were just fundamentally misaligned with other partners and participants in the ecosystem, right? We had this big battle with incumbents. That was just being young and naive. And obviously, it worked out in our favor. I think that there's a big difference with general medicine, which is I think we are ultimately trying to refactor the system around the consumer. And in many ways, that solves a bunch of problems for other participants in the ecosystem. So, that's probably the biggest strategic difference. I think tactically, like, we've spent 15 years, like, in the guts of the healthcare system, like deeply understanding all the underlying rails, like what's truly possible and not possible. And so, in many ways, that is probably the biggest difference is that you've got a decade and a half of understanding of what you can really do in healthcare. And then coupling that like really niche expertise with how does that manifest as like a differentiated customer experience, I think is what we're uniquely good at is sort of being— in the weeds over here, but then abstracting it away so that for the customer, it's just got— it was super simple, and they don't really need to understand, like, how hard it was to make it that simple.

The Jaguar Type 01 has leaked. Is this a good car for the modern healthcare CEO?

I think I'm out. I'm going to stick with my Raptor R, which is probably the right healthcare CEO car.

Yeah, I like that.

True.

Is there any type of— is there any type of, like, interesting why now? Because people are spending— like, it is such a better experience using an LLM to understand, like, a health issue you might be having versus historically, like, you know, Google to WebMD to Reddit, and you're, like, trying to find somebody else who's had, like, a similar experience or whatever it is. And now you can have a conversation about it. Now you can involve you know, whatever it is, like images. Now it even has like memory, so it can sort of tie things together. But it feels like that, like, you're— it feels like you're exactly the kind of business that should be advertising in LLMs, which is like—

Yep.

Because it can be done in a way that's not even personalized, which I think people are going to be a little bit wary of. But you can just, like, if somebody's using a free version of ChatGPT, you could just slot in and say like, hey, like we know you're kind of like, you can, here's how you actually solve your issue versus just staying in the sort of digital realm learning.

Yeah. I mean, I think obviously, the first iteration of this product is our storefront on generalmedicine.co. I think we assume that there's going to be all sorts of different ways consumers get access to these products and services. And going back to the retail example, Like, we know exactly what that looks like, right? A Shopify SKU pops up in an LLM, you can immediately add to cart and purchase in one click. Like, there's no reason that same CX shouldn't exist in healthcare once you have this infrastructure layer built. And so I think consumers are going to, whether it's an agent and they're literally never really interacting with us, or it's like Shopify and it's a SKU that pops a modal, like, we're— I think we're building the hard part of that CX, which is the underlying infrastructure to make that work.

Amazing news. Well, congrats on the new—

Glad to see you back in the game. You had, you had, you had a good few years as a capital allocator, just, just hanging out, begging people to let you give them money. But you're back, back in the driver's seat.

Yeah, worked very well.

The problem is I wasn't a good enough golfer. It's pretty depressing to play golf like 3 days a week and still suck. So I had to get back to operating.

Here we go.

Back at it. Awesome. Super, super excited for you and the whole band. Yeah, it's going to be a great run.

And thanks so much for coming.

Great to see you.

We'll talk to you soon. Let me tell you about Public investing for those who take it seriously. We got stocks, options, bonds, crypto, treasuries, and more with great customer service. Up next, we have Juwon Kim from Melius, the co-founder.

You know him.

You know him.

He may have triggered you live for the first time.

He used to work at Ramp. Now he's building his own company. Welcome to the show.

What's up?

How you doing?

How's it going, guys? Yo, congrats on the success and becoming rich.

You too. Thank you. And congrats on—

Always crude in your ways.

Yeah, sorry.

And always consistent.

But you got some funding. Speaking of money, how much did you get? How much did you raise?

Yeah, we raised $25 million. We actually raised a seed round.

I can't hear you. I can't hear you over the sound of the gong.

Congratulations.

No, I appreciate it.

General Catalyst, CRV, back-to-back.

Genius.

Geniuses.

Ben, Adam, the whole gang. But, uh, great line. Yeah, great lineup. What are you building? And, and, and go for it.

Yeah, what's up?

No, I was gonna say, like, you've, you've— it feels like you've been in— have you been, like, building for a couple years now? Is it more like a year?

I don't know, dude. I left Ramp back in September of last year, which is crazy to think. Yeah, it feels like it's been a while. It's been a wild ride. Yeah, it's been a wild ride. We raised from General Callis back in November. We actually started on a different idea in performance marketing, and then we decided to pivot away from that back in March. And then we landed on building the agent's lab for creative work in April. And then we raised our Series A in July.

Awesome. What does that mean?

Yeah, so I believe, well, one, I'll give a little bit of backstory. I used to wanna be like Nigahiga back in the day, if you guys know that Asian YouTuber that hit the 1 million subscribers for the first time. And then I kept telling myself like, oh, I need to do this. And my dad used to be like a wedding photographer or like a videographer. So I wanted to do, I wanted to learn Final Cut Pro and I just couldn't. My team is like going over the computer. But anyways, yeah, I just figured, you know, with the era of generative AI and people being able to build software like they've never been able to before, like that shouldn't be the case for creative work either. And creative assets, videos, the coolest motion graphics, animations, or bring your imagination to life. And, you know, with generative media models making it much cheaper than ever to create, like, but it hasn't been easier. And, you know, Melius is gonna be the harness to actually close that gap, uh, by building a suite, a creative suite of tools like the node-based canvas, agentic timeline editors, uh, even one-off generations all the way to being first to the MCP and CLI.

Very cool. So, so who are you guys? What kind of companies are you working with today? What's the focus? Are you focused on like certain types of assets? Like could be landing pages?

Yeah, we, we're primarily carried catering to a lot of businesses like in fashion, CPG, e-commerce for sure. Some notable customers are doors.com, Shown Media— shout out to the marketing agency in New York— and Skinny Dipped as well. We also provide— yeah, we also provide services for companies doing even more than $5 billion in ARR. So super grateful to be working with the group of customers that we are today.

More than $5 billion. It's time to start guessing, John. No, I'm kidding.

I want to know about $25 million. I mean, huge congrats, but it feels like you're going a sort of inference-light model. If you're going to be doing generations, you'll probably pass that cost through to the user and rely on other inference providers. Is that the thesis? Or do you see, yeah, at some point we're going to be training our own models, we're going to be doing crazy stuff, building hardware, racking servers?

I love that you asked that. I don't know if you guys had a chance to check out some of the press release stuff, but the theme of the fundraise is that we're building this so that anyone can make anything.

Yeah.

And we want to kind of leave an open-ended message there. Like, we're going to start in generative media for sure, but eventually it's going to be a play where, you know, like, I hope 10, 20 years down the line, people have like— this is a crazy thing to say, but maybe they have a microchip in their brain and then they just think of this something that they want to make and just materializes in front of your face. Like, that is the kind of thing that we want to strive to build. And we're not going to make any promises, but like, you know, as long as we have a goal that says, you know, we're going to build this company so that anyone can make anything, that is the right strategy, I believe. And this is just the first start.

First start.

I mean, it seems like at this point, a lot of the a lot of the customer base is essentially sales-driven or landing page-driven, news-driven. Are you optimistic about any of the foundation model plugin ecosystem driving, being the tool of choice for agents at some point being a growth vector? I've seen a couple companies go really, really vertical article off of that. But of course, there's always some trickiness to the relationship there. But how are you negotiating AI discoverability or AI top of funnel?

Yeah, I actually think that we were pretty— I guess everyone would say this, but we were pretty early to that take, especially in the generative media space. If you go back to March, we were still in stealth. And our first post was that we built first MCP to interact with generative media models on a canvas, and no one had done that before. And we positioned ourselves as the creative canvas for agents.

Sure.

And, you know, that had its own kind of like bipolar reaction from the market, I guess. But yeah, I mean, I totally agree that that is going to be where the trend goes. That is definitely where, you know, creative work is trending. Most of our customers are using Claude and ChatGPT plugins with the MCP at Melius.

Sure.

In addition to that, even the self-service market, it's growing at a very rapid pace.

What is demand for computer use like if somebody has successfully learned Final Cut and they want to stay there? Do you think it's important for you to offer an integration point there so they can sort of like watch the cursor do some of the work and then they can still jump in with the tool that they know and love. You can also edit the underlying files and then just vend them a project file.

Yeah, that's actually—

that's exactly it. Do you want to do it all or do you have a particular pattern that works?

Yeah. We're never going to be that walled garden that never lets people, like, do parts of things on different platforms. You know, I think plugins are really great, especially in this era of, like, multimodal computer use. You know, we're going to build the plugins for maybe After Effects or Adobe Premiere and whatever it may be. And we're also going to support all those features on our platform. The thing is, the reason to stay on Melius is that we built business conducive and long-term creator inducive workflows. That means like proactive agents, reactive agents, the timeline editor, the node-based canvas, and the digital asset manager. These are all things that like maybe people don't have out of the box on their computer. But if they don't need it, that's fine. They can actually go and use our MCP coupled with their plugins on their own computer when they are using agents with computer use. And so, you know, there, there's some value to using Melius on the platform, but we definitely are gonna capture a ton of usage off platform through MCPs, through plugins and, and, and tools of that nature.

Do you guys have a, like an agent or a workflow that can do this yet? Which is, I think it would be very powerful if a brand could, let's say, pick 5 to 10 companies that they look up to or they like their creative approach and say, anytime this company has a new ad in their Meta ads library, make 5 variations of— basically clone it for my brand, which is something—

Swap it up.

No, I don't think this is exactly what happened.

Yeah.

Like Skinny Dipped might see another company that, you know, in a different category with a, with new creative and, and marketers today are just going and looking at what their competitors or, or, or other companies are spending and just cloning it.

Yeah.

Yeah.

That's actually a common workflow, especially with the Claude MCP right now.

Sure.

What people are doing is they're having the TikTok MCP or TikTok Ads Library hooked up. And then every morning they set up a a Claude project or an automation to go look at the top trending ads and they'll just go dump the script and the transcript and all the keyframes into Malleus and then just reverse engineer everything. And so, yeah, I definitely think that we are conducive to that workflow and it's kind of scary what you can do with the tooling of today. But yeah, 100%, we can support that right now.

Interesting. Last question. What's your stance on taste?

Taste?

Oh, this is a hot take actually. I don't think there is a— There is taste when it comes to like UX. I think UX is like obviously you have to make it conducive to how a user perceives like the screen and like what the user experience should look like. But I don't think there's such thing as slop in terms of creative content. You know, there's a reason why Tung Tung Tung Sahur has billions of views on the internet. Yeah, maybe it's like considered AI slop, but like people find it entertaining. And like who's to say that you decide what good creatives are. If people find entertaining and they want to watch it and they find joy in it, then like, who are we to stop that? And which is why we want to say like, hey, you know, we don't want to bottleneck you on these like workflow constraints or these resources. If you have imagination, you think it's funny and you want to make it, go do it on Melius.

Love it.

Yeah.

Taste, like if something is tasteful or performant, like they, they can sometimes be the same thing, but very often they're not.

Yeah.

Yeah, at the same time, some of these AI slop projects are tasteful in the sense that, like, the person that created them or had the idea for a particular character, they exposed their taste in coming up with the concept. And then, yes, it is, you know, slop generative AI video that might have odd, you know, bugs and hallucinations, but they are expressing their taste and like, I think people want to see this cat dance or whatever.

Yeah, of course. Do you guys, uh, do you guys, uh, do the hotel lobby, um, trend a couple weeks ago?

Yeah, that's the Offset and Quavo rapping. Oh yeah, yeah, yeah.

I mean, do you think that was tasteful? I thought that was fucking hilarious.

Yeah, yeah, I, I thought a lot of that stuff was, was funny in that—

Wait, what did you just say?

Sorry, sorry, I, I asked you guys before I was like in the email. I was like, hey guys, is there something I shouldn't say? And I didn't.

Yeah. Okay, okay.

Anyway, congratulations to you and the team.

Congratulations on the fundraise.

Congrats to you guys too. Only only you know only maybe ten more raises to eclipse your former employer. But you know you're on the wrath. You're on just just two x every. Month forever and you'll, you'll be there.

There we go.

Right. No, I respect them. They did an amazing job. I look up to them for sure.

Well, have a great rest of your day. Congratulations. And we'll talk to you guys.

Goodbye.

Let me tell you about MongoDB. What's the only thing faster than the AI market? Your business on MongoDB. MongoDB. Don't just build AI, own the data platform that powers it. We have Rick Dungey from the National Christmas Tree Association, the executive director. Director coming on TVP. How you doing, Rick? Welcome to the show. There are 79 days till Christmas, and around here we absolutely love Christmas, and we wanted to talk to an expert. And so I would love for you— thank you for taking the time— I would love for you to introduce yourself and the organization, uh, and then I'm sure we'll have a ton of questions. But let's start with an introduction on yourself.

Uh, Rick Donche. So I've actually started working with Christmas tree farmers in 1998.

Whoa.

Yeah, no, um, so I've been around. I can remember going to my very first, yeah, um, conference, uh, of tree farmers and did a lot of farm tours. You know, it's a steep learning curve. I didn't know anything about it. I was a, I was a PR nonprofit management guy. But once you start going to tree farms and you hear them talk about how, what they got to do to actually grow trees that you put in your living room. You know, it's kind of cool. There's guys out there swinging big knives, and they got tractors, and they got trees growing in places you wouldn't think they could actually grow a crop. So, uh, yeah, it's been a steep learning curve, but I've been around them a long time, and, and it's quite a fascinating thing to grow Christmas trees.

What are the different categories of, like, growers? I'm, I'm well, well aware— I, I grew up going to a Christmas tree farm that was half a mile from my house, and they grew all the trees there. And then I moved to the big city, and I started going to places that sold Christmas trees. And I was like, wait, I felt like Buddy the Elf. I was like, wait, you guys aren't growing Christmas trees here. What's going on? Shipping them in from, I'm sure, all over North America. But I would love to understand the shape of the market.

So according to the USDA, there actually are Christmas Mm-hmm. Now, a lot of those are going to be small. Many farms are what we call cut-your-own or choose-and-cut, like you described, where you actually go to a local farm and cut down your own tree. That accounts for about 18% of all the trees sold in the U.S. All the rest of them are pre-cut, so they're grown in larger systems, larger farms on more acreage, and they're cut and sent to the stores. to other retail markets and sold as pre-cut trees. We call that wholesale. So there's either wholesale or cut your own. Those are really the 2 main types of farms that you can—

And for the pre-cut, where, what states or regions are those most commonly grown? Like, is there a place where we could fly a plane and just see Christmas trees for as far as the eye can see?

Yeah, Oregon, the Northwest. Oregon is the number one producer. They, they harvest about 6 million trees annually.

Wow.

In the state of Oregon. North Carolina, the western mountains of North Carolina, once you get up over about 1,000 feet of elevation, they have big farms. So the Appalachian areas of western North Carolina, western Virginia, and in all the Great Lakes states— Wisconsin, Michigan, Ohio, New York— all the way up through New England and Maine. Those are the main big production areas. That's where you're going to find most of your bigger wholesale farms that are selling And what type of work do growers need to do around demand planning and forecasting?

I know it feels like there's, in the same way that you see trends in fashion, there's trends in Christmas trees, but I don't know how much it fluctuates.

There was a war on Christmas recently. So that probably affected your hand. But I think we won the war.

We won.

In, in some people's minds, I suppose that's true. But, um, yeah, so the thing is that trees are grown like as any other crop. You know, you plant them in the spring and you harvest them in the fall. Um, really the main difference is you don't get an annual yield. So what you plant in the spring is not ready for harvest for anywhere from 4 to 12 depending on how big you want it, where you're growing, what species you're growing. Some species of conifers simply grow faster than others. In the South, some of the southern Christmas tree farms, they can get a tree up to 6 foot in like 4 years.

Oh.

But on average, it takes about 6 to 8 years to get one up to 6, 7 foot, which is the most common size. So that's really the main difference. And so as a farmer, in addition to all the other things that every farmer has to be an expert on soil chemistry, entomology. You got to be a bug expert, a soil expert. You got to understand nutrient levels, all these sorts of things. But you also have to be really good at managing your inventory, because what you put in the ground is not going to be ready to sell until 6, 7, 8 years down the road. And so you got to be thinking about what does the market look like 8 years from now?

Yeah, that's like—

How many trees I'm putting Yeah, I can imagine.

I don't know when it happened, but at some point, these Christmas trees that are sort of sparse, they're leveled out, I forget the exact name, became a lot more popular, it feels like, in the last couple of years. You're talking about Doug Fuerst? I don't know. Anyway. I'll try to find the exact name. But either way, it's very demand planning process where you're having to predict out 7 or 8 years in the future is almost impossible.

Not just how many American households will be buying Christmas trees, but also which ones they will be leaning towards. Yeah, that's another demand planning thing.

And then how lean are these farming operations? I imagine if you're making a product that effectively you're only selling 1/8 of your crop every single year, they have to be staffed in a pretty lean way.

Uh, yeah, so most of the smaller farms that are gonna sell trees, you know, one at a time, cut-your-own style, many of them have also supplemented revenue with other types of revenue streams. Agritainment is sort of a catch-all phrase. So, um, you know, they sell pumpkins in the fall, they sell hayrides, they've got all these other things and activities. So it's really more of an all-day-long entertainment kind of a thing. The tree is just also you happen to get while you're there. So those types of farms have really been able to maximize profits by finding other revenue streams. For a wholesaler, it's about scale.

Sure.

Yeah.

Um, so, you know, and shipping is going to impact your margins, of course, um, because you have to get trees from where you're growing them to where your retail customers are. Um, and some of them sell trees just on Now you might know the right term. So if you have a scanning system, you put a barcode on there, the retail outlet will scan a tree as they sell it at POS. But they only pay for the ones that actually sell. So any trees that don't sell, the farmer doesn't get any, any revenue for.

Yeah.

So that can be tough also. So it's really a matter of scale for most wholesale farmers, finding the right retail buyers that have the right type of market, they're going to be be able to effectively sell the trees that you're growing and you're sending to that retail customer.

For many, many decades, your organization has supplied the Christmas tree for the White House Blue Room. What goes into that selection process? I imagine it's one of the highest stakes, highest pressure kind of moments of the year presenting that tree. There's a lot on the line going into that. You know, selection.

Yeah, um, interesting timing because we just had a Zoom meeting this morning with the staff from the visitor's office at the White House about what day, when it has to be there, and you got all the process you got to go through and anything. Um, yeah, so I try to make sure and point out that, uh, not to put too fine of a point on it, but it is really good for our industry, for our growers, to be able to tell Hundreds of millions of people every year, that at the White House, at the people's house, their decorations center around a farm-grown real Christmas tree, not a plastic toilet bowl brush from China. I think that's a really important thing to point out.

I like taking shots.

Shots fired.

That's great. Love it.

I didn't realize that we, we had an enemy here. There is a war on the plastic toilet brush tree. We can't be I like this. This is great.

Yeah, yeah, we like to promote the actual real thing that farmers grow. Yeah, so yeah, to win, to win that contest is hard. It's a national tree contest. We hold it every other year. Growers actually bring a tree from their farm that is the best example of a tree that they can grow. They have to win their state contest first to qualify for the national contest. The national contest has a 2-stage scoring system. It's hard. It's really difficult to win. By the time— if you become a grand champion and get to supply and present the official White House Christmas tree since 1966, so we're on a 60-year tradition now.

Wow.

If you get to that level, what you can do is you can tell people, I am an expert at what I do, and there's no question about it. It really is sort of the pinnacle of being and having the knowledge skills and abilities to grow a really good Christmas tree.

Do, uh, how early is a grower getting the sense that they might have a championship tree on their hands? Like by year 3 or 4, are they kind of feeling like, hey, this thing's on the right trajectory, I might have, you know, a real contender here?

Yeah, you get, you get a wide variety of answers, but a lot of the growers that I've talked to that do compete, um, will say that they pick out about 15 or 20 potential trees on their farm once they're at about 4 or 5 feet tall. Once they're that tall, they start shearing, you know, and trimming the branch of the tree and making it grow the way that they want it to. But also by that time, by the time it's about 5 foot tall, it starts exhibiting the characteristics, the traits that you're looking for— straight branching, fast upward growth, good needle coverage around all sides. So yeah, a lot of them will start identifying potential contest trees as early as 4 or 5 years before. The trees that are in the contest have to be between 6 and 8 feet tall. So yeah, 2 or 3 years out, they start identifying potential contest winners that they can take.

Amazing. I figured. Um, what is— how, you know, we're headed into the Christmas season for us here on our show. We, we started our, our Christmas countdown You know, weeks ago. But when is like the optimal time to head to the farm to get your Christmas tree, whether you're going for a precut or your local farm? Like, I imagine there's a window where supply is starting to arrive or they're kind of opening up the good stuff to customers.

I feel like the whole don't get your tree until after Thanksgiving is kind of maybe fake news. I don't know.

Well, the real Christmas— The real Christmas enjoyers know that, that— but, but too early I think could be a bad thing.

Okay, well, let's hear from the expert.

Yeah, I mean, um, there's going to be a wide range of time that a cut harvested tree can maintain its moisture level and its look good and not drop a bunch of needles once it's inside of a home. Part of it is where it's grown. You know, trees that come from the north are hardened off. They go into a state of dormancy long before they're harvested, and so they can absorb a lot of moisture. I mean, I've had people constantly tell me that their trees— their tree stays fresh all the way through February. I would probably stick to 4, maybe 5 weeks at the most for most people, and that's even if you're taking really good care of it.

We're taking it to the test. We gotta put it to the test. We're gonna, we're gonna nurture it. We're gonna keep it alive.

I was thinking of going this week. I was thinking of going this weekend where we've been having, we've been having too much fun with it. Tree, tree, actual tree care once the tree's installed in your house. I, I've never known if it works that much. I feel like just respecting the tree, giving the tree some love, uh, that feels like it goes a long way. But how much should you be supplementing the tree, the water, all that stuff?

2 basic things, really. It really boils down to 2 simple basic things. One, you want to cut a small piece off the trunk before you get it in water. Quarter inch, half inch, it doesn't have to be that much. And the reason that you do that is it opens up the plant tissue that absorbs water molecules.

Yeah.

If it's exposed to air for long periods of time, it won't be able to absorb water. And then just simply add water to it every single day. Water, simple water. You don't have to add anything else, just simple water.

Okay.

My team just reminded me that it's October 6th, and we're maybe a little bit ahead of the curve here. But I do want to ask you about what you— Jordi just had to leave. I do want to ask you, like, what What questions are you getting from farmers as the head of the National Christmas Tree Association? Like, are they pushing you to negotiate tariffs on imported plastic trees? Like, what— at what level can they engage with you as a liaison to the government, to other businesses, to each other even? What else is top of mind for farmers? For your organization, what value do you bring to the community?

Yeah. So, our main mission for the members and for the industry is a role of protection and advocacy. And sometimes that does actually involve— we work on issues like ag wage rate, how much you got to pay laborers on the farms. We do work on tariff issues occasionally. Recently, the 2 main importers of fake plastic green toilet roll brushes asked for an exemption on their product from the US Trade Representative.

I feel like nobody's getting an exception. I feel like everyone's paying crazy tariffs now. That's ridiculous.

Well, what was crazy is their reasoning for it. They said there's no alternative American product, so they have to import them from China. And our answer to the US Trade Well, actually, there is an American alternative. It's a real tree.

Yeah, it's a real tree. Yeah, come on. Yeah, that's a great response. Okay.

Um, and we were successful in that. They did not get an exemption from the Section 308. Yeah, so lots of those issues. And, um, I would tell you, you mentioned growers communicating with others. I, I've been involved with other ag industries throughout my career also, and one thing I can tell you about Christmas tree farmers is they're really, really good at communicating with each other, at sharing knowledge, sharing expertise, sharing experiences. They don't— even though they're technically competing against each other, you know, they're all selling trees and they all want to sell trees, but it's more like they get together at meetings and they're very open about telling each other, here's what worked at my farm, here's what didn't work. Yeah, they're very good about sharing knowledge so that they all become better growers.

That's really cool. It sounds like a fantastic community. I know we're having fun because it's October and we're very excited.

How can people, uh, how can people help? Is there like— is it all funded by industry?

Buy a real Christmas tree.

Well, yeah, that's obviously number one, but our audience doesn't need to hear that.

But okay, sure.

Um, is there— is it all funded by industry or, or can people donate? Oh yeah, how can people help?

Um, yeah, so we're a C6 organization, um, so we're a membership association. Members pay Okay. To belong. Um, the other thing that I would mention, if I can get a shameless plug in here—

Please plug away.

The Tree Association back in 2005 started a charity called the Christmas Spirit Foundation, and the main program that that foundation runs is called Trees for Troops.

Yeah.

Where we donate Christmas trees every year to military bases to give out to families in the armed forces. Uh, treesfortroops.org, you can donate to that charity. That's a C3 charity. We are now entering our 22nd consecutive year. In 21 years, that program has given out more than 347,000 free Christmas trees to military bases all over the world.

That's fantastic. Treesfortrees.org. Go make a donation today.

We will be making a donation as individuals.

We will be getting matching donations from employers. These are more popular than ever in tech. So Do everything you can. Thank you so much for taking the time to come chat with us.

Having fun.

And we will be— we're keeping the Christmas spirit alive in technology. And thank you for breaking this down for us.

We want record Christmas tree sales in Silicon Valley this holiday season.

That's right.

Across the country, but everywhere, really. Thank you so much for taking the time.

Yeah, great to meet you, Rick.

That's what I like to hear.

All right.

Goodbye.

Cheers. Let me tell you about CrowdStrike. Your business is AI. Their business is security. CrowdStrike secures AI and stops breaches. We got Dylan Field, the CEO of a public company, coming on.

We've been keeping him waiting.

We've been keeping him waiting. Uh, we're very excited to have Dylan Field from Sigma, of course. Welcome back to the show.

Here we go.

We were having a lot of fun with the head of the National Christmas Tree Association.

I heard part of it, and it was funny because I was just thinking that Christmas came early. We launched We launched Figma Design Agent today.

So, pretty excited.

There you go.

Thank you. That is fantastic.

On that, I think that's appropriate to end. Figma Design Agent.

Presents are under the tree. Okay. I think we have enough Christmas nonsense. Tell us more about the product launch. Take us a click deeper, please.

I mean, it's been awesome. We've been working on this for a while. We've had the beta going and just been hill climbing. The team has done an incredible job.

Yeah.

Of not just working with our own first-party models.

Yeah.

Frontier models. But also really thinking about how do we make this great for our users.

Yeah.

And so much more to do. But today's launch, you'll see that the Figma agent is faster, higher quality. It has better ability to deal with context across your files.

Yep.

And more overall agentic capability.

Yeah.

I would say it's improved quite a bit on design systems. We still have more room to go on all the dimensions. Definitely design systems, too.

Yeah.

But really proud of the launch and finding myself really enjoying using the Figma agent on the canvas. And I think as we think about just kind of where it all goes, there's so much that you can do in the canvas. It's not the case that, you know, the non-developers, the designers, the PMs feel at home in the terminal and the IDE. I don't think text is the final representation of everything.

Hmm.

I love text. I don't think that that's it. And I believe that we're going to find more ways to go build software and use these amazing reasoning models to be part of that process. It's humans, agents together.

Are you generally, through this process, I mean, you're building an agent that is going to excel at one discipline and not be an AGI model, right? And so I'm wondering if you have a broader philosophy around ensembles of models. I mean, we've seen the Decisions API, JEV, There's smaller models for different purposes. And you can imagine this future where you're actually using more models, more models. I've been— for a long time, I was like, actually, let's get rid of the model picker. Just give me one thing. And maybe that happens, and there's an abstraction layer on top of some of these tools. But it does feel like there's a ton of value in having a team, research scientists, AI researchers focused on a particular discipline. shine the Eye of Sauron on something and go and deliver a great product that then can be the right tool for the job. But are you in a world where you're applying your lessons here to like the shape of AI as a tool broadly, or is this a unique design thing?

Superintelligence, it's, you know, it's the goal. I'm joking. No, man, we're, we're trying to make like the best design agent.

Yeah.

To empower designers. We are not trying to replace designers.

Yeah.

We're trying to make it so they are maximally empowered.

Yeah.

In their work, which does mean helping them make PRs.

Yeah.

Helping them go even further with their work, helping them get all the context they need to go make decisions.

Yeah.

And it does so happen that like after all this time building Figma, the team really does understand the needs of designers.

Yeah.

And what we need to do to go achieve that. And I think that is just as much a part of the special sauce over time.

Yeah.

And so, yeah, I'm very, very excited about what we can do here and also how the loops that we have in the product and the data available to be able to work with, in addition to just kind of like this very, singular focus of how do we make the best thing for design and the best thing for designers to use every day and make them more efficient. I think that is really nice to have that focusing power. And also, you don't want to be in this mindset of you're fighting against Bitter Lesson. You want to get the advantages of the general models, the ways they've been trained on all sorts of things, associations they've made, and you want to make it that you're really good at design.

Yeah.

No, that makes a lot of sense.

Yeah.

I mean, it feels like part of the challenge as a design company is you don't have the luxury of being able to produce slop even at anywhere at the same rate as some of these other companies where they might be offering a bunch of things and their whole brand is not built around design quality and taste and precision. And so, they can just make something that like only 5% of the time it's actually like decent and that's enough. Whereas I think for— so, I guess like, do you feel like you have to hold the team to a higher bar in terms of what products are shippable? Because— Like a new YC company can be like, hey, we made a design agent, you should try it. And if people even have an okay time with it, they'll be more excited about it. Whereas your community, again, has much higher than average taste level, capabilities, all these different things.

I think there's a bunch of points in there and it's like, they're all interesting. Maybe we'll start with just the overall setup that we have and what we're really caring about for this product, which is we don't— again, we're not trying to replace designers, we're trying to empower them. And as we think about that, what does that mean? Well, it means that as you're trying to get something in your head, uh, onto the canvas— we've talked about this before—

Mm-hmm.

You might be aided by prompting. And also it might be the case that that's just the starting point. You might have to do more work and that requires more integration with a global system of Figma. And so you kind of have to teach the models Figma first and make it so that they can understand how to go and work with Figma as well as do good design work. But I think that right now, you'll probably see just as much on the slop side across all products. And You know, in AI. I hope you'll see less in Figma. We are working to make it so that there's less and less and less and better and higher quality results. But also, it's— taste is a wild thing in that different people in different domains have different definitions. And so the ultimate is how do we make sure that we can understand what you're going for and give it to you?

How are you thinking about interfacing with AI as a top-of-funnel? I think everyone saw the Blender moment when everyone was modeling everything in their world in Blender. And the AI community and AI enthusiast community was downloading and installing Blender. And it feels like there's immense value if you can be a button there, or even just the top-of-funnel agent says, hey, you've been working on this business idea. I did a first draft of a logo, but you should take this much further. Why don't I hand off to Figma, get you set up there, even pass along credit card details? There could be a really deep integration, but then there's always difficulties on where the value accrues and whatnot. How have you been— dealing with those potential trade-offs in an increasingly AI agentic top-of-funnel world?

Well, first and foremost, we want to always be learning from the ecosystem. And I think that's essential. The ecosystem sometimes in these moments evolves so quickly that when you start to see everything come in, it's like, wow, there are so many ways that we can work with with other companies or other models or different setups here, sometimes it's the case you're like, wow, yeah, I predicted that like 5 years ago. And other times it's more of, I did not predict this at all. This is a second or third order effect.

Whoa.

And I think that if you don't understand what people are trying to do, you can't build for them. And also you need to work well with the ecosystem in order to make make all these new routes possible. And there's complex questions, like you mentioned, around where does value live, where does it accrue. I think that some of those, there's a category of questions where it's, there are not clear answers, and it's not clear there will be clear answers. There's other categories where I think that actually the way the space has evolved, will evolve, is more clear. And so you also have to have, I think, some macro bets about the way things will play out over time, but not get stuck in your head.

It's this weird balance. How are you thinking about motion graphics? I feel like we've seen some cool one-shotted demos. I've done enough motion graphics in After Effects to know that it does need to be pixel perfect, and sometimes you need to make small changes. And doing that via text would be extremely cumbersome. And so it feels like there's a need increasing opportunity there for web-based, Node-based, AI-enabled, MCP-enabled tooling. What's the current state of the strategy there? How do you see the motion graphics discipline evolving?

I think that the way that we'll see motion graphics evolving is parallel to all these other areas too, which is when you're facing a blank canvas or a blank page, it is hard to get started.

Yeah.

The sooner that you can kind of build to think, get something out onto the canvas, whether that's through design and, or it's even, you know, in a sketchbook or you're doing something with a coding model, that is good. It helps you get over that initial hurdle. And then from there you can react to it and go, okay, what do I want to tweak? What do I want to change? How do I make this mine? And how do I make this my vision? And software ultimately is just amazing creative medium. And we don't want the center of the distribution for everything. We don't want everything to converge to the mean. Rio, who was previously at Cursor and recently left after they were bought by SpaceX.

Yeah.

He had this amazing post recently where he talked about some of these ideas. And I think that Weave is another area For Figma, where we're really trying to figure out how to help people do this, both use models, get all the power from them, but still apply their own judgment, creativity, taste, uh, and really shape it, the model outputs like clay, rather than just trying to go for the one shot. The one shot's rarely where you want to land and end up.

Give us your design review of the new American passport. Do you like the previous version? or the current version more? Which passport speaks to you?

The new version I really like. I think that, you know, beautiful cover. And then I love the way that as you flip through it, the eagle animates.

It's like a flipbook.

We didn't even get to that. Yeah, there's a flipbook involved, which is interesting.

I hope I'm not spoiling something.

No, no, no. I did see that, but we didn't— we were just looking at the front. And the front seemed like a very moderate iteration. There was some debate on our team about the kerning on the word Passport. Tyler's not a fan of it.

This is too controversial for me. Let's go back to AI.

Yep.

Where do you— feels like people on the timeline have been doing Figma's marketing for you in some ways, where every other day there's somebody like, if you send me an AI-generated slide deck, I won't even take the meeting. Oh, yeah.

You said that.

No, I've been saying that for a long time. I've been saying that for a long time, just because I think you should be thinking about if there's a sea of just sameness, being able to prove to people that you can think differently, stand out, have a unique point of view, opinion, all these different things. in a sea of AI-generated decks, I would almost rather see, like, a Google— clear, obvious Google Slides template because that is signaling, hey, I don't actually care about design at this very moment, but I care about showing that I communicated, you know—

Words.

Very clearly what I'm trying to do. And the problem with AI-generated decks is that every single slide I'm second-guessing how deeply did you think about what you're communicating. Because I don't want to spend time trying to process and evaluate somebody and their thinking if it's possible they weren't thinking at all.

I think that's a great way to put it and have really been kind of on this thread myself with our team. At some point recently, I kind of gave some guidance to the team around AI writing, for example. And I thought about it a lot. It was a post that I kind of had queued for, I don't know, 3 weeks or something. And I kept rewriting it, ironically, because I thought this is really important to communicate. Like, we can't just ban AI writing in the company. Then we're not empathizing with, like, everyone else that's in tech.

Sure.

And everyone else that's, like, using our software. At the same time, I think that there's degradation of trust that can happen as you have all these folks interacting across the company and people are second-guessing. Did you even read what you're asking me to read? Did you think through this?

Yeah.

And because of that, I came up with this whole sort of rule for Figma, which is let's just have it be that you put written with AI at the top of messages.

Oh, yeah.

Where there's AI writing of any kind. And if you're using AI to spell check, essentially proofread, and you change a word, I don't care. If you write the whole thing with AI or there's a good paragraph in there, yeah, just put it right with AI. It's easy. No big deal. And that way we can trust that if you're sending something, you care about it and you've thought through it. And I think that in general, people should not give up on their own thinking. They cannot give that to the model. And I see so many people that are essentially looking for the shortcut, trying to make it so that they can have maybe less in Figma, more outside of Figma. But folks are eager for the model to automate for them, for them, the model to think for them. And I think that the more important thing you can do is to actually use the model to learn, to try things you wouldn't have tried before, to be more daring, um, and to kind of expand possibilities rather than contract them.

Yeah.

Uh, and I don't think— I think it's good to have default skepticism, not because models are all bad or something. That's not what I'm saying, but rather because if you're default skeptical, then you will understand better how models work, what to use them them for and what not to use them for, and you'll get better results. So, yeah, I'm all about thinking.

Yeah.

I've noticed in the process of making decks, like my best decks, always the outline took like 20 minutes max where I just like actually had clarified, like you want to— there's increased clarity that comes from trying to communicate an idea or a company or a product or launch or a campaign, you're gonna naturally get more clarity on the topic as you explain it in a way that is sort of logical and has a consistent narrative. And then there's times that I've opened this sort of my deck outline in Figma that I've shared at this point with over probably 100 founders, right? Because I spent so much time on this. I have example slides for each slide and And structuring and all this stuff. And there's been times I've landed in that and I'm just wrestling with it for hours and hours and hours, not making progress. And 2 things are usually happening. One, it's just a bad idea. Or 2, I don't understand it. I just don't understand it well enough, or there's just something fundamentally broken. And so, when people go to AI first for something as important as communicating your company, you're robbing yourself of, one, that sort of check and balance of, do I actually understand this deeply? Does the thing that I'm presenting make sense? And you're sort of accelerating to the next step, which is pitching the thing. And the last thing you want to go do is pitch an idea to a bunch of smart people that is not well thought through, that doesn't make sense. And so, you're Saving that time of wrestling with the idea, but then wasting even more time and wasting other people's time trying to communicate something that doesn't make sense. And so—

I mean, if you can't explain something, you don't know it yet.

Yeah.

I think is the basic rule I always have gone by. And that's pre-AI too.

Well, I don't know it, but my assistant knows it. And so you can just talk to the model now.

You don't even need to talk There's also the flip side of like— there's also the flip side of not just the, do you understand this? Can you explain it? But also like, what are the things that no one knows yet that we all need to explore? And, you know, that can't be fully understood. And I think that mystery in the world and sort of like the ways that you feel just When you encounter something that's really novel, new, or has an angle on something that exists, but just is fresh. Those are magic moments. And I think that in general, there's an opportunity right now to lean more into mystery in a time when everything's becoming a bit less mysterious.

Yeah, that's interesting.

Mic drop.

I love it.

Thank you so much, Jordan. That's a good place to end because I have follow-up questions to that encouragement of mystery, but you can just leave it in this sort of dramatic pause.

Yeah, you'll leave it mysterious.

Leave it mysterious. Thank you so much.

Great to see you, Dylan.

Always great to see you.

Congrats to the team on the launch. Cheers.

Have a good rest of your day. We'll talk to you soon. Goodbye. Uh, we have the founder of Vinci and the And the CEO, founder and CEO, Hardik Kabaria, is in the waiting room. Let's bring him into the TV studio.

How are you doing?

Hey, hey, how are you guys doing? I'm doing well.

Welcome.

We're doing great. We've had a wide-ranging show today. Nice conversation with Dylan, head of the National Christmas Tree Foundation as well. It's been all over the place.

Did you make it through that interview or not?

You actually broke. It took me to deep water.

One of us is a professional, and the other one's an amateur.

As it gets closer to Christmas, go watch our interview with Rick, the head of the National Christmas Tree Foundation. But today, we're here to talk about your business. But why don't you introduce yourself, and then we'll get into the company?

Sure. Yeah, my name is Hardik. I'm co-founder and CEO of Vinci. My background's technical physics and geometry software. I've been in that space for some time and started Winchy 3 years ago.

Cool. Yeah, PhD at Stanford and 8 years at Carbon. What were you doing at Carbon? What— actually, I'm not— there's a few businesses I wanted to know more about Carbon specifically and how it set you up for this.

Yeah, Carbon's a 3D printing company. I was— Sequoia Silver Lake-backed company. After my PhD, I was looking for something that can use geometry but will also closer to the real parts.

Sure.

So I joined as a software engineer. I was there for 8 years. Yeah, I was VP of software for the last couple years. So shipped software on manufacturing type, uh, manufacturing floors, um, in Asia, in Eastern Europe, in US, Mexico. So it gave me a lot of perspective on how software can help hardware.

Yeah. So, uh, narrow in on the current mission Semiconductor design, semiconductor engineering. I notice you're expanding into vehicles, aircraft, satellites, memory. Those are similar problems, but it feels like there's enough money to be made narrowly in semiconductors. So, I'm interested in the overall strategy and the level of abstraction you're operating at.

So, I'll start with something very basic, which is physics. Everything that is around us, whether there is human-made parts, nature-made parts, it'll be governed by laws of physics. Laws of physics are universal. Something as simple as energy in will always be energy out, momentum balance, and so on. Any part we design, we care about physics because ultimately the parts start behaving in the real world, and if they don't perform the way you expect it to perform, all goes to hell, right? What are we building is a foundation model for physics.

Okay.

And how it's meaningful. Let's say you create thousands of design ideas. If you cannot grade them with physics, it'll be very hard to pick which one's right. Right. So physics is the compilation layer for hardware.

Sure.

It's been manual. With our product, we are making it automatic and scalable.

Yeah.

Now this applies to semiconductor, high bandwidth memory. It also applies to satellite.

Okay.

But when the scope is this big, you gotta start extremely focused somewhere.

Yeah.

So we said, why don't we pick thermal problems in semiconductor and electronics and we build a software that allows our users to do it and ship it?

Yeah.

And, you know, make it true and then we expand. So that's where we are.

What is the sales process for a semiconductor company these days? Because I feel like they have something that's working so well. I thought the only job they had was make more of the thing that everyone wants. and raise prices, but they have time to implement new software solutions from third-party vendors? That's a good question, right?

So, you ask any— so, first, it's top-down sales. You cannot go to these large, large companies and say, I changed one engineer's mind. I mean, I would say good freaking luck, okay? You will have to convince a VP saying that, hey, this is something you should try and it's working for us. So, now to the second question, do they have time to try out? Like, who are we? I'm not even from semiconductor.

Yeah.

Why would they even give me a time of the day? I've never worked in semiconductor industry before. The problem is so acute that if you can really help them, it matters. So we start our sales process with asking what matters to you, and answer is consistently the same thing, time to market and performance. I do not want to miss my time to market. I have to launch a new chip, new product at CES end of the year, and it better be performing. And we work with the most complex parts people are making.

Sure.

High bandwidth memory.

Yeah.

CPU, GPU, right? And their thermal problems are so acute that the existing tooling doesn't take them there. So if you can really be more accurate, higher fidelity automatically, they are definitely interested in trying it.

Okay. So maybe a month ago at this point, we had the Navier-Stokes moment and it was very cool for people within the tech space. There was a lot of excitement. but average people, like, if you tell them we use a computer to solve a math problem, you know, they're not going to be super— it didn't feel like it moved the market is another way to put it.

Yeah.

My question for you is, like, how— are there a bunch of other moments like that happening at, like, a micro level within businesses where they're using models and they're having breakthroughs, but they're not raising raising their hand and saying, like, basically just publishing that because it actually is like, can become IP? Yes.

So I would say that it's, we have been deployed commercially for some time. So I'm gonna ask you a question from, I'm gonna start answering a question from there. So they would, in these companies, the IP sensitivity, who they work with, they are extremely private and secure about it. So they get started using a new vendor, new tool, new method, and they want to keep it to themselves. And that's the style of the industry, which, um, it helps us, it hurts us, but I would say we are part of the ecosystem.

Yeah.

So I like to go and meet customers where they are. So I would go far as to say, or even it's embedded in our deployment. We do on-prem deployment and we make our product work with zero forward deployed engineers, extreme data privacy. We'll meet you where you are, make sure our software is useful to you, and then sort of doesn't matter. What helps us is the industry is so small that everybody talks to each other, wink wink, works. That's fine by me.

Yeah.

On the data side, simultaneously, the laws of physics are completely open source, but the actual design of semiconductors It's not. It is extremely well guarded.

I personally discovered new physics with ChatGPT, but I haven't talked to anyone else.

Well, now it's in the training data. But you see what I'm saying? You have the most easy to implement, widely, look it up in any physics textbook. And then you have the stuff that is sent from NVIDIA to TSMC and is not making it to the open web. It's not in the pre-training data. Is there a gap to close? Are you hiring semiconductor designers to build RL environments for you? Like, what actually moves the needle? Or is just the laws of physics, is that enough?

So the laws of physics is enough, actually, because the laws of physics inside the firewall of TSMC is exactly the same as outside of the firewall of TSMC. I am yet to meet a company that invented physics, and I don't think I'll meet them. And we have trained a model on the laws of physics, and at the time of usage, it will, it will produce an answer that abides by it. So if that's the case, we don't need to know the designs that NVIDIA, Micron, TSMC, Samsung are working on. That's why we deploy and we make it work out of the box.

Sure.

Does it mean we have to— so we are today training on petabytes of data.

Yeah.

And we will go to 10x, 100x over the next couple of quarters. So what is the expansion? So the expansion is from thermal to thermomechanical to electromagnetics. So basically cover every physics modality that people can write equations about. So to the point earlier made, there is something new physics that somebody is going to discover. We haven't embedded that, but the discoverable physics is a freaking fair game.

Well, congratulations and thank you so much for coming on the show.

Hit the gong.

Oh yes, massive fundraise. $250 million.

Quarter bill.

Thank you.

Thank you.

Appreciate it.

Thanks for having me. Advent, Tomasek, and others. Wow. Amazing progress.

Coastal Ventures, AMD.

Yes. Only 3 years in.

AMD and Applied Materials as well.

Very, very good.

AMD and Applied, 2 corporate ventures.

Yeah.

Amazing.

Amazing.

Well, congratulations and thank you so much for coming on the show.

Great to meet you. And congrats Congrats to the team on the milestone.

We'll talk to you later. Have a good one.

Appreciate it. Thank you.

Cheers.

Thank you.

Goodbye. Well, I think that was a massively successful show.

Massive.

That was a hilarious interview. I learned a lot, genuinely.

And not, not, uh, Vinci.

Oh, yes. I'm going back to the Christmas tree one.

To be clear, I, uh, yeah, that was, that was one of the tough moments on the show for me. I was, you know, trying to look at the monitor.

The team was not helping.

Trying to look at Rick. The team was, was making my life— I was, I was in a, I was in a dark place. I couldn't keep it together, and I had to run off the set.

The team sent also—

And then Nick ran out behind me. He was, he was—

And the team sent this meme of the sitting And that was just the death of me at the time. It was rough. Uh, it was a good time, but we learned a lot. And that's not the end of our Christmas programming, folks. We got more Christmas-related guests coming on the show.

You didn't ask, we didn't listen.

Yeah, we got some good stuff.

No, it's actually fascinating.

No, it's so interesting. It's genuinely—

Being in a business where you're making something that you're gonna sell in 8 years is like an incredible challenge.

Oh, totally. Yeah. And, uh, yeah.

And competing in a market where there's a, there's a war on your industry.

It's true. Yeah, I mean, there really is. Once he kind of broke down the tariff perspective, it makes a lot more sense.

Um, anyway, the only thing I didn't get to ask is, do you still hear the bells? But I already knew the answer. Of course he does.

Of course.

Of course he does. We're going to be making a, a meaningful donation to their foundation, and we encourage you to as well.

Remember, that is treesfortroops.org.

This is financial advice.

This is financial advice. Do it now. Thank you anyways for tuning in to TVPN today.

It's been an honor.

Fantastic show. We'll see you tomorrow at 11 AM Pacific.

I can't wait.

Leave us 5 stars on Apple Podcasts and Spotify. Sign up for our newsletter at tvpn.com, and we will see you tomorrow. Have a wonderful—

goodbye.