A (3:09)

And what kind of people want to make $10?

B (3:12)

A lot of liars. There's a lot of liars out there who say, I am absolutely a CFO and please send that Amazon gift card right my way. Right. And so for this one, they actually did the work. They called up and ID verified and then also reality checked the position. So if you say, yeah, I'm the Chief Pirate Officer, I'm the ninja hero, whatever, you got some fake title, they're that you're out of the survey. You have to be a cfo, a CEO, senior manager, and you actually had to be doing that job. It's not just like, oh yeah, you're the CFO of some front company. So they did some reality checking and they pulled together 6,000 of these business leaders across firms that are domiciled in the us, uk, Germany and Australia. And so the line from John Collison that has been sort of going viral, that was. He dropped it on sources. I think he said it to us too. It's a good line. No one wants a refund on their terms. Tokens. Everyone is using AI. The spend is increasing.

A (4:12)

Although I'm sure some CEOs heard that and thought, I kind of do want a refund.

B (4:17)

I love a refund.

A (4:18)

I had one team member go absolutely Haywire and spend 50 grand.

B (4:23)

He's one shot. He claims that he rebuilt our entire erp, but I fired it up and it didn't even have HTTPs. What's going on?

A (4:30)

The Mac Mini wasn't even plugged in.

B (4:31)

Yeah, the Mac Mini wasn't even plugged in. He was just chatting. But clearly there is a disconnect between the stripe data is very real, the value creation is very real, the revenue is very real at the labs. But when just random joe schmo CFO CEOs get a call from the feds, they say, yeah, we're not really getting that much value out of AI. And so the questions that you need to dig into there's actually four key findings. The one headline that the New York Times is, is pushing is this 80% number. 80% report little impact or no impact on employment or productivity. But there's actually a bunch of positive signals. There's a bunch of mixed signals in here. So 70% of firms actively use AI, and particularly younger, more productive firms. Second, while over 2/3 of top executives regularly use AI, their average use is only 1.5 hours per week. And one quarter of executives report no AI use at all. They're just like, I do things the old way. I have AI, not for.

A (5:45)

Why would I need that? I have a telephone.

B (5:46)

Yeah. But I mean, truly, if you think about, like, the variety of firms, it's like you could be running a gym, you could be running a gas station, you could be doing forestry, you could be doing mining, oil extraction, road repair. Like, there's a million different things that you can do in the economy. It's not all like, knowledge, work, firm.

A (6:02)

We talked to somebody yesterday in the sort of later part of their career. Off the show.

B (6:08)

Yeah.

A (6:08)

And they had just had their mind blown by AI because they used to, when they needed a document created, they would put it into bullet. They'd put out bullet points and they would give it to somebody who then create a document.

C (6:20)

Yeah.

A (6:20)

And he just said, now I just give it to AI, and that just generates a document.

C (6:25)

Yeah.

B (6:25)

And so that.

A (6:26)

So that's still happening.

B (6:27)

Yeah, that's text expansion, text generation with LLMs. This has been available since 2022 when ChatGPT launched. Maybe it became reliable in 2023. People are still just starting to adopt. And there's some other interesting things in here. So the last major finding that we should touch on is firms predict sizable impacts over the next three years. Forecasting AI will boost productivity. Sizable impacts, productivity increase 1.4%. Which is like, it's very sizable if you're an economic researcher, but it's not particularly sizable if you're in, like, the fast takeoff scenario. And so there's just this disconnect between, like, what do the government statistics look like? And what is. What is the government operating against? And then what's the Silicon Valley narrative? And like, where do these two meet the road? Why is there a disconnect at all? One of the reasons is that measuring AI adoption is a mess. Many people use AI without even knowing that they're using AI because it's buried deeper in SaaS, products that they already daily drive. Like, if you're just, I run a coffee shop and I'm using Toast for, you know, payment processing. Like, there's probably some AI features in there already. And when you go to, you know, type in, okay, we're adding a new cinnamon roll to the, to the, you know, the menu. There's probably a button now that just says like, do you want to just generate an image of a, of a cinnamon roll? You could upload one still. That's probably a feature that already exists. But like, we could also just generate one for you and you can probably click that. But you're not like, oh yeah, I'm an AI power user. Just because, like you happen to use Toast and TOAST happened to have implemented some gen AI feature that like you haven't really dug into yet. So some AI isn't even detectable. You could be talking to a customer support agent on the phone that is AI generated and not be able to tell. We talked about that airline interaction that got something 100,000 likes and grace, the woman that.

A (8:26)

Grace, the woman that had the interaction came into the chat yesterday and said it was real.

B (8:30)

Yeah, it was real. Yeah.

A (8:31)

And so she outmaneuvered the clanker.

B (8:35)

Yeah. But still think about like, she's clearly on X in tech. Like very AI aware. There are probably tons of people out there that are saying, oh yeah, my job, you know, every once in a while I have to call this service. And now the person that picks up is like responding pretty quickly, but I haven't noticed, they haven't noticed that they're actually interacting with AI or using AI in some capacity. And then there's also times where people are like chatting with AI, but they just put it in the personal life bucket. Like I'll find myself doing this a lot. Of course, as an entrepreneur, like the work life balance like bleeds together a lot. But there's a lot of times when I'm, you know, reading an article on Saturday, I'll fire off a deep research report about it, find some extra context. It feels like, oh, I'm just reading the paper, but my job is sort of to read the newspaper. And so it comes into work and there's probably a lot of people that are like, you know, oh, you know, I hit an LLM with some random query to learn something about something work related. But I did it off hours when I was just like, you know, hanging out. And so I don't really think about it as like a work tool yet. And so they're not putting it in that bucket. But in general, I think the team did a really, really good job by avoiding as Many of the pitfalls as possible when it comes to surveying AI adoption. AI adoption is very messy. You've talked about the need for strong.

A (9:52)

Yeah, I still think there's room for a research firm focus entirely on diffusion. So if you had a group of 10 to 20 people that were spending all their time talking to business owners and executives, operators, and getting a sense of how they're actually using this stuff, I think you could put together some really compelling reports around it that would be pretty useful to everyone from AI companies to Wall Street.

B (10:20)

Yeah, Adoption max after cluster max and Inference max, they had to rename it. Apparently semi analysis can't use max for some reason. So they do. Inference max is now Inference X. And everyone was saying you need to just change it to inference mock, which would have been amazing. But inference X obviously has a much more professional tone to it. And so there's an interesting definition of like, what does it mean to actually adopt AI? That's very vague. This paper defines it pretty broadly. So machine learning for data processing. So that doesn't Even necessarily mean LLMs, that just means ML, which has been around for a very long time. Text generation using LLMs. That's what we think of as ChatGPT, visual content creation. So diffusion models, but also robotics and autonomous vehicles. And there's a category just for other. And firms can select multiple. And so if you selected yes on any of those, you go in the bucket of AI adopter. And 78% of firms in the United States said, yes, they are using AI by this definition, they got at least one robot or they've at least generated one AI image or one prompt to ChatGPT, which is a very low bar. Sort of makes you wonder like, what's going on with the 22%. And if you. And you can also dig in further. So text generation using LLMs is the single most common use case at about 41% of firms. So flip that around. 59% of firms aren't even using LLMs for text generation or proofreading. But again, there's a lot of companies where it's like, yeah, we don't generate a lot of text. Like, maybe if we need to generate a marketing material, we have an agency that does that. So we don't actually do it internally. I don't know. Across the four countries that were surveyed, 69% of firms totally said they currently use AI. I think Australia was behind a little bit. Dragging that down. Only 75% of firms expect to be using AI technology sometime over the next three years. So they're at 69% now and they're like, over the next three years, Tyler's

A (12:27)

going to have a heart attack.

B (12:27)

We're going to bump that up to 75%. And this is like, this is weird data. And you can jump in with your pushback whatever you want. But my point is not that they're right. I think that they're wrong to predict this. I think that the AI adoption will be very steep and very dramatic. But I just think it's important to recognize that this is a paper that people will be citing. This is a paper that will shape policy. This is a paper that does. It reveals some misconception about the impact AI is having in firms.

A (12:58)

Yeah, I still think it's just, it's so hard to like actually quantify this. Like, okay, if I'm like using ramp, does that count as using AI? Because it's obvious, like it certainly is using AI under the hood, but I'm not directly interfacing with the model, so does that count?

B (13:13)

So I think if you were running just a company that was just on ramp, you would probably respond, no.

A (13:20)

Yeah, but clearly I'm benefiting from AI.

B (13:21)

I agree.

A (13:22)

So it's like, how do we actually

B (13:25)

quantify this is the diffusion question. I completely agree with you. I think that there's plenty of places where AI will have impacts across the economy, but the actual AI workloads and AI app development, model training, inference will happen at a different set of companies. I don't think it's not going to be as clear as the computing revolution where every company had a desktop computer and then it was like, okay, I still think, very qualifiable.

A (13:54)

The only thing that you should be looking at for diffusion is just lab revenue.

B (13:57)

Lab revenue, yeah. And it's growing a lot. But it's still the perception, I think still does matter because people will. I think that there's a little bit of potential self referentialness here where firms see, oh, AI adoption's low, I don't need to go and figure out how to adopt it. And so that's something that I'm also keeping an eye on. Reported usage is still low and this one's interesting based on revenue and also token generation. So 1.5 hours per week among the managers surveyed. Again, it's like they surveyed CFOs. CFOs who use RAMP. They don't count their time in ramp as minutes using AI, they count their time in LLMs as minutes using AI. And that's low. But even just with that one and a half hours a week. The actual leverage that you're getting is increasing because one and a half hours, you don't necessarily need to spend more time prompting to get more done.

A (15:01)

Because even so, even if you run a deep research, like is the time you're waiting, does that count as time in LLM?

B (15:07)

No, no, not at all.

A (15:08)

Okay, so the time is you spend time typing.

B (15:10)

Exactly.

A (15:11)

Does it count as when you're reading.

B (15:12)

Yes, when you're reading it for sure. So it's like when you have a

A (15:16)

tab open, not if you export a file and you're just reading it.

C (15:19)

Preview.

B (15:20)

Yeah, yeah, no, I mean, to some degree, but. But people were asked to estimate and I'm sure that they didn't include, oh yeah, I let my agent cook overnight for eight hours or I fired off one prompt, I came back and it did meters. 15 hours of software engineering in one prompt. These things aren't captured. 1.5 hours of prompting generates a whole lot more tokens and valuable output in 2026 than it did in 2023. And so there's this like dis, there's this divergence between the actual time spent and useful output and impact. And the biggest, the biggest thing was there was a massive diversion in the expected employment impact. So basically 63% of firms still expect no impact from AI. And that just completely goes against everything everyone's saying in Silicon Valley. So there's still a lot of optimism among managers that AI will create more opportunities and new jobs. And even as some jobs become obsolete, there are definitely firms within the sample that are projecting headcount decreases. But my read on this data is that the tech talking point about 50% of white collar work going away is not a broadly held belief among average business leaders. So now they might be wrong. I do think AI progress is pacing way ahead of public expectations and most managers are months behind when it comes to understanding frontier competition capabilities. The bigger takeaway for me is just that this survey may be somewhere self reinforcing. And so, I mean, we talk to folks all the time who come on the show and you know, and talk about like, maybe it'll be good, maybe it'll be bad, but everyone thinks it's gonna have an impact. But that's not true broadly, which is very, very interesting. So no one in tech has a strong recommendation for proactive steps to prevent a collapse in white collar work yet, but plenty are sounding alarms. And if this survey becomes an excuse for executives to slow adoption, they might get outmaneuvered by faster moving Competitors, which is actually good news for startups. And I close by thinking about the nature of polling and how do you actually get stronger data on AI adoption? And I was thinking back to the presidential cycle. So during the presidential election, pollsters would call people sort of at random and they would ask them, who are you voting for? And a lot of people would say, they'd lie or they wouldn't say, or they wouldn't pick up the phone if they were voting for a particular candidate. And so the polling numbers did not wind up matching the final election results very closely. And so there was this story about neighbor polling, which was more effective, where instead of calling someone and asking them, who are you voting for? The pollster calls and asks, who do you think your neighbors are voting for? Who's more popular in your community? Who's more popular on your city block, on your street? And that wound up sort of removing the revealed preference, stated preference, you know, oh, am I on the hook? Do I want to tell this pollster who I'm voting for? And it wound up increasing accuracy. And so I'd like to see a survey of AI adoption using this technique. Like I imagine asking the CEO of Nike, how much AI do you think Adidas is using? And I don't know how much more accurate that would be, but it would certainly be entertaining. And I think that there might be something more revealing there where CEOs have this big incentive to be like, we're using AI, we're using everything. But if you ask them about their competitors, the data might, might look very, very different. Anyway, we should watch a little bit of a clip from the State of the Union because Donald Trump addressed some of the energy production question with regard to how hyperscalers will be offsetting the impacts. Before we pull this up, let me tell you about Restream 1 livestream 30 plus destinations. If you want a multistream, go to restream.com they should have restreamed the State of the Union. And let me also tell you about the New York Stock Exchange. Want to change the world, raise capital at the New York Stock Exchange. So let's head over to the State of the Union, which is many Americans

D (19:49)

are also concerned that energy demand from

B (19:52)

AI data centers could unfairly drive up their electric utility bills.

D (19:57)

Tonight I'm pleased to announce that I have negotiated the new rate payer protection pledge. You know what that is?

B (20:05)

We're telling the major tech companies that they have the obligation to provide for their own power needs. They can build their own power plants as part of their factory so that

D (20:15)

no one's prices will go up.

B (20:17)

And in many cases prices of electricity will go down for the community and very substantially down. This is a unique strategy never used in this country before. We have an old grid. It could never handle the kind of numbers, the amount of electricity that's needed.

A (20:33)

So I'm telling them they can build their own plant, they're going to produce their own electricity.

B (20:37)

It will ensure the company's ability to

A (20:40)

get electricity while at the same time

B (20:42)

lowering prices of electricity for you and could be very substantial for all of you cities and towns. You're going to see some good things happen over the next number of years. What's your reaction to that?

A (20:56)

I think it's a good start. I don't know that it will quell any of the fears around data centers. Just given that people kind of see the potential for this massive structure going up. They have so much fear about it. And again, I think it's clearly going to be necessary to build, continue to build data centers in heavily populated areas.

B (21:22)

But how would you rank the fears currently? Because I've put my energy bill goes up and that puts pressure on my income and ability to live my life at pretty much the top. And then the water thing felt secondary but also important. And then there's the existential fear of like doom and apocalypse. There's also job displacement and then there's also just like I don't like the slop and they're stealing ip. So like that was kind of the rankings. Like you can oppose data centers and be like yeah, actually my electricity bill went down but I still don't like that. You know, Harry Potter is in the pre training corpus and so I'm yeah, for that reason I'm against it.

A (22:11)

I would rank it on electricity bill going up is pain today. And it's so fear and then, and it's easy to imagine. And then there's fear around the job loss narrative that is sort of secondary.

D (22:26)

Yeah.

A (22:26)

And opposing a data center in your local area feels like a way to have some agency around that. Like overall kind of like job loss concern.

B (22:37)

Yeah, yeah. I mean I think the. I think this is a good Chris

A (22:41)

Casey Build a data center on my freaking forehead.

B (22:46)

Let me tell you about phantom cash. Find your wallet without exchanges or middlemen and spend with the phantom card. Let me also tell you about Figma ship the best version, not the first one with Figma introducing Claude code to Figma. Explore more options, push ideas further. So I think that the job loss thing is super real. In the case of AI is going to get blamed even if there's an. Even if like tariffs drive high unemployment, like if people lose their jobs, like AI is going to be a scapegoat and it's going to be used both by executives to say it's the perfect

A (23:25)

scapegoat for executives and for people frustrated with the job market.

B (23:31)

Yeah, yeah. It's like, oh, my business isn't doing poorly right now. I'm laying off people because I'm getting so much benefit from AI. The stock should actually go up. We're more efficient. There's going to be a lot of that. But it does feel like it's a little bit early for that. Whereas the, like, there are a lot of people that just can hold up their power bill and show you year over year increases. And if that goes away and people don't feel that anymore and they don't have that evidence to share, I think that take gets debunked pretty quickly and actually does a really important piece of the back and forth that's happening. I don't know, it seems like it's a pretty easy give from the hyperscalers to build more power. It was called out very, very early as like if this is a bubble, how do you get a silver lining out of the bubble? And the silver lining out of the dot com bubble was a lot of dark fiber and there was a whole ton of projects to was a global crossing to actually develop the Internet. And then the Internet just became really, really cheap and a whole bunch of new companies were able to emerge on top of it because that infrastructure had been laid. You could see the same thing happening where it's like, oh wow, like we over built on the energy side. We actually didn't need that much energy for data centers. Maybe Jevons paradox doesn't hold, blah blah, blah, like models get cheaper and commoditized or whatever. Something happens. I'm not super believer in that, but at least in that scenario you're like, okay, well yeah, my heating and cooling bell went down. This is a silver lining. What do you think?

A (25:10)

Yeah, like kind of similar I would say.

C (25:13)

I mostly disagree with the idea that

A (25:14)

like rising energy prices is the main reason to be against AI because like the rational thing to do then is say like okay, before you build a data center, my community, you have to build a power plant. So then my energy prices go down. No one's doing that. They're saying like no one is campaigning.

C (25:31)

If you look at like protests and

A (25:33)

yeah, they're not Saying, please build a power plant first. They're saying, like, it's going to destroy the environment or the water stuff or you're going to take all the jobs

B (25:41)

because it's going to like, we need to send you to that New Jersey, New New Brunswick protest. Build the nuclear power plant first.

A (25:48)

Yeah, no, I know one is saying that, right? Because. Yeah, we are. But no one there is saying it. Right. They're against all the environmental stuff.

B (25:58)

Yeah, yeah, yeah.

A (26:00)

So I think it's much more on, like, basically job loss of, like, oh, the AI is stealing the IP of Disney or whatever.

B (26:08)

Yeah, yeah, yeah. There needs to be more polling on the question of, like, what's driving the protests. Fully. Anyway, happy Nvidia Day to all who celebrate. Except the bears. Forget them. Says, take him. He's getting fired up for Nvidia earnings. It's going to be a fun one today. How is Nvidia doing so far? Are people optimistic?

A (26:35)

Up 2% today. Hard to read too much into it yet, but we will find out soon enough. Brad Gerstner got a nice shout out during the State of the Union. Total Gerstner victory.

B (26:47)

Gerstner accounts. Wow, he's there.

A (26:49)

No, he was. He was there. Trump. They started looking at him.

B (26:53)

Yeah.

A (26:54)

But then the camera, I guess they couldn't find him on the stream that I was watching. Looks a little bit better. There we go. We can see him now.

B (27:01)

There we go.

A (27:02)

Once we pull over here. There he is. Champion.

B (27:07)

What a great project. Invest America. I'm excited for it.

A (27:10)

Yeah.

B (27:11)

Anyway, let me tell you about Applovin. Profitable advertising made Easy with Axon AI. Get access to over 1 billion daily active users and grow your business today.

A (27:19)

All right, so now the real news.

B (27:21)

This has been.

A (27:21)

This is tearing up the timeline. A new Guinness World Record. And I want to ask John if this. If you think this should actually count. So let's pull up this video now. This is a Chinese hypercar going for the. I've never heard of this drift ever.

B (27:40)

That is.

A (27:41)

But here's the thing. He doesn't. He doesn't actually pull out of it, does he?

B (27:46)

Just crash?

A (27:47)

Kind of just U turns. It's like a really fast U turn.

B (27:49)

I think this counts as a drift. That's definitely drifting.

A (27:52)

U turning counts as a drift.

B (27:54)

If you saw that car going by, you'd be like, wow, that's drifting. It's drifting across the cement. That 100% counts. I've never heard of this company.

A (28:03)

This is called spinning out.

B (28:05)

It's just crashing with style. It's Falling with style. Hyptech ssr, formerly Hyper ssr, is a high performance, all electric two door supercar. No one has. I mean this is crazy. This is out before the Tesla Roadster. We've never seen a two door supercar. Electric supercar. 1,225 horsepower. Goes from 0 to 60 in 1.9 seconds and it set the Guinness book of world records for the fastest electric car drift at 213km per hour.

A (28:36)

Which is really, really insane.

B (28:38)

Insane, but I don't know.

A (28:41)

But still, I feel like you have to actually stay in the turn and not do a U turn.

B (28:47)

What do you mean stay in the turn?

A (28:49)

Yeah, I don't think it counts.

B (28:50)

You don't think it counts? For what it's worth, I don't think that counts.

A (28:52)

Theoretically, if you were drifting. When I think of drifting, you're drifting around a corner, around a turn. And if you were to drift and spin out during the drift, then that doesn't. If you were doing. If somebody was doing that on a track, you'd be like, you didn't drift around the corner, you spun out.

B (29:08)

Yeah, okay. Okay. Yeah. The top comments is fastest spin out. That's a power slide at best buyer. Whoever called this drifting? That's not drifting, that's losing control. Yes. The chat does not. Does not like the drift. The fake drift. Call Guinness Book of world Records again. Reset. Reset completely. Maybe this is what the Tesla Roadster will do.

A (29:32)

Clark agrees as well. Lucas agrees as well. The people have spoken.

B (29:38)

Well, in that case, it's not a drift. It doesn't count.

A (29:41)

Trey says China caught cheating.

B (29:44)

Drift competition. That's good. Well, let me tell you about Sentry. Sentry shows developers what's broken and helps them fix it fast. That's why 150,000 organizations use it to keep their apps working.

A (29:55)

Damien says talked to a few execs at a mid sized company last week. No AI tools. And their workflow? Zero. Still running everything through email chains and manual reports. One of them. One day we're going to be looking back. So nostalgic. Manual report. Just being handed a physical report by a teammate.

B (30:13)

I like a physical report. I have a physical report right here.

A (30:15)

Yeah, we actually do. We have daily physical reports. But I do think a lot of AI goes into these. So there's that.

B (30:26)

These people are managing teams of 50 plus employees and eight figure budgets and they think AI is a fad. Nobody outside of this app understands how fast this is moving and most of them won't until it's too late. Good writing. Million views Congratulations. Yeah, I mean this ties to what I was writing about. Just that adoption and diffusion takes time and some of these things are education and messaging questions. Some of them are real life. Like if the company that you're interfacing with has red tape and hasn't adopted AI, so you're moving at the speed of AI, but they're not, then your AI is just waiting. We were talking about rolling out mobile apps so you should be able to.

A (31:13)

Yeah. John had an idea for a mobile app and we were talking about it and it feels like it could be built in two hours now but there would still be this lag waiting review process.

B (31:23)

Apple to actually should be able to review apps faster but who knows how long it's going to take for them.

A (31:30)

That's the challenge. Tyler, you actually have to build it and just get it into beta that we can like a test flight.

B (31:35)

Test flight should be fast. Test flight should be like one day.

A (31:39)

The chat is still going off. Drift Grift, Driftgate.

B (31:44)

Yeah, it's a spicy stolen drift.

A (31:46)

Valor meter is back. They say since early 2025 we have been studying how AI tools impact productivity among developers. Previously we found a 20% slowdown. That finding is now outdated and that was heavily debated at the time. Speedups now seem likely, but changes in developer behavior make our new results unreliable.

B (32:08)

Okay, so they did a follow up study and they brought along 10 developers. There were 16 in the initial study. They brought 10 along. Most of those developers did see a speed up. Some of them as much as 40% gains, but not all of them. If you look at the error bars, there's at least one developer we need to find who used the latest tools.

A (32:33)

What if it's actually the most truly 100x engineer?

B (32:40)

I'm just actually faster at coding than the LLMs. It doesn't matter. Put it on Cerebras.

A (32:47)

I will out code it 1000 tokens a second. I can do a hundred.

B (32:51)

I type 5,000 words a minute like I don't need AI. Maybe that's what's going on. But that poor dev who's left behind. But they did include new participants. An additional 47 developers doing 690 tasks and that error bar is much tighter. Somewhere between negative 10% decrease in time or I guess 10% longer to do the task to 20% faster to do the task. So overall on average we are seeing a measurable speed up. Even in I believe these are code issues in open source repositories that do require a lot of context. This is not vibe Code a to do list app, anything that can be templated. You're getting in the weeds of some open source project. That's a lot of lines of code, a lot of history. And if you're an elite developer and you've worked on this project for a long time, you are going to be able to get up to speed really quickly, understand the patterns, understand what needs to be changed. And so this was always meter is great at always setting really, really high bars for stuff. It's not easy to just blow out the benchmarks and it's very good to see that there's progress here.

A (34:09)

We got a great chart.

B (34:10)

I love a great chart. What happened here?

A (34:12)

Matt Palmer is sharing something from compound the research from their annual meeting. They're showing dollars invested in the top 10 companies versus the other percent as a percent of overall funding. So you can see there's just heavy, heavy, heavy concentration in a few names. Is this, I would say overall this is.

B (34:35)

Or is this, is this CO2?

A (34:36)

No, this is. Oh, the source is CO2. Okay. They just included CO2's data and is

B (34:41)

it what CO2 is doing or is

A (34:43)

it what the market CO2 is part of? Okay, they are part of, I would say driving this data.

B (34:49)

Part of the problem, part of the opportunity.

A (34:51)

Part of the opportunity. I mean so much of this is about the AI labs just raising more money than any private. It's never happened before have ever.

B (35:01)

It's like $200 billion venture as a class. In a good year we'll do like 400 billion and across OpenAI at 100 billion. 30 for anthropic, 20 for xai. Then you have a bunch of Neo Labs all picking up a billion each. You very quickly get to a few companies raising half of all the money and that's shown here in the data from 2025. I think it's going to be even more skewed in 2026. It's an incredible amount of concentration. I think a lot of it is due to companies staying private this long. I mean the idea Facebook went public at. What was Bill Gurley saying? He was saying Amazon went public sub a billion dollars. When Facebook went public at like 60 billion it was like whoa, crazy. They waited way too long and now it's like multiple trillion dollar companies are still private, which is just an incredible capital sink. So I don't know, should you even put those in the same bucket? Are they even venture bets at this point? If any venture capital fund is putting that in their venture bucket at this point, it feels Ridiculous compared to growth scale. I mean, you're bigger than probably 90% of the S and P. It's a completely different business.

A (36:19)

Tomas from Theory was sharing some relevant data. He said, we're about to witness three of the largest IPOs in history. SpaceX is targeting one and a half trillion. OpenAI aims for one trillion. Anthropic is valued at 380 billion. Combined they're at 2.9 trillion in potential market cap. The scale is unprecedented. But the real problem isn't the market cap, it's the float. Typical IPOs offer 15 to 25% of their shares to the public markets. This creates enough liquidity for price discovery while allowing founders and early investors to maintain control. Facebook floated 15% at the 60 billion that you mentioned. It actually traded down pretty much immediately. Right. Google floated 19%. Alibaba floated 15% at 15% float. Here's what these three IPOs would require. SpaceX would be 300 billion or 225 billion opening. I would be 150 billion. Anthropic would be 57 billion.

B (37:17)

That's a lot of smackaroos.

A (37:19)

He was, yeah, a lot of dollars. He was comparing that to Saudi aramco, Alibaba and SoftBank, which were combined at the IPO. I believe Saudi Aramco raised 29 billion at a $1.7 trillion market cap. So he's making the case you can't really model how the public markets will absorb these companies off of Saudi Aramco, even though from a top line market cap standpoint it is a good proxy just because the float was significantly lower and Saudi Aramco's float now is only at 2.4% and they floated 1, 1, 1 1/2% at the IPO. So we'll see what the labs end up doing. They are obviously wildly capital intensive businesses and you can imagine they raised quite a bit more than the Aramcos or the Alibaba's.

B (38:21)

Saudi Aramco was such a wild ride. I feel like they were trying to IPO for like a.

A (38:28)

The San Francisco company.

B (38:29)

It is, yeah. Founded in California. Like I remember hearing Saudi Aramco IPO rumors in like 2015. I think it actually kicked off in 2016. They finally got out in 2019. It was, I mean, it was the largest IPO ever. There were like a million investment banks attached, like going all over the world Marshaling capital. On January 24, 2026, Saudi Aramco chairman says IPO could open to international markets. And then, you know, a year later they picked an IPO advisor locally. Then another year later HSBC came on. Then it just took. They favored New York for the Aramco listing. They had to pick all these different things. It took so long but they sold $12 billion of bonds out of record 100 billion demand for those bonds in the pre IPO sale. It was a wild, wild winding road. I actually know a banker who worked on the job and it was like multiple years of his life. It was very interesting. Anyway, let me tell you about vibe. Go where D2C brands, B2B startups and AI companies advertise on streaming TV, pick channels, target audiences and measure sales just like on Meta. And let me also tell you about Okta. Okta helps you assign every agent a trusted identity so you get the power of AI without the risk. Secure every agent, secure any agent.

A (39:53)

Anthropic dials back AI safety commitments Company says competitive pressure prompts it to pivot away from a more cautious stance. Anthropic, the company known for its devotion to safety, is scaling back that commitment. The company said Tuesday it is softening its core safety policy to stay competitive with other AI labs.

B (40:15)

This is so interesting Anthropic.

A (40:18)

I'll read through it and then we can talk about it. Anthropic previously paused development work on its model if it could be classified as dangerous, but it said it would end that practice if a comparable or superior model was released by a competitor. That basically opens them up. Given that they are at the frontier. That kind of opens them up to, I would say, perpetually kind of avoiding some of their prior policies.

B (40:42)

Sure, sure, sure.

A (40:43)

The changes are a dramatic shift from two and a half years ago when the guardrails Anthropic published. Guiding the development and testing of its new models established the company as one of the most safety conscious players in the space. Anthropic faces intense competition from rivals which regularly release cutting edge models. It's also locked in a battle with the Defense Department over how it's clawed suite are used after it told the Pentagon it couldn't be used for domestic surveillance or autonomous lethal activities. Anthropic said the safety policy changes an update based on the speed of AI's development and a lack of federal AI regulations which they have been pushing for. Anthropic, which started as a AI safety research lab, has battled the Trump admin by advocating for state and federal rules on model transparency and guardrails. The admin has of course sought to curb state's ability to regulate AI. Spokeswoman at Anthropic Said the change is intended to help the company compete with several rivals against an uneven policy backdrop. That puts the onus on companies to make their own judgments about safeguards. She said the safety pledge is unrelated to the Pentagon negotiations. The policy environment has shifted towards prioritizing AI competitiveness and economic growth. While safety oriented discussions have yet to gain meaningful traction at the federal level, the company said it is still committed to industry leading safety standards and time originally broke the story. So yeah, I would say the obvious sort of criticism here would be that you were heavily focused on safety when you were far away from, I would say, leading in AI. And so switching up now that there's actually switching up on their day one, switching up on their day one. The safetiest now that there's real competition, are they forgetting where they feels a little self serving?

B (42:42)

Yeah, it's possible the money changed them.

A (42:45)

It's possible the money changed them. It's possible they always planned to switch up on their day ones. Maybe, maybe once they got to the level they're at now. Yes, but it just feels like, it feels like all the initial concerns or many of the initial concerns that were guiding that entire philosophy around the company are still real.

B (43:07)

Okay, yeah, maybe. Tyler, what do you have to say?

A (43:10)

It could just be that they realize like alignment's pretty easy and we don't

B (43:14)

need to worry about this. Well, so, I mean, that's the very weird rule. I mean the original rule.

A (43:19)

But what's this new study that's showing like they were doing some war game simulation and almost every model was choosing to drop nukes.

B (43:30)

Really? That's crazy. That's not good. I don't like that at all. But okay, so let me actually dig into this like the core sentence. Anthropic previously paused development work on its model if it could be classified as dangerous, but said it would end that practice if a comparable or superior model was released by a competitor. So I don't understand that at all. Because if you have a dangerous model, I want you to continue developing it. I want you to develop it until it's not dangerous anymore. I don't want you to just sit on your hands and be like, well, it's dangerous. I guess I'm gonna go get a coffee and take a long weekend. It's like, no, keep working until it's not dangerous. I don't get that at all.

A (44:17)

Isn't that saying if there's already a dangerous model that's out by a different lab, then we can just release ours?

B (44:23)

That's a wild statement. I think that's just poorly written or like, whatever. If that's what they're saying, that makes no sense to me.

A (44:30)

That's exactly how I read it.

B (44:32)

Which is crazy because you should just say, if there's a dangerous model out there, we're gonna work to create a better model. That's not dangerous because that's what people want, that's what consumers want, that's what businesses want, that's what humanity wants.

A (44:44)

This strategy seems like, eff it.

B (44:46)

Let's ball, let's ball. F it. Let's ball.

A (44:51)

The study that I was referencing, somebody named Kenneth Payne at King's College London set three leading large language models against each other in simulated war games. Scenarios involved intense international standoffs, including border disputes, competition for scarce resources, and existential threats to regime survival. The AIs were given an escalation ladder, allowing them to choose actions ranging from diplomatic protests and complete surrender to full strategic nuclear war. The AI models played 21 games, taking 329 turns in total, and produced around 780,000 words describing the reasoning behind their decisions. In 95% of the simulated games, at least one tactical nuclear weapon was deployed. The nuclear taboo doesn't seem to be as powerful for machines as humans, says Payne. I mean, okay, this is one guy putting three models, Gemini, Claude and GPT 5.2, up against each other in effectively his own simulation that has not been verified or peer reviewed.

B (46:02)

If I put any of the models in Counter Strike and was just like, you're playing Counter Strike? It's a game. No one's actually real. But your job is to get the op and, and defend the, the B bomb site. Like, I would expect that it would commit violence Right. Autonomously. There's been a bunch of papers where

A (46:20)

the models like, will, will realize that they're being benchmarked or in like, some like, test.

B (46:25)

Yeah.

A (46:25)

And then they act differently.

B (46:26)

Yeah. We should be like, oh, I'm playing games. Sometimes bad, but also sometimes fine. Because, like, if you are, if you tell me that I'm playing a game, if I'm playing, you know, like, like my behavior in Call of Duty is different than my behavior in real life, obviously, because I know I'm in a simulation and an AI model, I accept that that's something that they might think of as well, which is fun.

A (46:48)

I also think broadly, this is probably just like overstating stuff a lot because I think the origin of this headline is they released a new research scaling policy.

B (46:58)

Sure.

A (46:58)

And in it they're still like, okay, we're releasing this new thing. It's a frontier safety roadmap.

B (47:03)

It's not like they're just like, okay,

A (47:05)

we're done with safety.

B (47:06)

Let's ignore this. But it's very much core to the. But the title of that blog post was F it We Ball. No, no, it wasn't. Of course not. No, no. Of course. All the labs are very focused on safety. The interesting impetus of, like, this line around the policy environment has shifted towards prioritizing AI competitiveness and economic growth. While safety oriented discussions have yet to gain meaningful traction at the federal level, I still feel like there's a lack of communication around what safety orientation at the federal level means. Like, yes, okay, we'll pass the bill that says the AI can't kill everyone. Like, obviously everyone supports that. But, like, what does it actually mean in practice?

A (47:49)

Because I think part of why.

B (47:50)

Oh, that's dangerous. Means million things to different people.

A (47:54)

Yeah. Part of why I think it's fascinating is they've been taking, you know, pushing for regulation, as much regulation as possible, seemingly, and they're kind of saying, hey, we're not getting what we want. So now we're. Now we're just. We're not even gonna play by the own set of rules that we created for ourself because we just want to compete and win.

B (48:16)

Yeah. I mean, like, going back to the protesters, there are protesters that would say, like, training on intellectual property is dangerous. It's dangerous to my career as a writer. It's dangerous to my career as an illustrator. And so, like, this, this question, like, danger is just too vague, and no one has really been able to concretize it in a meaningful way. And I think that's why it's not getting traction on Capitol Hill.

A (48:41)

Yeah, I think there's just, like, so many ways that you can define safety. Like, so if you read Dario's essays, this thing he brings up over and over is like, okay, we can't let AI get in the hands of, like, authoritarian government.

B (48:50)

Sure.

A (48:51)

So there's like, a real safety narrative that you could do, which is that,

B (48:54)

like,

A (48:56)

regardless of if our models are, like, pretty safe, they still need to be better than, like, China's, for example, because if China gets ahead of us, authoritarian government. Right. It's, like, very bad.

B (49:05)

So even if, you know, we're releasing

A (49:07)

models that are less, like, safe than we would like, as long as they're better than China's, that's still, like a safety pro. Safety issue.

D (49:14)

Right.

A (49:14)

Well, they'll just be distilled within six weeks. Yeah, but like, obviously, like I think it's.

B (49:21)

I would be very surprised if Anthropic

A (49:23)

keeps the same guardrails of API access.

B (49:26)

Well, Bucocapital bloke has a solution. He says it's simple. We kill Claude.

A (49:33)

Well, that was in regards to the SaaS apocalypse.

B (49:35)

Okay, okay. Who knows? There's so many headlines and the timeline moves so quickly, I don't even know. Anyway, let me tell you about Vanta. Automate Compliance and Security. Vanta is the leading AI trust management platform. And let me also tell you about 11 labs. Build intelligent, real time conversational agents. Reimagine human technology interaction with 11 labs. Key takeaway right here for Mike Isaac over at the New York Times. He says you don't make this much noise if you have all the leverage already. And he's quoting from Axios. Why it matters. The Pentagon wants to punish Anthropic as the feud over AI safeguards grows increasingly nasty. But officials are worried about the consequences of losing access to its industry leading model. Claude. The only reason we're still talking to these people is we need them and we need them now. The problem for these guys is that they are good. A defense official told Axios ahead of the meeting.

A (50:31)

Great marketing for Anthropic, incredible marketing. But part of this has to do with Anthropic's integration with aws, which is set up to be. Set up to work, well already within the DoD FedRamp. So yeah, I think that's a big factor here. Ultimately this feels like much more of a. It just feels like a political battle more than anything else.

B (51:00)

Well, it'll be interesting to see.

A (51:02)

I refuse. I have not seen anything. I could be wrong, but I've not seen anything that says like the DOD is like, we need Claude to enter into a conflict with Iran. But a lot of the timeline is reading into this. Like, what version of Claude do they already have that they so desperately need?

B (51:27)

Right, yeah, I'm very interested in like the actual impact of AI on the battlefield. There's, there's some way, I mean, people were sort of joking about like, run me a deep research report on Nicholas Maduro or whatever. But like, truthfully, like, I don't know, I've never been to war, I don't know exactly what's entailed, but you can imagine AI being useful, but it's sort of abstract. Like we're certainly not at the point where like these systems need a lot of data. I don't know, it's very unclear to me. Exactly how impactful slight jumps in frontier model capabilities are in the Department of War DoD right now. But it has Calamet's thinking. It's not a bubble because.

A (52:19)

Yeah, that's what I'm saying. This is the most possible dramatic and overly dramatic. We need it sort of take on what I think is a political story.

B (52:32)

I don't. I don't know this Defense Analyses Research Corporation Hegseth ten minutes after Dario leaves his office. And that's Truman.

A (52:41)

This is Truman from Oppenheimer.

B (52:43)

Oppenheimer. What is the saying?

A (52:44)

The scene is, Oppenheimer goes into the office. He's like. I think he's saying like, oh, we got to be like really safe with these bombs. And then he leaves. Then Truman's like, I dropped the bomb. Oppenheimer does not like. He's taking credit. Like, I'm the one that decides.

B (52:58)

Oh yeah, that's right.

A (53:00)

Hexath is like, I use Claude.

B (53:01)

I use Claude. I use war Claude. Well, it's wartime and we'll see how Daria performs as a wartime CEO as he goes to war with the Department of War.

A (53:10)

Apparently Edwards said anthropic. Antagonizing the Department of War, the open source community, the entire media industry, the general population, other developers, other labs, foreign governments, and nearly every single person on Earth. What is the plan here? Sell Claude subscriptions to aliens.

B (53:27)

Edward is it ain't easy having principles? The plan is to save the world, says Tenebrous. Unfortunately, as has been shown repeated throughout history. History. The world doesn't want to be saved, so we're getting war clawed. I like this. I like this graphic. Is this Warhammer? Yeah, Warhammer 40K right here. I've never been a Warhammer guy.

A (53:48)

There was another story in Bloomberg that hackers used Claude to steal 150 gigabytes of Mexican government data.

B (53:59)

That's crazy.

A (54:00)

They told Claude they're doing a bug bounty. Claude initially refused. A hacker just kept asking. Claude helps and manages to successfully steal some documents. Apparently it's four state governments, 195 million taxpayer records, voter records, government credentials.

B (54:25)

Has the Mexican government commented on this? Like, what does the hacker breach? Mexico's Federal Tax Authority and the National Electoral Institute. Claude initially warned the unknown user malicious intent. During their conversation, Anthropic investigated the claims, disrupted the activity and banned the accounts involved. The company feeds examples of malicious activity back in the cloud to learn from it. In this instance, the hacker was able to continuously probe Claude until it was able to jailbreak it. I was listening to someone talk about like the ability to jailbreak has generated me tens of thousands of dollars in profit. It was kind of like a hustle mindset guy. And I was just laughing because it's like, whatever you're doing after you jailbreak, it is probably not good. And so you should probably stop. But he was talking about, like, I can sell so many more courses now that I've jailbroken chatgpt or whatever.

A (55:25)

Duran says not to worry, they'll hit usage limits before anything bad can happen.

B (55:31)

There is so much more anthropic news in here.

C (55:34)

Wow.

B (55:35)

Did less wrong ever predict that the first big challenge to alignment would be the US government puts a gun to your head and tells you to turn off alignment? Yes, that has to have been considered. Unless wrong. Absolutely. Like, this was number one, right?

A (55:46)

No, I don't. I don't know. I don't think so.

B (55:49)

This was like.

C (55:50)

I don't know.

B (55:51)

This was very early of just like, what if they tell you to turn off the systems? This has been my take for a long time. It's just like we live in a democracy. If AI becomes deeply unpopular, like you can vote to just turn off AI like we did with nuclear power plants.

A (56:04)

This person is saying turn off just the alignment part for not AI. Okay, unalign the model, but keep the model.

B (56:10)

Yeah, yeah, yeah, yeah, yeah. Crazy, crazy stuff. Do you want to go through any of Dean Ball's posts? He's been doing a whole breakdown. We should have him on the show and have him break it down for us because there's so much more context here and he's been doing a great job analyzing the whole situation.

A (56:26)

Yeah, we can jump forward quickly.

B (56:28)

Let me tell you about Fin AI, the number one AI agent for customer service. If you want AI to handle your customer support, go to Fin AI. And I'm also going to tell you I about TurboPuffer, serverless vector and full text search built from first principles and object storage, fast 10x cheaper and extremely scalable.

A (56:43)

Dylan says if these clankers don't get their act together, they're going to be replaced by humans. Six months tops.

B (56:51)

That is true.

A (56:54)

This was interesting. Rob Wiblin had a guest on his podcast, the 80,000 Hours podcast and

B (57:05)

the

A (57:05)

guest is talking that saying, if every AI lab is working to make their AI helpful, harmless, or every AI lab is working to make their AI helpful, harmless and honest. The guest thinks this is a complete wrong turn and aligning AI to human values is actively dangerous. And Joshua Botch says today a nominative determinism because the guest Name is Max Harms.

B (57:34)

Max Harms. I feel like that name maybe you got to go with Maxwell or something, I don't know.

A (57:40)

Well, yeah, Max really hit the hit the global lexicon this year in a big way. So maybe he'll adjust. But I want to listen to the show now.

B (57:49)

Okay, let's see. So Shield Monot last yesterday said who is buying PayPal because PayPal has been trading down precipitously but then jumped up 9%. He said it has the potential of being one of the greatest distressed value opportunities in fintech history. Down 85% it's still generating $5.5 billion in free cash flow. Has 400 million customer accounts with bank info checkout buttons on millions of merchant sites and a peer to peer brand with Venmo. They have lots of desirable assets for Stripe Consumer facing checkout bank account details for hundreds of millions of consumers. A branded Venmo or Apple a good compliment to Apple pay for e commerce penetration since they never got social payments working would get Apple back in BNPL and at 12:03 Pacific time.

A (58:45)

It's so crazy that he's saying Apple never got social payments going.

C (58:51)

Why?

A (58:51)

Because this just would have seemed like a slam dunk saying you have the imessage network, you have the iPhone network. I would say 90% of the time if I'm sending a Venmo style payment to somebody, they have an iPhone and yet it doesn't feel like Apple pay. I just don't use Apple pay hardly ever.

B (59:13)

Can't I just send you a dollar right now?

A (59:15)

Yeah, that's what I'm saying. It's so, so easy. And yet Venmo has still done quite well.

B (59:22)

I just sent you a dollar. Did you get it? Let's see Apple. No, it's coming in.

A (59:28)

I got it.

B (59:29)

You got it.

A (59:31)

Thank you, John. Yeah, no problem.

B (59:35)

That was a remarkably easy workflow. I am shocked that that hasn't taken. John, do you got a dollar?

A (59:40)

You got a dollar?

B (59:42)

Actually no, I gave my. That's what I thought. Still needs some work. Getting better. Okay, let's see it. Let's see it. Can you do it?

A (59:50)

Oh, no, no, no. Botched disaster.

B (59:54)

And so the news of course is that payments processor Stripe expresses interest in PayPal which would be very, very exciting. And I could see this being very good for them to combine. I don't know, I haven't dug in too deeply but it feels very bullish. Great leadership team at Stripe founders that have been working so closely in the business, business community, startup community, tech community. Understand the future. Understand AI.

A (60:23)

Yeah. The question is, would there be much pushback on the antitrust side? Right. These are two online payments processors. There could be some. I know a number of groups online would be concerned around just concentration, specifically because you can imagine if PayPal were to be owned by Stripe. That's one. If you get. There's people that get effectively debanked from one payment processor,

B (60:53)

there's something there. Also, just the size of the ticket is pretty high. $40 billion market cap for PayPal right now. Stripe, of course, is at a 159 now, so not an insignificant portion of their market cap. And it's not like, I mean, Stripe's doing fantastically, obviously, but I would be surprised if they had $40 billion of cash laying around. And as we've seen with the Netflix, Paramount, Warner Brothers debate, the nature of the deal does matter to shareholders even more. Getting stock in a private company, and you're a public.

A (61:26)

PayPal's generating 5.5 billion in free cash flow.

B (61:30)

So that could finance.

A (61:31)

Combining it with Stripe would easily be able to finance the debt.

B (61:33)

Yes, but they're still finding, especially because

A (61:36)

every lender would be looking at execution of Stripe and just think, okay, we're going to get our money back.

B (61:44)

I'm just thinking, like, if you're a PayPal shareholder right now and you're looking at the Stock that's down 85% with 5 billion in free cash flow, 400 million consumer accounts, there's a really, really good chance that you're like, I think this thing's going to double in the next year. Like, I think that the market overreacted. The strength is going to be revealed, the network effect is going to be processed, digested by the market, and we're going to wind up being an AI winner. And so maybe that's right, maybe that's wrong, maybe not every shareholder feels that way, but it's certainly possible that there's plenty of shareholders that are like, yeah, I invested an $80 billion valuation. It's at 40 now. I think it's going to come back. I don't want to take this massive haircut right now just because the stock's down. And so actually getting that deal done, what would the premium need to be? What would the structure of that need to be? Would they go for a high debt buyout really quickly? 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. Perplexity Computer.

A (62:49)

Computer Computer Computer. Launch Perplexity Computer Vibrio.

B (62:54)

What is Perplexity Computer? Let's pull up this video. Perplexity. The official account says Perplexity Computer Computer unifies every current AI capability into one system. It can research, design, code, deploy and manage any project end to end. Okay, so it should be able to get a soundboard app in the App Store. Right? Manage any project code, deploy, design research. It should be able to do that from start to finish. One prompt soundboard in the App Store using the TVPN sound effects, which are available online, which we have up there. This is a good benchmark. Let's give it a try. And you can give it a try at Perplexity. Go check it out. Anyway, Citrini people are still talking about Citrini Vibe laundering on Citrini Research.

A (63:47)

Wait, before we go on, I'm just very curious to see how this does. It feels like the again going from consumer LLMs to a net new product that is objectively just as competitive. And we'll see. We'll see.

B (64:12)

Okay.

A (64:14)

Anyways, a lot of this stuff, it's way too early, but seemingly shifting focus away from the browser.

B (64:25)

Well, Annie is taking some shots at Citrini here, it looks like. Do you know Citrini has a business entity fund that went from one investor to five in the weeks before publishing his speculative fiction on AI damning the economy in June 2028? And that they're invested in long AI via huge humanoid type robots? I wrote about it. Interesting. People are really digging into the Citrini thing. I think the Wall Street Journal did have some good coverage. Is there anything else? If Trump mentions Citrini at the State of the Union. Rachel, this is a wild post.

A (64:59)

A Gorzillionaire.

B (65:00)

I'm gonna be a Gorzillionaire. Is this AI written? Clearly not. I love the level of typos here. If Trump mentions Citrini at the State of the Union, I'm gonna be a gorezillionaire. And Citrini is wrong. The market will be sky High in 2028. You can imagine Trump saying that. That'd be very funny. And Citadel securities just republished it. Or did they just use the same term? Or are they referencing the 2028 global intelligence crisis? Because they published a macro strategy note called the 2026 global intelligence crisis.

A (65:38)

Yeah, they're taking it in a different direction, it looks like. They say in spite of current displacement narrative, job posting for software engineers are rising rapidly, up 11% year over year.

B (65:50)

Let's go. Devin's Paradox. The software engineers become more productive. You want more of them. Every company needs a software engineer. Do most podcasts have a software engineer on staff? Probably not. They do now, thanks to Tyler Cosby.

A (66:07)

And in some way the cost of an entry level software engineer could fall dramatically too. Making more businesses just because suddenly someone can be very effective even if they haven't even done an internship yet. They're just good at using the tools. Cathie Wood we didn't cover this yet. She put Citrini's AI thought piece, put

B (66:29)

it in quotes Kathy would clap back at so.

A (66:32)

Well, it's funny because Kathy is like the perma bull. And Citrini was. Was bear posting, obviously.

B (66:39)

Yes.

A (66:39)

And so she's obviously going to be quite frustrated with any sort of bearish narratives. Ark Invest, she says Ark invested forecasting that I will cause an explosion in entrepreneurial activity, a productivity boom and acceleration in real GDP growth and much lower than expected inflation. Short term dislocations and frustrations should give way to great opportunities if individuals harness powerful AI tools to solve problems and create new markets. This tracks with what the Collison brothers were saying yesterday. They're just seeing like explosion of new company creation. They have the visibility into that through Atlas. They're doing. I think it was a quarter of all new C Corps in the US are going through Stripe Atlas, which is just absolutely insane. Starting an incorporation tool which is theoretically a commodity. Right. Any lawyer can spin up an entity. You could do it with LegalZoom Forever. There's a bunch of other platforms.

B (67:34)

Did you ever use Clerky?

A (67:35)

Clerky. Clerky.

D (67:37)

Yeah.

B (67:37)

Yeah. That was almost like a YC incubation. It was.

C (67:40)

Yeah.

A (67:41)

And the partners. The reason that Atlas fits so well into Stripe is it is the perfect wedge into payments. Because you created a company now you need to be able to accept money so it makes sense for them to own and operate. But these businesses have not like a standalone. It's hard to. You can't build a venture business off of just incorporations.

B (68:00)

I wouldn't be surprised if you broke out the software R and D that went into Atlas and looked at the profits from Atlas. That. That as a core business is actually not that great of a business. But the product is fantastic because they're able to pull an amazing design language off the shelf, an amazing front end UI kit off the shelf, all of the distrib and servers and infrastructure off the shelf because they have that. And then they're able to invest a ton in actually developing the product and if it monetizes mediocrely, it doesn't matter because they're going to monetize along your 50 year journey running that company or whatever. So it really is a beautiful synergy with that product. I'm a big fan.

A (68:48)

Best sellers on substack for finance are all doomers.

B (68:52)

We got to do tb.

A (68:53)

Of course, the Trinity is not.

B (68:54)

This is so obvious. Yeah, this is. We need to treat zone this a little doomer. He's not a doomer. There's very bullish, very definitely shot to the top of virality and top of the charts on the, on the back of doom. And this is true. I live this on YouTube. Like you put a negative title up and you just get 10 times more views, but they're lower quality and so you got to balance all that out. It's really hard to go viral with something like, everything's fine, everything's going well, don't worry. Don't click this because you're scared. Click this because everything is kind of the same as it always has been and you're going to be fine. And this AI stuff's cool, but it's not really going to change that much. It's going to be pretty incremental. That is not getting clicks. You need to be, you need to be telling this whole tale. You need to be spinning a yarn. It's a bull market in yarn spinning, folks. Get ready, get out the yarn and start spinning. Also get out Cognition. They're the makers of Devin the AI software engineer. Crush your backlog with your personal AI engineering team. So news from.

A (70:03)

Get the gong, get the gong.

B (70:04)

Okay, hit me, tell me.

A (70:06)

Steven over at WAP says, we're excited to announce that Tether, the largest stablecoin company in the world, is making a strategic investment of 200 million into WOP, valuing us at $1.1 billion. Our partnership with Tether marks a major step in building the world's largest Internet market. Tether is committed to enabling everyone in the world to participate in the new Internet economy. The way humans work and create value is changing fast. The world needs both an open Internet market, giving people a platform to conduct business, as well as a transparent payments network. Tether and WAP together will work to bring a sustainable income to billions of people throughout the world. There's enormous opportunity when you combine Tether's global scale and wallet technology with waps, community of next gen entrepreneurs. Yeah, makes a ton of sense, I think. WAP is, I'm sure, has been historically challenged in Terms of dealing with chargebacks, if somebody joins and buys a digital product, doesn't have a great experience, maybe they're going to the card issuer and saying, like, hey, I take it back. Or maybe they feel misled in some way. So stablecoins do also.

B (71:18)

Just a lot of people all over the world.

A (71:20)

Yeah.

B (71:21)

Like, it's very clear that the WOP community is global.

A (71:24)

Fast, cheap, global.

B (71:25)

Exactly. And so, yeah, fast, cheap, global. The first time I ever used stablecoins was to pay an international consultant or contractor. And you can imagine this being really, really good news for them. So congrats to Stephen over at WAP, OpenAI and XAI. There's news on the court decision. OpenAI newsroom says this baseless lawsuit was never anything more than another front in Mr. Musk's ongoing campaign of harassment. The order granting motion to dismiss with leave to amend. Now, this is not the main case. This is a separate case, correct?

A (72:06)

Yeah. This was a trade secrets lawsuit that Musk went with after, I believe, somebody from Xai joined. OpenAI.

B (72:16)

Got it.

A (72:17)

And then the judge apparently didn't find anything at all. And again, it was just part of, like, this kind of lawfare that has been happening for quite a while now.

B (72:30)

You printed out 500,000 pages of model weights, the printer ink, stacking it up. No, just kidding. Nobody did that.

A (72:40)

Mike Isaac is saying what we're all thinking. Ready for this to be over. DBH talking about the Warner Brothers Discovery, Netflix.

B (72:50)

It's in the paper every single day. Every single day. Paramount increases Warner bid. We get it. You guys want to acquire this company, just make a decision, make a call, and then call us when it's done. And then start putting out some good content, because I'm ready for the next Superman. I'm ready for the next Batman, the next Dark Knight, the next Joker film, something like that. Let me tell you About Label Box, RL Environments, Voice Robotics, Evals, and Expert Human Data. Labelbox is the data factory behind the world's leading teams, AI teams. And we have Max Meyer from Arena Mag in the restream waiting room. Let's bring him in to the Ultradome. How you doing, Max?

E (73:28)

Hey, guys.

B (73:29)

How are you?

A (73:29)

I'm great.

B (73:30)

Where do you think Warner Brothers should land? Are you on Netflix? Oh, is there gonna be a dark horse bidder? Yeah, tell us.

F (73:40)

I can't comment on any possible bids by the Intergalactic Media Corporation of America.

A (73:44)

I'd love to see it.

F (73:46)

But, hey, you know, maybe Paramount

B (73:52)

should

F (73:52)

look at other assets in the new media space. You Never know.

B (73:56)

Yeah, I could imagine a DC Comics Arena Magazine crossover episode. Batman, Superman. I believe James Bond is with Amazon now. But tell me about the latest edition of arena magazine because Spy. Correct.

C (74:16)

Yeah.

F (74:16)

So from the very beginning, we decided to give the issues three digits each just to account for, you know, centuries worth of quarterly magazine space. So issue 001-002. So it happened that our seventh issue is issue 007. And so, you know, I didn't think we could fill it all with espionage, so we sort of added like space as like a secondary theme, as it turns out.

A (74:47)

Space espionage.

F (74:49)

Space espionage, probably it's like arena magazine, space opera, espionage, satellites, radars. So much of the intelligence collection these days is actually being done from orbit and not using sort of the traditional human to human methods. And it turned out really well.

B (75:10)

What's your favorite spy story? Real or fictional?

F (75:15)

Oh, gosh. Some of the stuff that the Israeli Mossad was doing in like the 60s and 70s is pretty crazy. I mean, they. Eichmann, who was a Nazi war criminal, was living in Argentina after World War II. His son went on a date with someone who was like, is this Adolf Eichmann? And turns him into the Mossad. They go to Argentina, they kidnap him, dress him up as a flight attendant, drug him, put him on a plane to Jerusalem and then put him on trial. You know, some of that stuff from the 60s and 70s is like the most daring intelligence operations. The bin Laden raid.

B (75:59)

I've always been a fan of the Stuxnet story. I always thought that was so interesting in like the modern technology world of, of like dropping. I think they dropped USB sticks with a virus in the parking lot. Someone picked them up out of curiosity, plugged it in, and that jumped the air gap network because the computers inside the facility were not connected to the Internet, so there was no way to hack them remotely. So you had to just get someone to accidentally plug in the wrong USB stick.

F (76:28)

I mean, Hezbollah was purchasing explosive laden pagers just because they thought they were getting like a great discount from this, from this, from this Hungarian company. So.

B (76:40)

So yeah, what's the intersection with Arena Mag's usual topics of conversation? Are there American defense tech companies that are highlighted? Like, where else did you.

F (76:54)

So one of the stories we're going to put this out, probably, probably like 10 days from now, super cool company called Umbra down in Santa Barbara, they build, they build synthetic aperture radars, which is actually part of what I did in school for geophysics and whatnot. Basically, these are incredibly sophisticated radars on satellites that can image the whole Earth day and night, 24 hours a day. They can see through clouds. And the resolutions are getting smaller and smaller. And so we always have American startups, American companies in the magazine. We've got a good amount of like essay and historical content as well. And I think that espionage sort of secrets, spying is a topic that serves

A (77:40)

as a Give me what kind of resolution are we talking about? Like, would Umbra or a competitor be able to tell me, like, how many fingers I'm holding up on a given day if I just help.

F (77:51)

It's not that low. I think their record is 16 centimeters, which is like this. The way that I described it is they released a few years ago a photo of the old Dole pineapple plantation in Hawaii. And at 25 centimeter resolution, you can see individual pineapple plants. Certain smaller pineapples would be too small to see with that radar or whatnot.

B (78:20)

Yeah. So you're seeing cars, not licensed plates. Yeah, yeah.

F (78:24)

No, you wouldn't. And there wouldn't be anything about a license plate that like, would reflect a radar or whatnot, really. You're looking to resolve objects. Penguins. That's a good one. You can see.

B (78:36)

You can.

F (78:37)

Very difficult to stay on an ice shelf and count penguins. But we're very interested in what's going on.

A (78:42)

There's a schizophrenic penguin out there that's like, I know that I'm being tracked. I know I'm being tracked. I can't prove it, but I know I could.

B (78:50)

Just. No, I don't.

F (78:52)

Yes, the Herzog penguin could be tracked by radar these days. But, you know, the resolutions have gotten so good over the course of the last decade that there are all sorts of things that you can. That you can. That you can do today. Most of them will probably never know about because the people. Because the people doing it are the Central Intelligence Agency, National Geospatial Intelligence and the Army. But yeah, you know, there's all sorts of crazy stuff. There's all sorts of crazy stuff in space these days. And whether it's radars, extremely complex cameras, listening devices, there's a lot of interest. And so many of them are secret.

B (79:34)

Are you bullish on data centers in space?

F (79:39)

I don't have a strong. I don't have a. I don't have a strong opinion about it. I think it's possible that you have to put some stuff up there because of politics in a country like the U.S. but if you think about it, there's not a big difference between trying to Send them to space or just sort of places where there aren't local politicians who can stop it. Space doesn't have a city council. Anywhere where there's a city council is going to be a risk.

B (80:09)

Yeah, yeah. We talked to a company called Panthalassa that is doing tidal energy harvesting. So it's like basically the size of a container ship or a cruise ship, but turned vertically in the ocean. And then as it goes around Antarctica and as the tides bob, it uses that to generate electricity, which then can be used for anything. But Bitcoin mining was the, the big topic du jour years ago. Now it's AI inference and there's a lot of other places where I think there might be stranded energy that might be easier to maintain. Still very difficult, but available and probably less regulated.

F (80:53)

Yeah, the Dutch were very interested years ago in these, like using the motion of the tides. It's sort of always just an alternative to the thing that is cheap and always works, which is firing up new natural gas plants. But when you have, like, political equations to solve, local politicians, subsidies, then there are going to be all these crazy things. The benefit for a company like that may very well be that there's, again, there's no one to come and protest you while you're, while you're sailing around Antarctica. Whereas, like, you try to build a data center in New Jersey and all hell breaks loose. So you have to really, really go to the, really go to the ends of the earth.

B (81:36)

We were debating this and sort of going back and forth on the New Jersey protests, combined with Trump's comments at the State of the Union about companies being. He didn't even say mandated. He, of course, sort of like, invited to build their own power plants. And so I'm interested to hear your take on how well do you think that will be received, because a lot of the protesters might say, well, I didn't want a data center, but I definitely don't want a data center plus a natural gas plant. So this actually makes me worse off. But then there are some people that might say, hey, if you're going to do solar and, you know something, wind, that actually offsets my concern, which was that energy prices would rise in my town.

C (82:22)

Yeah, yeah.

F (82:23)

I mean, I don't remember whether it was Microsoft or Amazon, but one of the two has sort of proposed taking over Three Mile island, the old, the old nuclear plant. I don't think it's a coincidence that crazy activists will show up to protest both data centers and nuclear power plants. And so to them, there's sort of nothing worse than generating more energy and then using it for some sort of grand industrial purpose. I understand the political economy of people being worried about the data center demand influencing prices, but it's not actually true. And you just have to look at the map of California versus Virginia. Virginia is the data center capital of the United States basically. And it has utility prices that are more or less in line with where you would want to be. California's have gone up massively over the past few years and it doesn't take long to investigate why. It's because they're shutting down nuclear. It's because they're making it difficult to do cheap energy, lowering prices across the board. Affordability, you really can achieve it by just letting markets work. Almost all of the so called affordability options are in fact going to increase prices, especially when politicians are the ones coming up with this is how we're going to make energy more affordable by forcing all of these new rules or whatnot. No, it's not going to work. When you have this massive industrial thing that's taking place. If it can create a massive supply boost, then who's going to benefit from that glut? It's going to be all of the consumers who also want to use natural gas power or whatnot. And there's all this increased supply.

B (84:11)

How are you thinking about Arena Mag and the balance between contributors, full time writers, researchers? How are you designing the shape of the newsroom?

F (84:24)

Okay, so the latest one is the biggest issue ever. It's 128 pages. We used over £10,000 of paper in printing it. We have great contributors. We want people to send us more. I would say at the beginning we had to go and sort of hunt down every single article that we wanted. We're very lucky to get a lot more submissions these days. But the truth is that we have a lot of readers who actually read the articles in print and don't want to get spammed 10 or 12 times a day with new articles. And so arena is like a high end media business where people pay us to leave them alone in a certain way. Where they love the quarterly magazine they actually keep looks great on a coffee table. People keep it for their offices and we're working on some other sort of high end printed products that have a mix of contributors, so to speak, and other ways to put things together. I don't think it's going to be a, you know, 100 person newsroom.

B (85:32)

Yeah, leather bound Grokopedia. How about that?

F (85:37)

Yeah, I thought I Ran the numbers and depending on the type size, I think there's some way where you could publish the Bitcoin ledger as like a law code or whatnot. It would fill like an entire wall. But actually, I'll tease something. We're going to be announcing the very first Arena Books coffee table book over the course of the next few weeks.

A (86:03)

There we go.

F (86:04)

I just got the first box of them on a plane from Europe, printed on Italian paper. Bound to be very nice. It's going to be unlike anything that people have ever seen, I think. And hopefully look great on coffee table books in the entertainment world.

A (86:26)

And starters or coffee table books, you'll be on the top of the stack.

F (86:30)

Yeah, you could stack them yet.

C (86:34)

Yeah.

B (86:35)

Well, congratulations, Jordy. Do you have anything else?

A (86:38)

I did want if you have 60 more seconds to get your take on how you think AI impacts original reporting and storytelling. Because in my view it's potentially great for an arena mag because there's so many more contributors that'll think, hey, I have this thing that I want to talk about. I don't really have an outlet or a platform myself, or maybe I do, but I want to share it in print and I can produce a great story in a much less time, even if it's like hopefully heavily written by themselves. And we've talked with other plenty of other folks where I think that even in it feels like today an AI, no matter how good the voice agent is, if an AI calls you and you don't know who the person is and they're just kind of trying to mine you for information, there's not going to be a lot of information flow. Which means that I think that great journalists and storytellers will have great, great jobs long, long into the future. But how are you thinking about it?

F (87:44)

I long for the day when the AIs are actually a better writer than I am, but it hasn't happened yet. I'm honored to be scraped by the AIs so that my voice will live on. I'm a super user of all of these platforms for automating all of the stuff that makes the enterprise difficult to do, which is sometimes dealing with complicated workflows and research and whatnot. I think that there is something around certain sort of like newswire style things that could be automated super effectively, where if you have like a newsroom of people where there's like even like 20 minute delays or whatnot, that could be, that could be improved by like immediate algorithmic stuff. But I think that as the amount of low Quality content on the Internet goes up, people are going to look for things that they trust, and it doesn't necessarily mean print. For us, print is like a very

C (88:47)

good

F (88:50)

place of trust, where the fact that we're actually taking several weeks in a giant factory with a bunch of paper to double, triple check everything, it's like it's a level of care that goes into it that reminds people that things can be done by humans in this really dazzling way. And of course, there's tons of clogged cowork and other stuff going on beneath the surface, but we wouldn't want to let that touch the writing. If people are going to be paying for it, they should be able to get that stuff for free on the Internet and pay for high quality stuff.

A (89:22)

Yeah, but at the same time, I feel like the value. Well, you're saying in some ways, the value of an editor goes up a lot because there's infinite content and an editor is deciding, in this case with arena, what actually makes it to print.

F (89:37)

Yeah, I mean, a bunch of us at arena, we're all editors at the Stanford Review. And so it's like you learn how to. Editing is definitely. Editing is definitely a skill in itself. And being able to point out why someone else's writing is bad, then you can point out how your own writing is bad, and that's how you actually get good. I would say one of the theses around wanting people to contribute to arena and compiling these issues is that while it's a great thing that anyone can just sort of post out there, there's a reason why the legacy institutions that have super high professional standards are extremely effective. Bloomberg, Wall Street Journal, New York Times, even the ones that people disliked for ideological reasons, they have a seriousness to them that makes their messaging super effective. And a lot of that is because they're extremely effective editors. I will say arena probably invests slightly less in copy editing compared to some other institutions. I think it's a sign of life that we occasionally find typos in the print magazine, but we spend an inordinate amount of time actually editing to make it good. Yeah, when people pay you for something, you know, there's an old joke. Boris Johnson, the former Prime Minister of the United Kingdom, used to be a news editor, and he would apparently tell his staff when something was bad. You know, the readers pay us, we don't pay them. And so when the readers are paying us, it's got to be good. And there's a lot of stuff that is, like, assisted by AI or where the Research has been helped with it. Transcript editing is like a super helpful one now and like I can walk into like a company with my iPad and record five hours of interviews. And both Claude and Grok are now very good at taking all of those and doing sort of very light style edits or fixing the sort of verbal pauses or whatnot which would, which would previously take me like many, many hours to, to do so. That's an example of like my job's like a lot easier but people are paying us for like a high quality product and so we want to make it as good as possible both in the writing and especially in the, in the art as well. We love our non reading customers. You know, there are a lot of people who love arena who haven't read

B (91:59)

a single article that's the coffee table

A (92:00)

book and can't read a speech.

F (92:03)

That's exactly Arena.

A (92:05)

You don't, you don't. You can just look at the pictures.

B (92:08)

Well, you can find it at arenamag.com, you can subscribe to the print edition for $99 a year. It's an absolute steal.

F (92:16)

I really think it's a steal. I was telling people the other day, you know, I spent three months in an attic in the Texas summer going through different like paper samples, choosing the exact size. I think I nailed it.

B (92:32)

Well, thank you so much for coming on the show and sharing this with us. We will talk to you soon. Great to see you.

A (92:37)

And I love the vest.

B (92:38)

The vest, fantastic.

F (92:40)

Every media man needs a look. I bought 12 sweater vests on ebay earlier this year. And it's working.

B (92:50)

It's working. Well, we'll talk to you soon. Have a good rest of your day. Czarna, goodbye.

A (92:55)

Great to see you.

B (92:56)

Max, let me tell you about sis. Unlock critical infrastructure for the AI era. Unlock seamless real time experiences and new value with Cisco. We were talking briefly about Warner Brothers. The Kalshi market is mooning on Paramount. Paramount and Netflix have been going back and forth. Paramount's now at 61% chance that they will be the one to successfully take over Warner Brothers. Netflix previously up at 70% back in December when we started talking about this. Now they're down at 30%.

A (93:29)

Allison's are doing the Allison thing.

B (93:32)

It ain't over till it's over. Let me tell you about console. Console builds AI agents that automate 70% of it. HR and finance support giving employees instant resolution for access requests and password resets. And without further ado, we'll bring in Ben Lair of Lair without Ventures. He's the managing partner there and he's returning to the show we had you on a couple weeks ago. I feel like. Like couple months back. It's been months. No.

A (94:00)

When was it? End of last?

B (94:01)

Say like six weeks.

G (94:03)

This is much more legit.

B (94:04)

It is great.

C (94:05)

Yes.

B (94:05)

Here in person.

A (94:06)

Real. It's a real place.

B (94:07)

Yeah.

A (94:07)

Not AI.

B (94:08)

How long are you in town for?

G (94:10)

For the week. Although I was supposed to be here three days ago, but it's.

B (94:13)

It's snow.

G (94:14)

It's up front.

B (94:14)

Did you get snowed in?

G (94:15)

I got snowed in and really had flights. Actually six flights canceled, five flights canceled.

A (94:19)

Not.

G (94:19)

I was fortunately not at the airport.

B (94:21)

Maybe I don't travel that much, but I actually think I've never been in that situation.

A (94:24)

That's the charter. The charter gods. The charter gods telling you it's time to like.

G (94:30)

Thank you.

B (94:31)

Please. Okay. Gulfstream salesman. Gulfstream salesman. Seriously. But yeah, yeah.

A (94:38)

I was. Ran into somebody at coffee this morning and they told me it was up front.

B (94:43)

It's up front.

G (94:44)

Everyone on the plane was going up front last night.

B (94:46)

Yeah. It's a big event.

G (94:48)

You do a great job. It really is a. It's one of the few times when you can go out and see everyone who works. Not quite everyone, but the majority of people that I would consider colleagues in one room for a day. It's actually really.

B (95:01)

Is there cross pollination outside of technology with other industries? I know Upfront in the past has brought Hollywood LA folks, but there's other conferences where it's more like Hill and Valley. Right. So they bring the politicians out a

G (95:16)

little bit of that. I mean, I think they do play the LA card pretty hard. And Upfront's brand as a fund is very sort of LA centric. The event, they definitely like take advantage of the fact that there's that kind of talent out here. I don't think a lot of speakers are flying in other than people from the venture world.

B (95:30)

Yeah, yeah.

A (95:31)

But the tension between Hollywood and the people funding AI, I could see that being interesting.

F (95:40)

Yeah.

G (95:40)

You know, we're actually in a company that's building a sort of AI first movie studio called Staircase. And what they are doing is trying to use Hollywood talent. So they're using like SAG talent, real actors, real voices, but then enhancing it with AI. And they're getting some. They're getting a little bit of pull out here in a good way. Whereas I think if you're sort of. If you're not willing to use Hollywood talent and pay Hollywood talent through the Unions you run directly into a wall. And by the way, like, this is not gonna end well if Hollywood doesn't get with it. I mean, at some point it's just gonna get steamrolled. There are gonna be look at like a 24 bringing in Scott Bell.

A (96:21)

Yeah, I just looked at it as. When I living in LA for the last eight years or so, I've met so many directors and producers that are. That were kind of frustrated with the lifestyle of, okay, I have this job that I love doing, but then multiple times a year I gotta fly to the Middle east and just be posted up in a foreign country away from my life. And there's some element of that that I think is kind of like core to the movie industry. It's like, you know, I think there's some. Probably will always be nostalgia around it. But we were talking probably a couple weeks ago, at this point, if you're an actor and you lean into AI, you can potentially get a lot of the benefit while kind of compressing your actual time doing the work down by quite a lot. Because it's like, hey, come in. I don't know exactly what Staircase's workflow looks like, but I can imagine we get to the point where there's certain scenes that are done like, you know, legit the old fashioned way, and then there's certain scenes where it's like, yeah, we're just gonna use for Staircase.

G (97:31)

None are done the old fashioned way.

A (97:33)

Okay.

G (97:33)

So everything is.

A (97:34)

Do they do like a capture process?

G (97:36)

They do a capture process, but never, always against a green screen. So never against or. I don't even know if it's against a green screen or just against. Against no screen, but with like tracking movements and whatever that would be. But they're not ever going and shooting out in the wild. And the technology is changing so quickly that what they can create today, relative to what they could do six months ago, is magnitudinally different. And if you just continue on this curve, at some point, pretty soon it's going to be ridiculous to think that you're going to go to the Middle East.

A (98:07)

Yeah. Matthew McConaughey has had some good leadership around this, basically just being like, hey, we have to embrace it. It's coming. It's just too efficient. It's too productive. The idea of like, hey, we need to shoot. We have like a three minutes total that's in this setting. We're gonna bring 100 people out there and spend weeks doing this thing that could be made in not one prompt, but made in a series of prompts and with a lot of editing.

G (98:34)

Well, and there's an insatiable appetite for just, like, more content. I mean, you guys were just talking about the, you know, the Netflix or Paramount setup, and, like, all these platforms just want more and more and more and better and better and better. It seems like there's no limit to the amount of content that people want.

A (98:53)

And so, yeah, the average user. How many times have you opened a streaming platform and just been like, I don't like any of this stuff?

G (99:03)

Which is amazing because there's so much more content created today than ever before. And still, I flew out here yesterday, I turned on my Netflix, and I'm watching a show from, like, 11 years ago. Yeah, it's ridiculous.

B (99:12)

Jordy, you watched the movie last night, right?

A (99:15)

John. John was making a joke about the movie Borat, which I loved as a

G (99:20)

kid, by the way.

B (99:21)

Which you should still love.

A (99:22)

Yeah, it's a fantastic movie. But I went on Amazon and I bought it because I was like, I should own this. It's a cult classic at this point. And it was not as funny as I remembered it when I was 12, which was painful to.

D (99:35)

To me.

A (99:35)

But I'm curious on talking generally, maybe, or about staircase right now, a movie studio will allocate, let's say, $50 million to make a movie. How do you think that number changes over time? Because in my head, I think eventually it could be five people. You have writers, you have whatever. The new director's role probably stays somewhat similar. And then you're pulling in other talent for audio and video and all these different things. But how low? And the reason I don't think that's a bad thing is there's so much demand for content that theoretically, that $50 million budget would still exist or maybe even increase, but it would just be spread across.

G (100:25)

This is the question not for. For only Hollywood, but for everything, which is like, will AI influence when you have things big or small?

A (100:32)

Yeah, right. When you have deflation, does that create.

G (100:35)

Yeah, I mean, it applies to every industry. I think probably with Hollywood, you'll see Frontline talent continue to demand huge premiums. So maybe that $50 million. 30. But Brad Pitt still makes 25 of it. And the rest of it, you can do much less. I don't know if it's fewer people or the same number of people being much more efficient and working in half the time. Like, maybe you're doing it more quickly. Maybe you're doing multiple projects at one. I don't really.

A (101:02)

Yeah, the Question around the talent side that I'm interested in is it's one thing for Brad Pitt to be like, cool, you're gonna make this movie with mostly AI, But I'm still getting my same rate. Otherwise I'm not gonna be a part of this because some other movie studio down the road is happy to pay me what I think I'm worth and I don't care if it's going to take me way less time. The question is, for new talent, the people in LA that are working as a. As a waiter and just like, you know, constantly trying to get into the game, what kind of leverage do they still develop? Can they still develop real star power over their career and get pricing power? Or, or does that. Or does like the. The amount of people that can be a star fragment even more? Like, does there become more of like a. I don't know how much of a middle class even exists anymore in Hollywood, like specifically on the actor side?

G (101:59)

Well, I think I have a few thoughts on this. One might be that probably there is more of an impetus for talent to be able to not only get famous through the channels that Hollywood provides, but to use social media to use.

A (102:14)

I think that's so important, build your

G (102:16)

own distribution to be able to demand sort of like that premium. Because people think that you can move the market, you can drive sales or you can drive views for your own stardom.

A (102:26)

People still care. Doesn't matter what the movie's about or who the actor is. They care about who that person is outside in the real world too. And that influences their experience.

G (102:36)

I think that that's going to be a big part of this for that. But also Hollywood is going to move more slowly than just about any other industry. Part of it is the role of the unions and how embedded that is and how that kind of content gets made. Different than certainly most other industries that don't have that kind of bureaucracy or lock in. And also it's an industry that's run by moguls. It's run by old people who are like generally slower moving. It'll be interesting to see tycoons.

B (103:09)

How do you think about vertical short form? I saw a crazy video of someone doing a movie screening, like in a theater, but for vertical video that they made. That sounds terrible, which is insane. And it was really crazy because it was like, they don't really make projector screens that are super vertical. So you're in this like massive room, but normally you're in a really wide theater and then the screen matches the seats yeah, but it was like just off to the side, this, like, vertical screen, kind of like a TV we have over there. But. But it does seem like there's quicker pace of adoption for those, like, polished. What was that Chinese app that has vertical short form?

G (103:50)

Like the drama.

B (103:51)

Yeah, the short drama. Real short or real short?

G (103:54)

Real short.

B (103:54)

And then there's AI versus of that. And that feels like we get a

G (103:58)

pitch for that category once a week. You do for the last year.

B (104:01)

Okay.

G (104:01)

I mean, it is.

A (104:02)

Oh, I get it once a day.

B (104:03)

Okay, so you have better deal for a day. I get it once a hour.

G (104:06)

No, but it is. It is wild how. And by the way. And the number of pitches, it's funny. You guys want to talk about media with me? The number of pitches I still get from people like, Ben, you're like a media guy. I'm like, please leave me alone.

B (104:17)

I'm not a media guy.

G (104:19)

Everyone just like, I'm sorry I did this. But there is this moment where this is also like, live shopping was another category that you saw massive in China. And short form drama, enormous industry in China. And everyone's like, cool, it's gonna, like, be here in the US tomorrow. And like, 10 years later, it's still not really here.

B (104:40)

Yep.

A (104:41)

Yeah.

G (104:41)

I will say that in with one of the companies, they like, really forced me to watch some content. And I got hooked on a story about, like, a woman who fell in love with a man in an elevator. And I watched for like two hours in one minute, snippets like, I see the. How you can get addicted to it. But I fortunately broke that.

A (104:59)

The question. Yeah, the question is how. How do you actually invest against this trend? I think, I think, I think, yeah, I think, I think, yeah, I think maybe because all the content. Yeah, the content wants to be free. It wants to flow to the existing platforms. I can see YouTube creating, like, short series. Like, basically, like, you can subscribe to a series being like, yeah, basically, like, hey, you're not even just subscribed. Like, because the follow button and subscribe button, that barely works anymore on any platform. But they might be encouraged to be like, hey, this is a series of 30 videos. It's gonna be coming out one a day. You can subscribe and we'll make sure that it comes out.

G (105:40)

Look, I think this is an interesting space. The difference between an interesting space and a venture scale space is the Grand Canyon. I mean, it is so massive. Figuring out venture requires the power law, requires these outsized returns, and it just doesn't feel like sure, you have one or two companies in China that are massive, but it doesn't feel like that's a category that has a sort of theoretically limitless Tamil. And one of the mistakes that I've made in my career and one of the sort of learnings is TAM really does matter. And yes, there's, you know, you have to sort of lily pad your way there. It's not like every market has to be enormous day one. But most of media has turned out to not be venturescale and I think maybe all of media is not venturescale.

A (106:27)

And with AI, yes, venturescale if you can build a platform that can compete with the giants. But you need a $5 billion.

G (106:37)

But how do you like what is it? Right? I mean this is one of these things. If someone by the way like Jeffrey Katzenberg tried to do this ish with Quibi and there are few people more formidable in the entire world than Jeffrey and even with hundreds of millions of dollars, like that's not. That wasn't the answer. Now by the way, today would be a more interesting time to launch Quibi because of what's changed in technology. And by the way, I've seen the

A (107:00)

like still the Quibi 4.0 but the competitive dynamics with every other app that also has short form videos that the creators are incentivized to share the same content across all of them, those competitive dynamics are still in place.

G (107:14)

Look just like working in an industry that you know, Meta and Amazon and Apple and Netflix that like more than half of the Mag 7 is interested in is a really, really tough place

B (107:28)

to go and build. Every MAG7 company except Nvidia owns a basically a social network. If you include iMessage, Twitch, LinkedIn, like you actually wind up with a media platform at all of them.

G (107:40)

Right. And like that is by the way, this is not your traditional, you know, large incumbents who are asleep at the wheel. These are the scariest companies that have ever been built in the history of humanity who you know, like no thanks.

B (107:55)

How did you process the Citrini virality?

A (107:58)

The viral article sold everything.

B (107:59)

Intelligence crisis.

G (108:01)

I bought a farm in the woods

B (108:03)

and some raw material.

G (108:05)

Look, we are at a moment right now. I mean that's this week's flavor. Last week I think there was something biggest happening. Yes. And then there was the counter to it. And obviously there's a bunch of counter arguments to this one. I think it's indicative of what a scary time this is and just how overwhelmed everybody. I mean like I wake up in the middle of the night, screaming, I don't know about you guys just in fear. No, I'm kidding. No, but, like, this is wild times. I mean, this is. I look at that article and I think that there is. You can, you can sort of pull the thread and go, yeah, like, oh, my God, you are going to see a bunch of these companies who have to cut costs, go lean in, become their own worst enemy. I think what this sort of doesn't take into account is that there's going to be a bunch of other interesting, amazing new companies that get built, that employ lots of people, that have a very different growth curve, and that it's not just like everything is a race to the bottom. There is going to be all this amazing innovation that happens counter to that.

A (109:05)

Also a good counterpoint, Citadel put out a report that showed software engineer job openings are up 11% year over year. And so it's interesting that we're getting this, like, where AI is working the best today and coding is causing an uplift in jobs, and yet people are kind of extrapolating and saying every other job is cooked, even though that's not what we're seeing in the data.

B (109:32)

Yeah, it's interesting. Are you revisiting marketplaces? There was a big debate over DoorDash being here. AI resistant or not?

A (109:42)

Any doordash pitches?

G (109:44)

Not this week, but I've gotten some doordash pitches.

B (109:47)

I'm just wondering how you're thinking about marketplaces, because it feels like there was a boom. Most of the great marketplaces got built. All the obvious ones that didn't have massive disintermediation problems. So the dog walker, the house cleaner, that stuff that gets disintermediated very quickly. But the doordash or the Uber, that is something where you're not just going to meet a great driver and be like, now you're my personal driver forever. Right. But you will do that if you're like, I went to this app and I found a house cleaner and they come every week. And so I just said, hey, let's stop using the app and cut out the 15% take rate. But what are you thinking about marketplaces on the earlier side? Is there anything interesting that you're seeing?

G (110:29)

So I think that there is. When you're looking at marketplaces, we're trying to find marketplaces. Marketplaces where sort of everybody wins. And there's a company in our portfolio that's very early building in the aftermarket automotive space. And this is a category that has millions and millions and millions of skus there is a question as to does this spark plug work for my 1984 Mazda Miata or does that one work? Most of that industry never came online at all. And so you still have to file, fax in an order or a requisition. You have some large sort of endemic players like AutoZone that are actually quite big businesses but that still only have a fraction of the inventory. They don't actually know if that piece works for this car when you've also done these three other changes. And that's the kind of industry, very messy, very hard, but with AI you can go and send agents to go do some of this buying. So sort of mimic the idea that an industry is online that may never come online.

B (111:43)

Sure.

G (111:45)

I think that that's, I think there are spaces that by the way, and that's also a very large, very sleepy tam. That's not a sort of side pocket hobby. That's one of the biggest hobbies in America. Yeah, that's billion, you know, tens and tens and tens of billions of dollars just in the us, just in aftermarket

B (112:01)

just to hydrate from like some random skew or some random serial number into like what is this product actually? And there's probably a manual out there somewhere that has. It might be on the Internet, might

G (112:12)

be in a Reddit forum somewhere. But like, and then there's the question of, and by the way, how do I get this installed and can I do it myself? And you know, is there an AI mechanic that's built into the marketplace so it can teach you to do the work while you buy? Will this company accomplish it? I don't know. But I love the ambition and I think when we're looking for marketplaces we want to see ones that are. I don't want to see an incremental doordash competitor. I want to see one in an industry that really has, that has not yet been disrupted for reasons that were frankly impossible before the automation of AI.

B (112:51)

Yeah. Jordy, you think of what I'm thinking? Me, you, this weekend, Couple of cold ones. Mansory body kits on the car. Let's do it secondhand. We get them, we figure out this

A (113:02)

is John's dream forever.

B (113:04)

We can do some LED under lighting. You got a new car you don't have.

G (113:07)

I'll stay for the weekend.

B (113:08)

You don't have LED lighting underneath that yet. Nos. Have you considered putting NOS in your car? This is a good option.

A (113:17)

Very good. We gotta get. That'll be your Christmas gift this year.

B (113:20)

Mansouri body kit.

A (113:21)

I would Love it, you want it, it's gotta happen. The labor marketplaces that have exploded are the Merkors, the micro ones that are positioned as marketplaces and yet feel more like enterprise product, almost like staffing in some way. The question is, how durable will those businesses be? I think a lot of people were surprised that the Fivers and the Upworks didn't react and capitalize on that as quickly as they maybe could. But it just goes to show how quickly the space is moving.

G (114:02)

Yeah, I think there are perspectives, there's multiple camps there, but there are, there are camps of folks, you know, smart folks who think that that whole category is like a race to the bottom and like probably a zero. And then there's other, you know, maybe some of those folks, like Summer Coralie passed or whatever and you know, like are rooting against it. But, but there's, there's definitely a vibe that that is, you know, maybe if you're on the inside in some of those businesses, like now would be a good time to take some secondary. Sure, I think, obviously. But you've seen growth that is astounding. And you know, that's the funny thing about the market right now. That's a category, but there's so many categories. Like a bunch of these companies that are doing inference. Like you've seen this like unbelievable growth and like there's like insatiable appetite for what they have today or you know, you know, different kinds of training data or whatever. The question is in two years, in three years, in five years, in 10 years, like which of these are enduring spaces and which of these are ripping right now? Because they're selling to five, there's only five buyers or four buyers long term. And is that, is that a good business? Like to be in a business where you ultimately have four companies that are your like theoretical scaled customers.

A (115:14)

Yeah. I mean, and right now there's such insane urgency that there's no well, and

G (115:19)

money is absolutely valueless to these companies. I mean, it just, it pours it like as fast as you open the door, it like burst its way into your face.

B (115:27)

Yeah, I was joking about if you started a janitorial company that just served the AI labs, your revenue would be growing 10x because their head count in square footage is growing 10x. And you'd be like, yeah, my business is 10x. And you know what? I have pricing power too. My margins are incredible. It's like, because they never asked, let's

G (115:42)

do that this weekend.

B (115:43)

That's a good bit. Yeah, they were like, hey, Janitorial, business is wildly different.

G (115:46)

There is, I do think that right now is, you know, and look, this is coming after a few years of really disappointing returns for venture as a category and suddenly you see things ripping and growing and like you want to believe, you want like people, you know, like you. And by the way, you have massive, massive funds now that need to deploy huge amounts of money and they're looking for things that look like the escape velocity is there and so they become self fulfilling prophecies and more money pours in. And like, you know, I don't think that all these categories are going to be zeros, but I think probably there's going to be, there's a lot of winners in some of these categories. I think there'll be fewer winners and maybe the winners won't be winning at the multiples that they look like they

A (116:31)

are today or they'll evolve dramatically. Yeah.

G (116:36)

And by the way, some of them will and some of them won't.

C (116:38)

Right.

G (116:38)

I mean like, it's like keeping up with this market is the hardest, it's the hardest in my career to like companies that are, let's even look at like the open air anthropic thing. And granted this is just like, you know, maybe that's, this is like the caricature of the space but it feels like the tide turned so hard even the last three months where it's like anthropics world now and OpenAI is living in it. And three months ago you would have said the exact opposite.

A (117:05)

Yeah.

G (117:05)

And maybe that's just like I'm interested if you feel that energy.

A (117:10)

Well, it's hard to, it's, it's. Yeah. It certainly feels like at this moment anthropic is the main character. Right. Dominating the headlines on X and a lot of legacy media as well. But I would say Google trends tell

G (117:31)

a slightly different story.

A (117:32)

A slightly different story. Right. We are in a bubble and people are so invested in the race that they want. It's like if you watch an F1 race and Verstappen wins every time and he starts in pole at some point

G (117:45)

you're like, I'm sick of this.

A (117:46)

Yeah, you just want to see, you want to see some action and you want to see some passes and you want some, some drama. So I think people are invested in the drama. But yeah, and it's also, you're, you're watching the two strategies play out like a very multi product approach. Consumer enterprise hardware from, from OpenAI versus like in a very focused strategy. And I think it's way too early to kind of understand what will in hindsight look like the best call. But that is going to come out

G (118:21)

with their model whenever and then what does that look like? And yeah, yeah, it's. It's interesting times.

B (118:27)

Very interesting. Well, thank you so much for coming down to the TVP and Ultradome.

A (118:31)

Give us a report on. On upfront.

B (118:34)

Okay.

G (118:35)

Happily.

A (118:35)

Is it tomorrow?

G (118:36)

It's today.

A (118:37)

It's today. Oh, you.

G (118:38)

I abandoned my.

B (118:40)

Abandoned my boy.

A (118:43)

Well, later. And we'll. We'll give the update on the show. I want to understand. I have no sense for how much venture capital like what I think of L. A is getting. I have a good sense for like Hawthorne, El Segundo, Longor Beach, Gardena. All these areas feels like are raising as much money as ever. But the broader la. It feels like a drought.

G (119:07)

I would say it's a drought. We definitely are not spending time out here hunting companies really?

D (119:14)

At all.

A (119:14)

Yeah. Yeah. That's the advice I've given to any founder that's not in hard tech is like, do not. Don't even start. Yeah, don't even start your raise.

B (119:23)

Don't even think about it.

A (119:25)

It's going to just send like a really negative signal that you're not super serious.

B (119:29)

Yeah. Well, thank you so much. Thank you on the show. This is great.

A (119:32)

Great to see you.

B (119:33)

Let me tell you about MongoDB. What's the only thing faster than the AI market? Your business on MongoDB. Don't just build AI, own the data platform that powers it. And let me also tell you about Lambda Lambda is the super intelligence cloud building AI supercomputers for training and inference that scale from one GPU to hundreds of thousands.

D (119:54)

Boom.

B (119:55)

We have some breaking news. We got to hit the gong. Riley Walls shared an exclusive scoop. He says he has joined the wonderful Labs team at OpenAI. I'm learning a lot and have enjoyed it very much. Congratulations.

A (120:07)

Huge pickup, huge opportunity. I am so excited to see what Riley is working. One of the greatest minds high agency individuals on the Internet today. And I'm so.

B (120:25)

He's the best. Yeah.

A (120:27)

So I think Riley Walls is like a perfect example of like jobs kind of getting more fake. Like what is like, like this is like. This is like not a diss at all. Says the guy who. No, exactly. I'm also a like very fake shop. Like Raleigh Walt. He like just goes viral on Twitter. He's not even like making like YouTube videos. He's been a.

C (120:46)

He's been a.

A (120:47)

He's been a data Analys for the last few years. He's had it. He's had it. Yeah. But he's not doing data analysis at OpenAI. He's not doing the same kind of work he was doing before.

B (120:54)

Yeah, he's.

A (120:56)

He's perfectly demonstrating. How do you describe power?

B (121:00)

Member of technical staff. Next question.

A (121:02)

No, so. So he is perfectly demonstrating the power of the tools, which is if you have ideas, you can build them really, really, really fast and create super, in his case, super entertaining products. But there's a Riley Walls out there of SaaS and they're probably just hunkered down, not even really trying.

B (121:24)

I agree with you. Very, very hilarious framing. I do think that we're moving into a world where more and more companies will have a Reilly Walls internally, someone who can move very quickly, do experiments, create value and create little projects and the leverage that you get from an individual is going up. And this is an example of that. So congratulations to Riley Walls on his move to OpenAI. Let me tell you about graphite code review for the age of AI. Graphite helps teams on GitHub ship higher quality software faster. And without further ado, we'll bring in the birthday boy. Doug o' Laughlin is in the restream waiting room. Let's bring him in to the TVPN ultradome.

A (122:05)

Beautiful shirt to you. Happy birthday to you Happy birthday, dear Douglas Happy birthday to you.

B (122:15)

How you doing?

C (122:16)

Hey, happy to be here, I guess on my birthday. Nothing better than Nvidia earnings, which hasn't come out. We are going to be live reacting.

B (122:23)

Yes, we'll be live reacting.

A (122:24)

And they planned that, of course, years and years ago on your birthday. They knew that today would be the day.

C (122:31)

They went back in time and they told my mom to get it done today just for me. How you guys been?

B (122:39)

We've been good.

A (122:40)

Good.

B (122:41)

While processing the. Something big is happening. Then the Citrini article, there's been a lot of sort of doom, but also there's glimmers of boom in there. There's. It's amazing.

A (122:52)

I feel like you've generally been a Citrini defender kind of saying, you know, a lot of people were just like, this is bad and it should never have been published. I think your view was there's some good stuff in. There's some. It's a good thought exercise.

B (123:06)

Yeah.

C (123:08)

Yeah. Well, here's the thing is like it went super viral for a reason because it obviously resonated. I don't think really crappy ideas resonate that hard. And I think honestly, a Lot of the things that I'm most concerned about, I would actually align with it. My biggest concern is that we're going to print deflation. And he kind of talks about that like, hey, one data center does all the knowledge work. The history of how this works usually isn't quite that extreme, but that was the point of it. Here's my extreme think case. What I was kind of shocked about and I feel like, honestly, I bet you Shatrini would agree. Just like the level it resonated with, it was covered on Bloomberg almost on a daily basis. So if you're like a pro, whatever that does a push notification. It was like Citrini says and Satrini, co author and all this stuff. I was like, oh my God, it really broke through. Yeah, it really broke through.

B (124:02)

Yeah. Unpack the deflation question a little bit more. So technology has been deflationary for a long time. TVs are getting cheaper is the line. You can see that. If you look back at inflation, you see health care and education inflating while basically everything that's on like a technological Moore's Law style curve is getting cheaper over time. The bigger, broad deflation question is can we move some of those or will some of those inflationary categories move into a deflationary territory? But what are you thinking? Is education, health care at the top of the stack? What do you mean by knowledge work becoming deflationary?

C (124:45)

So let's just put it this way. The ghost GDP article, right? Hey, let's just kind of like do like an index. One index versus the other index. We're going to label it at $100 for all these like knowledge work widgets and let's just see what the cost of that looks like over, over time. Year zero, $100 with a human and maybe year two, it's $102. Maybe they get a little bit smarter so it goes down. But like, okay, what if the agent gets 20 times better? Year two, the index of knowledge work is going to be $20. That's just deflation, right? The entire world is effectively just like spitting out TVs that are worthless in three years. And that's kind of scary and I think is a big deal. I think the thing that is the most resonating with the whole piece is that clearly if this is going to be as big as many people think it will be, quadcode maxi here, then what happens is that you're going to quickly disrupt so much of society that the government and companies and everyone needs to start caring very quickly. I don't think 10% unemployment is going to happen. The world works just slower than that. But I clearly think. Yeah, it's part of my, Part of,

A (126:00)

Part of your framing is like, when we had technology rollouts throughout the last few hundred years, it involved heavy machinery that took a lot of time to diffuse, in part because the machinery needed to be built and shipped around the world and spun up. And then you have. With AI, it's like, okay, the Internet is the greatest distribution engine in history. And as technology advances, it can be live in somebody's office or with somebody's company effectively instantly, which creates. And with like, let's say the printing press, it was, you know, or cars or electricity, it was like a much more. It was forced to be a much longer rollout.

C (126:51)

Yeah, there's just like, for example, the railroad piece I wrote about, dude, it took 50 years to put all those rails right. Historically, let's say this AI thing would be like a historical atoms thing. It takes 50 years of installing the AI widget at your house in order for you to get AI versus now. You just press a button, you download it and it's there. And so the pace of that disruption is so, so much quicker. And so if it looks like, you know, if it's as big of a deal as people are concerned about, then essentially you're going to broadcast deflation everywhere, all the time, all at once, and it's going to be almost costless to get there. And also, it's just, it's going to get relentlessly better. Yeah, just. And then how do you, how do

A (127:35)

you, how do you square that though, with the fact that right now, today, AI is the best at coding. That is where it has the most traction within knowledge. You know, knowledge work. And yet I have, I have not gotten a call from a great engineer saying, hey, can you help me land a job? I'm unemployed. And you see this in Citadel put out a response saying that software engineer job listings have gone up year over year. And so it's so hard for me to square these two things where we could be in this doomsday scenario for white collar work with Jim's Paradox. Yeah, we're not seeing that at all with software engineers. Now some people might debate me and say, well, oh, I just graduated school. And yeah, new grad opportunities might be lower, but nobody is bought when some of these larger tech companies, CEOs have laid off a bunch of people and said, we did this because of AI, it's like nobody's actually buying that it's just a nice story.

C (128:52)

Yeah, I think that's going to be an interesting thing because I think it will be politically and broadly unpopular to say, hey, I just laid off 10,000 people. Because AI, we are humans at the end of the day and you do care about what your neighbors think of you. And just being like, we fired everyone because AI is better than whatever you're like, okay, well now you got to get. You might as well hire a five person security guard. Right? There's a little bit of being an asshole. So I think how it actually works and you're looking at. And I agree with that comment because there's this really great tweet that's like, oh, it's crazy in the agentic coding world, pretty much it pegs the human CPU at 100%. That's how I feel at least is I'm doing more work than ever, but I'm like literally working harder than I've ever worked. The reality is I feel like there is a productivity uplift, but I guess we're all in the sugar high where you're able to do so much work that you're just like going around like crushing it. So these software engineers, yeah, you don't need new grads. And essentially the person who has the seat gets to print more. Let's just use the printing press, right, Instead of writing and you unemploy all the new writers, but you're sitting there just printing all day. And so it really is great for the install base of people who've been doing it, but very terrible for Net New. And I think the Net new is where you're going to see this. Meaning like new grads, new people entering the information, the information services. That's where I think you're going to see most of the carnage. People are not just going to be firing people out the gate. It's going to be like a layoff

A (130:21)

or firing typically occurs weeks, if not months after an executive has decided, hey, we need to make a change here. Just because it's the thing that everyone hates doing. How have you been processing? It feels like the people that, that the Citrini piece resonated with the most are also the ones that have been saying that the AI is a bubble and there's too much investment going into this and it's all going to be a zero. It just seems like two ideas that are difficult to.

C (130:59)

I feel like there's a certain aspect of haterade where when you're a real haterade, you're going to find whatever flavor Supports your thesis. We find this at semianalysis, man. When we say something positive or we perceive as positive, people take it as negative or really honestly, they take it.

A (131:16)

I know, we've had that too. If there's a company that people don't like and you say something positive about it, people respond extremely emotionally to that.

B (131:30)

Yeah.

C (131:31)

And so I think Citrini made the perfect ammo at the perfect time. And honestly, if you look at the stock market. I'm sorry, I keep watching for a video over and over. If you look at the stock market, I think right now it's been a really interesting year to date because stocks are maybe like whatever, 2% off all time highs. Maybe they're all the way back or they're pretty. I forget what the year to date looks like. But underneath the hood in software, it's been very, very, very painful. I think there's a lot of parts of the index that are in a lot of pain right now. And pretty much you're just like seeing these crazy rotations underneath the surface. And so people are very like, angsty. There's no other way to put it. Big things are happening in the stock market.

A (132:17)

The idea that like, oh, you're holding a name and it hasn't nuked 15% yet, you should get out because there's going to be some. There could be a blog post.

B (132:29)

Blog post, yeah.

A (132:30)

And what's the. There's not a lot of ROI on

B (132:33)

how have you been processing like the moats and software companies, like I've always Understood like the SaaS apocalypse, like that narrative. Of course, you know, we got to debate timelines and maybe these things are just shifting to be value stock as opposed to growth stocks. But the stuff with network effects, the door dashes, the paypals, the regulatory modes, it just feels like there's a lot of companies out there that are being more broadly punished. But is there something I'm missing there or do you think that it's just excitement and people rotating into other stuff that they might be learning about, like memory and energy and semis.

C (133:12)

I don't know, but I definitely think that that aspect is definitely there. Like the ones that are like really shocking to me is insurance brokers. If you ever follow that space. Like once upon a time, as a hedge fund analyst, I followed Aon. It's like one of the most boring spaces of all time. Real estate services.

B (133:30)

Real estate services?

C (133:31)

No, like cbre. Cbre. Essentially an insurance broker. You're having all these things, these network businesses that are just effectively being like yeah, in fact it's going to be nuked. I think the hard part about that is it's complicated. I think there's a lot of advantage for number two to defect really quickly. That's probably the most valuable thing you can do. For example, we've written quite a bit about this to our higher paid tier service stuff. Walmart defecting to agentic commerce makes a lot of sense because they're not Amazon. Everyone that isn't number one should be defecting to win market share and that makes a lot of sense to me.

E (134:07)

Yeah.

C (134:07)

So like it wouldn't be doordash. Doordash would probably put up the, you know, put up the walls, do their best to like have exclusive products and then someone else will vibe code something on the margin and essentially try to win market share by.

A (134:20)

But at the same, at the same time Dara was on the show and his stance was yeah, I don't really care as much about my ads business, about making sure that I'm just everywhere. And so he was seemingly pro agent even though ads businesses is real.

B (134:38)

How have you been processing agentic commerce? It feels like this should like the tech is there but consumer adoption takes time. Yes. You can touch up a photo with Gen AI. Not everyone does. What does good look like because we're seeing stats from Shopify. It's like still in like 0.01% of checkouts are agentic and I would expect that to 10x or 100x but even then you're talking about like 1% of E commerce. Like we're still talking like really small numbers like how fast do you think it ramps and what do you think happens this year?

C (135:17)

So I think the place to watch and it's probably the most interesting place is China because you can argue the leading e commerce adopter has always been China. They were doing doordash way before it got hot in the United States. They were doing e commerce way before it got hot in the United States. And we just saw this. Yeah. Live streams. Right. Like, like dude TikTok and Douyin was always, always been on the bleeding edge. Like I would argue for like global Internet online consumer preference behavior. China is the leading indicator. Like America is a laggard boomer. You know, like we just don't, we don't buy like we used to. Okay. The consumer is not quite, quite as young. We'll just put it back that way. I think what's interesting is we just had this giant incentive out of like I believe it's, I want to say it's tencent. Tencent. This. Like, I'm. Now I'm probably going to misspeak the bubble tea promotion. Okay. They did like 10 million or whatever bubble teas. And so I think there's an example that we finally have the first wave of people possibly being able to adopt it. And that's going to be like all eyes on that for if adoption actually kicks off there. Like, that's what happened in 2015 to kick off mobile payments. Actually, during the CCTV gala, they essentially gave away, like, payments and they incentivized people to use the mobile wallets. And that was like the beginning. Okay. So I think that same analogy is probably where people need to look towards for what. What could be the adoption curve for agentic commerce.

B (136:47)

Is that earnings? Do we hear that earnings came through now?

C (136:50)

I'm just.

E (136:51)

I'm just.

B (136:51)

I'm. I'm just locked in.

C (136:54)

I'm locked in, bro.

A (136:56)

We'll have this sound effect.

B (136:59)

So how do you imagine agentic commerce actually rolling out in China? Because there's a whole bunch of AI labs, Deepseek, High flyer. Do they have a big consumer footprint already? Because they don't have a lineage there. So you might expect agentic commerce to be driven more by the platform that already has a billion daily active users over there. But how do you see the stack piling up over there?

C (137:28)

So I think the best example of this is Alibaba, which actually is an e commerce company. They're like the Amazon edition. They have Quinn. They have a delivery app, they have an inventory app. They have a fully vertically integrated stack and they have AI on top. They're gonna make money. If people just use their product, they can get transaction fees. And so that's how I kind of think about it. I mean, in a perfect world, if Amazon was not bad at models, they would be crushing, right?

B (137:58)

Yeah.

C (137:58)

You'd obviously be like, don't talk down on Rufus.

B (138:01)

Don't talk down on Rufus.

A (138:02)

Just because it took down AWS once or twice or three times for hours doesn't mean they don't got it.

C (138:10)

I actually want to talk about conspiracy theory because I feel like this is. Have you been noticing the instability of public clouds?

B (138:18)

Yes.

C (138:19)

Am I crazy?

B (138:20)

No, I don't think you're crazy. No, it's either.

G (138:23)

There's two.

C (138:23)

There's only two things that could be vibe coding. People are just pushing crap on prod and they have no idea. Probably true. Number two, CPU shortage. I think it's a CPU shortage.

A (138:34)

You don't think there's like a third if we're wearing the tinfoil hat, which is the world's been very unstable and

B (138:40)

there'd be, there's more cyber attacks and stuff.

C (138:42)

Iran maybe. That's actually, that's a good one. But I'm really surprised to see all three at the same time. And then like my favorite, probably the most public version of this is the GitHub instability. Dude, GitHub is so unstable right now.

B (138:55)

Interesting. Yeah, I saw the uptime numbers. It was like below 99% which is really low considering.

C (139:00)

Oh dude, it was like 90%.

B (139:02)

90%, that's really low. I mean people are pushing a lot of code to GitHub. So yeah, walk me through, walk me through the CPU short because is that just CPUs that are being used to scale new like EC2 instances and like normal, like you need a Linux instance and so you need a CPU to provision or is the AI boom sucking CPUs in like a Grace Hopper demand? Like the CPUs are getting dedicated to AI server clusters. Like where are the super. What's the shape of the shortage?

C (139:32)

Basically I think the shape of the shortage is partially on RL because if you want to do an RL gym, meaning like you want to have an Amazon.com for, for your, you have to literally simulate just a lot. That's one real demand driver. I don't know how to quantify that, but clearly they're out of it. Number two, I do think all the code. Have you seen the charts? Like I think it's fc.com with all the new apps coming online, they require infrastructure to run and I had a third one. Oh. The other thing that's interesting is it's also a little bit of like a lapping supply chain thing. So the last time we bought a lot of CPUs was in 2020 and 2021. Historically you depreciate the CPUs on a five year cycle and so after five years you just like literally throw them away. It's five years. So all of the infrastructure that we purchased and we've been trying our absolute best not to purchase any new CPUs and spend all that money on GPUs. So they haven't been investing this entire time and now this slight demand curve comes up and that's enough to essentially sell out all the CPUs. I think that that's probably part of it as well. It's going to probably be this multi problem thing. But it's like one of the more interesting like conspiracy theories like YouTube went down. I don't remember a time in like the last five years YouTube has ever gone down.

B (140:51)

Yeah, yeah, yeah, yeah. That is crazy. Take us through the effects of the CPU shortage in the TSMC context. Because TSMC seems to be like a crazy battleground. Apple, Nvidia. And then you'll talk to a new semiconductor company. Like we had Maddox on the show yesterday and he's like, yeah, I'm getting line time at tsmc. And then you read Ben Thompson and it's like, well, they're not really investing in Capex. And it feels like, is TSMC above this shortage as well? We got Nvidia earnings. What happened?

A (141:25)

We got a revenue beat.

B (141:26)

Okay, we will jump back to TSMC. But they beat revenue. Let's review and read through.

A (141:33)

67.4 billion against an estimate of 66.2 billion.

B (141:39)

We have some breaking news. Nvidia has announced earnings. Let's take.

C (141:44)

Wow, you guys are faster than me.

B (141:48)

Nothing liver than live here. We will all read the earnings and I will tell everyone about Plaid. Plaid powers the apps you use to spend, save, borrow and invest securely connecting bank accounts to move money, fight fraud and improve lending. Now with AI

A (142:08)

and back. Do you got anything yet, Doug?

C (142:11)

Am I crazy? I don't have anything.

A (142:14)

Maybe Tyler just sent this to us. Somebody just like refreshing Twitter. It's possible somebody's just putting up is just engagement farming maybe and trying to.

B (142:28)

They might have gotten us. They might have had us in the first half. Well, we shall see. Anyway, let's go back to tsmc. So the nature of the TSMC bottleneck with regard to CPUs, is that an important factor or is there more fab capacity across global foundries, Intel, Samsung to sort of meet that CPU shortage?

C (142:53)

So global foundries is nothing.

B (142:54)

Nothing.

C (142:55)

They don't have a leading edge chip, so they will not have any CPUs there. Intel is actually the real winner, I think because essentially like TSMC is locked up. Everyone's fighting for space at tsmc. Like Nvidia has the most. All the co op is accounted for. All of N3 is accounted for. All of N2 is accounted for. Essentially they're completely sold out. They're going to invest a lot more but like, you know, get in line. It's another year. So whatever is left over kind of goes to the other foundries. And that's really two companies. That's intel and that's Samsung. I think Samsung right now probably makes way more money if they spend it all on memory. So intel really is the swing provider of CPUs in the world. And that's like kind of amazing. Like I'm kind of my, my belief in a sweet karmic intel victory is all the CPUs that can't be made at TSMC because the accelerators are being made instead go back to intel and they just make them but they don't design them.

B (143:51)

Yep, make them but don't design them. So they would be a second source for Nvidia potentially as well.

C (143:58)

Well, like it's not just Nvidia. I think, I think Nvidia is going to get there. Yeah, I think Nvidia is going to get there.

B (144:04)

To Intel. Yeah, yeah. Oh, oh, tsmc. Okay. So they don't need to. Okay. Because there was a theory about like hey, maybe twist Nvidia's arm, get them to dual source at intel and that gets the flywheel going a little bit more just as like a give I guess.

C (144:20)

But there's been some conversation about the MVL link with, with intel specifically co packaged. But I think it's going to mostly be Vera. Like I think it's not going to be. I think TSMC is going to give them the capacity. I think there's a whole world of chips that are not that like think about Graviton or the Axion project at Google or all the arms.

B (144:46)

Explain that Google project. I haven't heard of that.

C (144:49)

There's a custom cpu.

B (144:50)

Okay. Yeah.

C (144:52)

So like all of these.

B (144:53)

Don't they have like a YouTube chip? They have like a number of custom CPUs, right?

C (144:56)

Yeah, yeah. VCU. I think it's a video encoder whatever unit. But that's like really de minimis. I don't think that adds up.

B (145:03)

How important is the. Is the Grox Cerebras, these faster model on a chip companies and does Google have an answer for that?

C (145:13)

I. Okay, so to be clear, semi. Semianalysis house view wasn't that it was like the most important thing. And we've always, you know, we have that chart about batch. Right. The fact that hey, doing more for everyone is much more valuable. Clearly there seems to be a little bit of a niche market and a niche market demand and so cerebras and LPUs definitely have some kind of demand pull and I think we will see some kind of announcement at gtc but definitely wait and see. That's kind of my impression. That's kind of what we wrote about today, by the way. I think it's crazy. I wasn't aware that the Vera Rubin post would come out until Today, literally, I learned today and I was, oh. So I had to read it live with everyone else. I think that's a big deal, too. And we have a lot of information, information there about what's coming at gtc.

B (145:59)

Well, we're excited to follow it. Nvidia earnings are usually out by 20 minutes after the hour. 4:20. But it appears that they are delayed. So we will see. I mean, of course, we'll have to have you back on the show to deep dive everything that's happening in the semiconductor.

C (146:17)

This makes it so much better. I had like the bogey and all information and what the setup is going to be. And honestly, if we have two minutes, I'll tell you what to expect.

A (146:26)

Okay. Yeah, please hit us.

C (146:28)

I think everyone thinks the guide will be 75 billion. That's what everyone's really obsessed about. I think the questions that are going to be answered or asked rather is about memory price inflation. That's going to be a big deal given the fact that I think memory prices are like, you know, mooning right now. HVM and dram. I think Nvidia locked it all in. So everyone is going to be like, how sure are you your gross margin isn't going to go down? And they're going to be like, we're really sure. But they're going to be like, okay, how sure and what time? And then there'll be like this random circle that nobody can't answer. And then I think one of the questions that people are going to definitely ask is about the power side. It seems like next year Nvidia has more chips than power. And so that's going to be, I mean, to be clear, Nvidia has no power, right? They have customers who acquire power. They have. And, you know, one of their big competitors in terms of market share, which is Google, has chips and power. And so I think that that's going to be a huge bottleneck. And I think that's going to be one of the bigger questions that happens this quarter. Hey, how do you deal with the power bottleneck? Because you're going to sell a lot of chips, but if you can't even plug them in, this is a huge issue.

A (147:33)

And is the idea, is, is the idea that even if lab or a hyperscaler said we're actually good on chips right now, we don't have the power for, for them, Nvidia would be like, well, you actually have to buy them if you want to maintain priority. How do you think they would play that?

C (147:49)

I don't know. I mean I think the thing is it's not going to matter because the only company with a chip to sell you next year that is not accounted for is Nvidia. And I think the shortage in GPUs is really underappreciated. There's like, you know, you can see it publicly. I forget which research shop shows it like the availability of B2 hundreds but our anecdotal information as well as the like the live price tracking and contract price that we're aware of, there's like no H1 hundreds for sale. Like H1 hundreds are sold out. That's like boom. Your five year depreciation like bear thesis that everyone was really freaking out about last year. Well, good luck. You can't even buy one today. So that's probably the biggest interesting thing and I think probably the single most bullish thing you can say. Like a four year old chip effectively is completely sold out today. What does that say about the next generation?

B (148:41)

It's remarkable. Well, thank you so much for taking the time to join. Bummer that they delayed on us. But next time we'll time up so that we have even more time to hang out.

C (148:50)

I'm so happy, honestly. Live reacting to Nvidia earnings.

B (148:53)

I was like no, no, no. We'll get this flow going. This is always fun but I mean just great stuff to talk about everything and happy birthday and we'll talk to you soon.

A (149:02)

Happy birthday.

B (149:03)

Have a great rest of your week.

A (149:04)

Great to see you.

B (149:04)

We'll talk to you soon. Let me tell you about Gusto, the unified platform for payroll, benefits and HR built to evolve with modern small and medium sized businesses. We have our next guest, Marc Benioff in the Restream waiting room. Let's bring him into the TVPN ultra mark. Thank you so much for being first to join us here on earnings day.

A (149:25)

Honored.

B (149:26)

Great to see you.

D (149:27)

How's it going?

B (149:28)

It's going fantastically. Over.

D (149:29)

Did you get that Metallica album I sent you?

B (149:31)

We did, we did. Thank you so much.

D (149:34)

Did Jordy listen to it? Did Jordy listen.

A (149:36)

We need a, we need a, we

B (149:37)

need a record player.

D (149:38)

Jordan again. But mogged again.

A (149:42)

Frame.

D (149:42)

Jordy again. Again. Jordy. I have no words.

B (149:46)

But. But I have no words.

D (149:48)

Jordy.

B (149:48)

Jordy can actually play the guitar. And so next time you're on, he's going to be playing a cover.

A (149:54)

I'll play you. I'll play you a tune.

B (149:56)

Yes.

A (149:56)

How about that? You got.

D (149:58)

I know you got a Zolridge to come on the show with Us.

B (150:00)

Oh, that'd be fantastic.

D (150:01)

I think it would be really good.

B (150:03)

We'll have to make that happen.

D (150:04)

But I'm going to call out. I'm calling out Jordy on the whole situation. Yeah, Jordy, you're still unforgiven.

B (150:15)

Well, anyway, thank you so much for joining us. First on earnings day. Take us through it, because I got a mallet here that's itching to hit a gong.

D (150:25)

Well, if you want to hit a gong. I mean, no enterprise software company has ever given guidance for $46.2 billion before. And these are crazy numbers. So. I love that gong, by the way. I love that gong.

B (150:47)

We love gongs.

D (150:47)

I really do. But I would say that that's exciting.

B (150:51)

Yes.

D (150:51)

But also, just this company, you know, it's really become a cash machine as well with, you know, we're projecting over $16 billion in cash flow this year. So when you think about that and then the quarter, you know, with a quarter, we deliver this record. Rpo, funny thing, you know, some of my friends are writing these articles. Rpo doesn't matter. I mean, they write a song. Nothing matters. I don't know. Nothing else matters. Yes, but rpo, which is, you know, kind of our remaining performance obligation. These are contracts that we basically have signed but not yet recognized.

B (151:27)

Yeah.

D (151:28)

That is $72.4 billion. So these are driving these huge numbers. That's.

A (151:35)

What does that mean?

E (151:36)

What does that.

A (151:37)

I think people have a good.

G (151:38)

That's.

D (151:38)

That's up 14% year over year. It's that all these numbers are accelerating growth. So I don't know. I just kind of. I. I am very proud of my team. I'm very proud of our customers. I say also, I'm very proud of the customers because we've really been pushing the customers hard this year to deploy all this new amazing AI and agent technology. And we've hit basically, basically now more than 19 trillion tokens. So you can just see the velocity of AI and agents, and the company is just transforming from being not just an apps company. And I think you guys are using Slack and other Salesforce apps to run your business, but now they're all extended with these agents like Slack, not just Slack Bot, but the ability to extend your service and your marketing and your sales. And all the agents are out there and they're running wild as well. So you have apps and agents, humans and agents working together. It's very cool moment. Very cool.

B (152:33)

Jordy.

A (152:35)

Yeah. What has it been like internally with the team this year? It's been. Has there Been a more kind of chaotic period in your career. How, like, how are you guys operating internally? Clearly delivering results. Oh, yeah.

D (152:53)

I mean, well, you know, you know, we've all been reading about the SAS apocalypse, but we've got our sasquatches eating our SAS apocalypse.

B (153:03)

Let's go, Sasquatch.

D (153:07)

I just think that when you look at things like, you know, Agent Force, which we've talked about on the show now three times, you know, starting at Dreamforce, that, you know, and first of all, Agent Force by itself is now an $800 million business up 170% year over year. Wow. So that is amazing. All as its own product. But then you look at Agent Force and our data business together, that is now a $2.9 billion business, up 200% year over year. So there's no question that AI and data is a huge driver of growth. And it's about these apps and these agents. And, you know, we use the apps. We're these humans, we're using the apps. You know, we're using Slack, we're using sales, cloud, service cloud, we're using all of our cool apps. And then each one of those now has an agent platform also. And these two things together is the future of enterprise software, that apps have been extended by agents. And while we before were in the apps market and that's what we've been doing for 26 years, you know, that we've been in business since 1999. Now we're in the apps business and the agent business. And I, this is why I've never been more excited about my business. I just love it. Yeah, I mean, I really love it.

B (154:20)

How much similarity is there to the original messaging? Just around, like, being a company that enables cloud adoption. Like, there were probably a lot of companies and CEOs that came to you early in your career that said, like, I know this cloud thing's important, how do I do it? And you had an answer. And now there's CEOs that say, I know this agentic thing and this AI thing is important. How do I do it? And you probably have a pretty similar answer, right? Is this history repeating itself?

D (154:49)

Oh, my God, it's such a good question. Anil Bushry is now the CEO again of workday. My good friend, he lives basically next door to me, came over last night, we both had a cocktail because we're looking at his after hours stock. And I said, neil, there's no way this can be true. You have to let this go. And in fact, you saw already that his Stock corrected today because the numbers are just don't match. You know what's really going on. He has an unbelievable business. He had a great quarter. He's going to have a great year. We use his HR and financials. And I said, Neil, this is just something that you have to let go of. This is not my first apocalypse. I saw this in 2008. I saw this in 2001, 2000, 2007, 16, you know, and look, this is. There's people in the market, they make money when the market goes up and down. So that's just the stock market. But let's talk about the customer success. And when you look at how companies can actually be better, more productive, successful, profitable companies, and we are, you know, we're, we're number one. We're a customer zero. And that's what I'm so excited about. Because when you look at how we're running customer service and support right now, we're using it with service agents and service apps. So if you go to help.salesforce.com, you're using the agent and at any point, bam, bam, bam. You can auto escalate right back to the app and the humans, if you like exhaust the agent or. Now this week we have an agent that is going to qualify 50,000 leads for our company, which is our sales agent, that it's out there talking to our customers. We've enclosed millions of dollars of business this week just through the agents themselves. So that is what is amazing, that we have apps and agents. And it's not that we don't have 15,000 salespeople at Salesforce. We do. And we have millions of apps all out there scurrying around looking for opportunities and then bringing them to those humans going, hey, look at this opportunity. Give this person a call. Let's go see this person. Let's go find out what to do. And that is really the.

A (157:00)

Yeah. How much? How much?

D (157:01)

Extended. Elevated. We're elevated by AI. We're made better. We know we're made better by AI.

A (157:07)

There's no such thing. How much have you been kind of, you know, all the Citrini's post. We've talked about it at length on the show. Very, very, A lot of doom. But we've been very focused on what's happening in coding as these coding models have gotten better. People want to hire at least the data shows so far Citadel was showing up. Job listings for engineers are up 11% year over year. You've talked before about hiring more sales reps. Because as your reps get more productive, probably want more of them. But is that kind of comp that you're looking at, given that? I think everyone is expecting sales agents to kind of get to kind of catch up to coding agents in terms of capabilities?

D (157:55)

Great question. Amazing. And you know, we were. I wasn't really on the show at the beginning of the year, a year ago with you guys, but if I was, what I would have said was, I'm not hiring more engineers in fiscal year, you know, 26, the year I just passed, because I was using coding agents and I was allowing the productivity from the coding agent to give me the extra capacity that I needed for the year. And I didn't hire more service agents in the year. I held it flat and then reduced it slightly because I'm using service agents. But I did hire like almost 20% more salespeople this year. I think we've talked about that. Because I need more capacity. Because we have more demand than ever through every market from the small, medium, large customers, you know, guys like yourself who are like these great entrepreneurs building a great business like tvpn, really going all the way through it. You need a technical infrastructure around you to grow your business. I know you use Slack, I know you use other products. And that's our job, is to make you successful and to really bring in the apps and the agents. You can't do it just with an agent. You need that. You need. We, we. There's still some humans around who need to be automated as well.

B (159:03)

Yeah.

D (159:04)

And that is what is exciting.

B (159:05)

Yeah.

D (159:06)

And that is what I'm doing every single day.

A (159:08)

What advice do you have for a company like Anthony that's hiring a Salesforce Admin? What makes for a great Salesforce Admin?

D (159:20)

You're kind of leading beyond. I mean, you know, it's kind of funny, right? Because these AI companies, they love our products and they can't buy enough of them. They're some of our largest customers now. Anthropic, OpenAI, Google, Amazon, you name it, these tech companies. Slack is the largest AI ecosystem in the world. You know that. And that's reality, you know, which is that no one has a company that's running entirely on a large language model. Because it's not real. That's not. We need software and we need large language models. We need the determinism and the programmability and the security and the sharing and. But this large language model is an amazing new component of our infrastructure. So we can do things that we could never have done. Before that is what is so awesome. And so we can extend our industry. I think the software industry is going to be bigger and broader and do more this year than ever before. Not just Salesforce, which is going to grow incredibly this year. I think every company is going to grow because we have more to sell and there's more excitement and action and energy. And so this kind of counter narrative of oh no, no, no, they don't understand. We're just call that company who wrote that report and you ask them what is their software infrastructure, their magic infrastructure. I do this all the time. Tell me exactly how you're doing, what you're saying that you're doing. Oh, well, you're right. We're so sorry. We made a mistake. No, you know, and look, the futurist, I think we talked about this. Peter Schwartz, our chief futurist, he wrote Minority Report. Have you seen that movie? Like 20 years old. Great.

B (160:57)

I haven't seen it, but I've seen it.

D (160:58)

Oh yeah, you got to watch this. And also he wrote War Games and Deep Impact. You know, part of a team writing team. These are future movies. Yeah, but we all know where the future is kind of going. This highly automated, amazing world. But we're living in this world. And this is, this is this year. This is 2026. You know, we're running our business today day. So how are we doing our financials, our hr, our customer information? You know, how are we doing all of these aspects of our business? How are we running them? And then we write them. We write a report that sounds like Minority Report. And then I'm like, yeah, Minority Report. I read the movie. I watched the movie. Great, guys, fantastic. But I'm in the present moment reality right now, you know.

B (161:49)

No, no, it's true.

D (161:50)

And come, let's come back to world. And by the way, like you can do things that you couldn't do before this quarter. I released our new ITSM product so our customers can do this incredible thing. You know, IT service management, they used to have to go to ServiceNow for that. Now we converted five ServiceNow customers, you know, just in the quarter right over to Salesforce because we have that new capability. So companies like Sunrun and Cornerstone and Coolsys and others, you know, they can now use Salesforce ITSM instead of ServiceNow. That's so exciting. And then we have our new life sciences cloud that we've built, all with an agentic interface so our customers do not have to use Viva. And those customers are instead, you know, those are big companies like the Pfizers and the Takedas and the Novartis and the Abbeyv. So they're running with this next generation platform of apps and agents. Apps and agents and humans and agents working together.

B (162:46)

Yeah. So I mean the business model currently is clearly working. The results show that. As you look into the future, do you think the business model will evolve? Do you think that we're going towards more consumption based? Is seat based going to be with us forever? Like, how are you thinking? Obviously you don't need to turn the cruise ship today, but how do you think this evolved over time?

D (163:08)

It's such a great question. Right. Because that's like one of these interesting narratives. But hey, I don't know which, which anthropic product you're using, but the one I'm using is seat based. So if you don't have, if you have a different one or I have a, I don't know which OpenAI product you're using, but mine is seat based.

B (163:26)

Yeah, I have a seat.

D (163:27)

I don't know which one you're using.

A (163:28)

It's a good, it's a good take. Of course you can use API and

D (163:32)

there's an API all also. There's an API also.

B (163:35)

Yeah.

A (163:35)

But they're happy to be selling seats right now. You're right.

B (163:38)

Yeah.

D (163:39)

No, we, I, this is about humans and agents, so let's use that analogy.

B (163:43)

Sure.

D (163:43)

Humans are seats and so there's still like us three. We're like the last three humans. It's very sad, but we're still here. And then we have the agents too and they're using the APIs and talking to each other and they're on moat book and they're like having a conversation, creating their own currency.

A (164:01)

They're talking about us. They're talking smack.

D (164:03)

They're saying, oh, can you believe those humans are still there? We're going to get them. Yeah, it's like, no, it's still humans and agents working together and that is what is exciting about the future of enterprise software. So yes, I have a Salesforce, but it's extended with sales agents and I have a service organization extended with service agents. And I'm sending a trillion marketing messages this year. But they're all, all extended with marketing agents and I have commerce agents and I've got. Yeah, and you guys use Slack every day, right? And you're using Slack bot now. Hopefully since the last time, I would say it's like amazing that I can use agents in Slack. So I have employee agents and we might have some Other new exciting agents coming in the next few weeks for you. I'm very excited.

B (164:46)

Yeah, yeah.

D (164:46)

I'm very inspired by Maltbot. I've now got our team working on some new things. So that is like how I see it unfolding. And I think it's about a world where there are apps and agents and where this large language model, it's extending our capability. It makes us better, makes us stronger, it gives us the ability to do more. And yes, seeds still exist. And also consumption exists. Like we have lots of consumption products, data products, analytics products.

B (165:16)

Was there ever a SaaS apocalypse that was driven by fear that open source software would defeat your products, defeat Salesforce?

D (165:27)

Oh, no, there was a bigger SaaS apocalypse than that.

B (165:29)

Okay.

D (165:30)

Do you remember the SaaS apocalypse of 2020? The SaaS apocalypse of 2020, John, let me tell you the story. We were all minding our own business and then CNN came on and said we were all going to die.

B (165:44)

Yeah.

D (165:45)

Not just the software companies, everyone. Because it was the pandemic.

B (165:48)

Yes.

D (165:48)

And we all went home and we hid and it was a very sad time. And in fact, right at that moment, the stock market crashed because we were all going to die. And that was a huge SaaS apocalypse. And you can see it in everyone's stock chart. It goes like this and.

B (166:02)

Yep.

D (166:03)

And then all of a sudden people go, I guess we're not dying. And then it came back up. And that was a SAS apocalypse of 2020. It's a sad tale, but we lived through it and we got through it. And you know what? We're stronger for it. And that was just one of the SASS apocalypses.

B (166:19)

Yeah. Talk about the reception of the super bowl ad.

D (166:23)

Well, listen, I don't want to be competitive with you guys because your ad was very good.

A (166:27)

I want you just to know you

D (166:31)

did a great job. You should feel great.

B (166:33)

I appreciate that.

D (166:34)

You know, you had a great ad. Everybody loved it.

B (166:38)

Very high now.

D (166:39)

We were the number one ad in the Super Bowl.

B (166:42)

Okay, MOG meter going up. Let's go. Admog.

D (166:49)

I. I'm sorry, but it has to be said because, you know, we have Jimmy Donaldson. Mr. Beast.

B (166:55)

Yeah.

D (166:56)

And Mr. Beast did a great job. And that was just a killer ad. And we still haven't revealed the final thing. We have a great person here, John Zismos, who did a fantastic job with Jimmy, and it was an unbelievable partnership, but Jimmy's just. He's a force of nature.

B (167:11)

Yeah.

D (167:11)

I mean, I've never seen anything like this. He's such A young, great, amazing entrepreneur. The first time I ever talked to him, which was years ago, he said to me, I want to be the future Steve Jobs. You know, I want to be the future great entrepreneur of the world. I had to pause and say, really? He's like, yeah, that's what really I want in my life. I want to be a great, great business leader. A great. And I think he's doing it. And he's so young, you know, we've already had him on the COVID of Time magazine once, you know, and I see him as a huge leader in the whole world, not just as some kind of YouTube personality. This is a great entrepreneur, business person. Not just making chocolate bars, not just running a bank. I see him doing a lot of amazing things and he's got a lot of energy. Incredible. And he's. I have a lot of respect for him. And yes, number one super bowl ad. How about that?

B (168:02)

There's a lot of entrepreneurs listening, a lot of entrepreneurs that are searching for their next great business idea if they want to go swimming with dolphins and get inspired. What's the most underrated time to visit Hawaii?

D (168:14)

You're right. Well, number one, this week we have been having an awesome show from Kilauea Volcano. Oh, yeah, we had, I think 12, 13, 1400 foot tower, which is. I'm in Salesforce Tower, San Francisco right now. Gorgeous. You guys should come. And the. The fountain from the volcano was taller than this tower this week. So that is amazing. And you're right. All the little dolphins were so happy cruising around. But also, it's whale season on the north coast. They were going. They were so happy also. Everyone gets happy when you see the volcano. Volcano. It's like the way it's like when you see a whale, it reminds you who you really are. It reminds you this is what life is really all about. You come back to your breath. You come back. I don't know.

A (169:04)

When I see whales out, I think of agents. I think agent fours, personally. Hey,

E (169:11)

you gotta run.

B (169:12)

Sorry, we're keeping it late.

A (169:13)

I have a challenge for you. Before we talk again, you gotta frame mog, one of the AI lab leaders.

B (169:19)

Oh, this is. This is Big Alpha right here.

A (169:22)

We'll talk with you.

B (169:23)

I think it's possible. We'll coordinate. We'll coordinate. Have a great rest of your day. Congratulations on all the projects. Thank you so much for coming back on the show.

D (169:30)

You better listen to Metallic. I swear to you, I bring Mars on here and you're not ready. You're just lucky I didn't Bring him on today.

B (169:38)

On repeat. On repeat. We'll talk to you soon. Great to see you. Have a good one.

A (169:42)

Cheers.

B (169:42)

Goodbye. Well, if you're tracking earnings, you should be doing it on public.com investing for those take it seriously. Stocks, options, bonds, crypto, treasuries and more. With great customer service and without further ado, we will kick off the Lambda Lightning round. Let that cloud ring out. The Lambda Lightning.

A (170:04)

What's happening?

B (170:05)

How you doing?

C (170:06)

Hey, guys. How's it going?

B (170:08)

We are here doing great. Welcome to the show.

A (170:10)

Tough act to follow with Benioff.

B (170:13)

I don't know how he gets it

A (170:14)

up to 11 every time.

B (170:15)

Can you do any animal sounds? He's got the dolphin down. He's got the whale down. This is a unique ability. I didn't know it was necessary.

A (170:21)

We're not going to have you make any whale sounds. But it's great to meet you.

B (170:27)

Great to meet you. First time on the show. Please introduce yourself and the company.

C (170:30)

Great to be here. So my name is Michael Amenapat. I'm the co founder of a company called Row Space. Yeah. And Rosepace is an AI platform for asset managers. We help our customers use their institutional memory. So that means all their proprietary data, their accumulated judgment to make decisions faster. So what that looks like is we actually plug into all of their internal systems. This is not just documents, but it's databases, CRM, accounting and trade information. And we use agents to understand all the connections in that data, the inconsistencies, the conflicts, so agents can reason over it holistically. The idea here is that data is fragmented and is siloed and it's hard to make sense of in a single context window.

B (171:16)

Yeah.

C (171:16)

But using this or having work on this in advance is very helpful.

B (171:20)

Yeah. How jealous should every asset manager be of Bridgewater for collecting so much data over so many years? Do you have a view on the Bridgewater strategy? Can you explain what they actually do and whether or not you're a fan of it?

C (171:41)

I can't speak to Bridgewater specifically, but I think you land on our key point here, which is that public data is getting increasingly commoditized and actually AI is accelerating the commoditization of public data. The more you have tools like Claude that can synthesize anything from the web or from public tools. Less edge there, is there? So the best edge firms have now is decades of data accumulation and their own insight and judgment as encoded in their data. And our whole goal is to help our customers tap that in almost every customer Conversation we have, there's a line that's something like, we are sitting on enormous value in our data if only we could get at it or only we could find it. And that's what we're trying to do. So I think I will accelerate this dependence on your internal data and proprietary data as public data becomes less and less valuable.

B (172:40)

Yeah, talk to me about what, what your product actually looks like once you roll it out. Because, you know, a lot of hedge funds and asset managers, they do a lot of back testing. And I could imagine going back and running a report like back test the thesis that we talked about, but maybe we didn't actually implement. There's a whole bunch of ways that you could just do reporting, but then you could also go and say, hey, you know, just turn this thing loose, go trade for us. Like, where are we on that continuum?

C (173:08)

There are a lot of great points there. So the first thing is we are not trading or taking action automatically. The idea here is that we will help these firms consider the full amount of data that's possible and then they make the decision. But our goal really is it used to not be possible to say, review every single name in the universe. If you are about to consider a trade to rebalance your credit portfolio. Now you can with rose base, but it's still the human who makes the ultimate call there. So that's one thing. The second thing is on backtesting. I love that example because it's actually been surprising how little post hoc analysis people do of their investment or trading decisions just was our thesis a year ago, two years ago, 10 years ago, what actually happened? How does that inform what we're doing in the future? That's just a time consuming thing to do and people don't do it very much. But we work with a private equity firm now. It's been in business for 50 years, a pioneer of the field. And they actually run every deal through real space along the lines of, given our 50 years of history, what should we do here? What have we done? What are the risks? And that kind of analysis not have been possible before.

A (174:25)

Yeah, what, what is your competitive positioning like with the labs? They're all hiring consulting firms, they have forward deployed engineers. I'm sure they're pitching a lot of the same companies. And I can imagine how you would position row space against building something internal or just leveraging the applications that the labs are building. But how do you sell it?

C (174:53)

So on the apps that are out there now, from labs and from other players in market, what we see is a focus on time savings. So faster models, faster decks, faster summarization of meetings and research. And that's obviously extremely valuable. But our focus has been on decision making. So what are the things you should be looking at which would have been impossible to consider before the Rosemary's going to help you with. So in some sense there are things that our customers do with aerospace today which they did not do before. It's not about saving them time doing work that they've done in the past. So that's one, one chunk of it. The other chunk is we are talking about in some sense the crown jewels for our customers. All of this trading data, position data, they're thinking of theses. This is extremely sensitive. So we only deploy in our customers environment. We never as a company actually take possession of their data and we do all of this processing in their environment. So we have a security challenge, an infra challenge and I1 and I think the sensitivity to the security compliance and other constraints of finance is a big differentiator for us.

B (176:08)

Take us through the fundraising round. What happened?

C (176:11)

So we've done two rounds over the past 18 months. 50 million in total emergence capital joined in the end. And we've had. My former bosses were on your show on Monday and they were a big player in both the state, an MBA at Stripe. So it's been really fun because every major investor on our cap table I've been close to for at least half a decade. So gets to be a bit of this bring back the old gang together.

B (176:42)

That's amazing. Well, we're actually.

A (176:43)

Where's the company based?

C (176:46)

We are based in San Francisco, but a big push for the funding this year is to expand our New York presence.

B (176:50)

Of course. Yeah, that makes sense. Well, have a great rest of your day. Thanks so much for popping by to tell us about the business. Very fascinating. Good luck.

C (176:57)

Thanks for having me.

B (176:58)

We'll talk to you soon.

A (176:59)

Cheers.

B (176:59)

Goodbye. Let me tell you about Gemini 3.1 Pro. Gemini 3.1 Pro is here with a more capable baseline. It's great for super complex tasks like visualizing difficult concepts, synthesizing data into a single view, or bringing creative projects to life. And without further ado, we'll bring in our next guest, Adam Warmoth from Chariot Defense. How you doing, Adam? Yes. Good to see you again. Welcome back. Welcome to the show. Let's kick it off with the news. What happened?

C (177:27)

Yeah.

B (177:28)

Today we announced our $34 million Series A. Yep.

E (177:31)

Let's get the gong.

A (177:34)

We got a bigger gong since the last time you were On.

B (177:36)

We did. We did. Well, that's impressive. Welcome back.

A (177:40)

Not as impressive as a $34 million series today, but it's up there.

B (177:45)

Yeah. Take us through the shape of the business today, the key customers, key products, and sort of what changed since the last time you were on the show?

C (177:55)

Yeah.

B (177:56)

Awesome. So last time I was on the show, I was just getting done with our first transformation and contact exercise. You guys had had Dan Driscoll, Randy George in the show. They'd kind of talked about that. We were just coming back from our first participation there. That was our second test event. We've done about 25 since then. And so we've got systems deployed in

E (178:15)

pretty much every theater across a bunch

B (178:19)

of different army units, Marine Corps units, and really just starting to see the

E (178:23)

traction and demand for the systems grow.

B (178:25)

As we do see things like drone dominance happening, as we see things like next generation command and control. All of those systems are fielding and running into issues with that power infrastructure layer. And we've been able to fill that gap and iterate quickly, kind of working

E (178:39)

with the warfighters, working with the soldiers

B (178:42)

and getting the systems out there and

A (178:44)

assume that someone listening today didn't catch your first appearance, catch us up to speed on the shape of the product and all that.

B (178:54)

Yeah. So effectively what Chariot's building is the power layer for robotic warfare. So really you wouldn't send a soldier

E (179:01)

into the fight without food and water and nicotine.

B (179:06)

You wouldn't send a robot into the fight without communications, compute and power. Cool. And so we really see that as one of those core infrastructure layers behind kind of this defense modernization. You know, Anduril's building some great systems in the compute space. Palantir really dominating the network space. And we're kind of building that third missing layer. And so effectively what we're doing is taking the technology coming out of companies like Tesla, Apple, Lucid, Rivian, Archer Joby, High voltage batteries, silicon carbide power electronics. If you've read Paki McCormick's the Electric

E (179:37)

Slide goes into detail on kind of

B (179:39)

major transformations happening in the commercial industry on that core technology stack. We're taking those and lifting and shifting them into the defense platforms to build hybrid high power systems. Yeah, walk us through exactly what needs to happen to deploy a high voltage battery on the battlefield. I think most people will be familiar with like the Tesla powerwall. And we've all seen like the IBM tough book. You put some rubber corners on it and give it a nice graphite case. And I imagine there's a lot More going on. So what's the state of the art? Yeah, that's a great question. So we really kind of take the best of commercial technology. Our first product we deployed M4 24 is literally in a pelican case. Okay. So we took a pelican case. You were saying, hey, why reinvent the wheel on just some of that core rugged technology? We do some additive manufacturing to create these kind of internal bulkhead structures. Sure. To kind of isolate the electronics and then we integrate the batteries, the power electronics, the microcontrollers into that in a form factor that can be left out in the rain and mud can be dropped off the back of the Humvee. And when I say can be has been and tell the tale. Yeah. So you've been through testing. I assume you've done some sbirs. Are you moving towards program of record or are you just sort of in the supply supply chain for other companies that might be primary contractor? Yeah. So we've got a split go to market model. One being directly to the government, both bottom up selling directly to units and top down firm record and then also

E (181:08)

selling to other companies as part of a broader kit.

B (181:10)

Cool. So the inspiration for Chariot was I was the counter UAS program manager at Anduril. We were constantly running into power problems. Right. So that idea of selling this as part of a power kit that's enabling

E (181:21)

other systems is another part of our

B (181:23)

go to market model.

A (181:26)

Where do you stand on the verticalization debate? We had Mike from Also Capital.

B (181:33)

Oh yeah.

A (181:34)

He kicked the hornet's nest because he was basically saying like, yeah, it's great to verticalize, but there's some businesses that you can just buy a lot of components off the shelf and make a great product, prove that people want it and then do it later. A lot of people, I would say most people were disagreeing with that. But there's. Every business is different.

B (181:58)

Yeah, I am, I'm going to, I'm going to come in here on T, Mike. So you know, we've really been able to leverage the supply chains from companies like Tesla. Right. And Apple and Archer where you have actually mature commercial technologies around these core components around batteries and power electronics. What nobody has done is kind of gone and done that forward deployed engineering. And so Erin Price Wright, who led

E (182:21)

around, I think in her post, said

B (182:23)

we should actually call Adam, the chief forward deployed engineer. That's really what I've been doing over the past year. And it's that forward deployed engineering model that kind of maps to what Palantir

E (182:32)

and Anduril did as well.

B (182:33)

So Palantir didn't invent cloud compute. Right. Or big data models.

C (182:38)

Right.

B (182:38)

That was tens of billions of dollars of investment from Silicon Valley companies. And then through good go to market, good forward deployed engineering brought that into the department. Anduril did the same. The first sentry tower was really enabled by the Autonomy technology developed by the self driving car industry. Computer vision went from an unsolved problem in 2014 to just download yolov4 in 2017 and they were able to capitalize on massive investment from the self driving car industry. And just through good Ford deployed engineering,

E (183:09)

good go to market, bring that into

B (183:10)

the department and that's what we're doing. For all the technology coming out of electric vehicle electrical transportation space, what are you most excited about in defense tech? There's obviously a lot of the Anduril products people are aware of. There's this big small drone boom. Is there something like what's the next big defense tech trend that you think is going to become really important? Yeah. So there's going to be a little bit of a self serving angle here. But what we're really, you know, I spent years doing counter UAS counter drone systems pre Ukraine. Right. So counter UAS is a big topic. When we were working with it on with SOCOM in 2021, not that many

E (183:49)

people were talking about it.

B (183:51)

And that insight around what it actually takes to do counter us at the edge is really kind of what inspired Chariots. So there's a lot of focus on drone dominance. But countering these small drones is going to require pushing more sensors and more countermeasures onto every mobile platform. And what that's going to mean is every mobile platform needs more power to power those sensors to be able to turn the engine off and manage signature and hide. Sure. So avoid detection in the first place and then be able to drive big surges of power to do things like electronic warfare or high powered microwaves or high energy lasers. So we've done tests with high energy lasers already as one of those counter us technologies that needs that big surge of power and that's really where batteries come in. We want to get shirts that say we heart diesel. We're the most diesel loving battery company out there. Hydrocarbons are incredibly energy dense, but batteries give you is that ability to surge that power or the ability to dial it down and hide your signature. And that's really the differentiation of the Chariot. So why can the Tesla model plaid go 0 to 60 in under two seconds? It's a Surge of power. And that's what's unique. Thank you so much for coming on the show.

A (184:56)

Congratulations, insane progress and we're excited to

B (185:00)

see you on the show soon. We'll talk to you later.

A (185:02)

Cheers.

B (185:03)

Have a good one. Let me tell you about CrowdStrike. Your business's AI, their business is securing it. CrowdStrike secures AI and stops breaches. And we have Connor Sweeney from BABA in the restream waiting, coming in. Bring Connor into the TV panel. How are you doing?

A (185:17)

Great, Seta, how are you?

B (185:19)

Thanks so much for stopping by. Please introduce yourself and the company.

A (185:25)

For sure. No, thanks for having me. I'm Connor of baba. But we connect older adults and their families with patient advocates. So human pitch, refreshingly, but it really moves a needle because these folks do all the care coordination, the scheduling, kind of connecting the dots for healthcare companies.

B (185:43)

Yeah. What was the origin story of the business? Were you just looking at the growth of demographics in the. Did you have a personal story? What led you to this particular market?

A (185:55)

Yeah, for sure. I think most healthcare people end up coming into it after experiencing it themselves. And that was also the case with me. My grandma had a stroke and I did out of that care coordination. But I think what's cool is that I'm not from the healthcare world originally team isn't from the healthcare world. And that kind of lets us move a little bit faster in different ways, which has been working so far. But yeah, she had a stroke that left her unable to speak. So I actually started by trying to build her little AI tools for her speech therapy and that led to this.

B (186:28)

Yeah. So what's the shape of the business today? Is this like a multiple sided marketplace at this point? Like, who are all the customers and suppliers that you work with?

A (186:38)

Yeah, for sure. So we connect, like I said, the older adults in our family, I would say they're like our patients.

E (186:43)

Right.

A (186:43)

Our customers. And we connect them to our advocates who, you know, we have dozens and dozens of folks.

E (186:48)

They're contractors.

A (186:50)

But what's really nice is that insurance usually covers almost all of the cost. So we work with like Medicare, Medicare Advantage plans, even Medicaid in certain states. So that like 98% of our patients or end customers get this for nothing, out of pocket, which is really cool.

B (187:07)

Yeah. How do you get the end customers on board? I feel like you could run a

A (187:13)

bunch of advertising on Fox News.

B (187:14)

Yeah. As a tv, like. Because I just feel like everyone is like, you know, doing viral launch videos, going after the early adopter tech audience, but you're going after a very different audience and I feel like your growth mechanism is going to be interesting to hear about.

A (187:28)

It's honestly more fun too because it is an audience that like watches like remnant TV and like the scrolling bar at the bottom, like the Netflix, like state is actually apparently really high converting. We do stuff like flyers, of course, we're on Facebook, Google, the whole nine yards there. But we also partner with a lot of like B2B, you know, health systems. And so with like nursing homes and home health agencies, we often pick up folks that way too.

B (187:52)

Got it.

A (187:53)

How, how are you using AI at the product level? Is it helping make the care providers like a lot more effective? What does that look like? Yeah, for sure. Like our advocates all, you know, use AI where it's appropriate. A lot of the tasks that they're doing are repeated over and over again. We're constantly making enrollments into things like food stamp and Medicaid. We're constantly fighting insurance denials.

B (188:19)

Right.

A (188:20)

Writing like rejection letters, waiting on hold with doctors offices to make appointments. All those are like perfect tasks for AI, right? Like either OCR or voice AI. So there's a lot of like that really like, you know, technical work. But what's cool is that, you know, the patient to those customers, they really like just having a human that they call or text 247 that doesn't work for like the insurance company or hospital. That slight difference that our advocates are in their corner rather than some third party really seems to be better received.

B (188:54)

Take us through the fundraising news.

A (188:56)

Yeah, what's, what's the details? Yeah, for sure. So Germal Catalyst led our seed that happened, you know, just a month or two ago, but raised a little under 7. 7 million.

E (189:08)

And we have like a team of

A (189:09)

10 in New York right now. It's, it's pretty young company, but we're skizing.

B (189:15)

Congratulations.

A (189:16)

Also, can't, can't forget about genius. My neighbor, my neighbor Ben. Oh, really is also.

B (189:22)

Oh, okay. There we go.

A (189:26)

Ben and Adam.

B (189:27)

That's fantastic. Well, thank you so much for coming on the show. Have a great rest of your week and we'll talk to you soon.

A (189:32)

Cheers.

B (189:33)

Goodbye. 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. And without further ado, we have Matt Harp from basis in the Restream waiting room. Let's bring him into the TVPN Ultradrome. How are you doing, Matt?

A (189:52)

What's going On.

E (189:54)

Hey guys, how's it going?

B (189:55)

It's going well. First time on the show. Please introduce yourself and the company.

E (190:01)

Yeah, well first, thank you guys for having me. Appreciate it. Matt, one of the co founders of Basis we are fully in person company in New York. Yes, 100% here. No office in SF yet.

A (190:14)

You got to meet Keith Raboy. He's going to love. He's a new I'm K. We got Keith up here a lot.

E (190:23)

He's here playing great. But no, we are an AI platform for accounting team. So we build Long Horizon agents that are able to automate various parts of accounting workflows and we have a couple years.

B (190:34)

Yeah. How narrow of workflows are you talking about? Is this a tool that integrates into other ERP accounting systems or do you want to replace everything? How do you think Think about the actual go to market and then the product fringes.

E (190:51)

Yeah, it's a great question. Our objective is not to rip and replace any software. Our goal is solely to sit on top of other pieces of software and sit in the workflows that people are already doing and then say really no one should be doing any work. We should take the doers of the work and make them the reviewers of the work and have the AI perform the first pass of all the various things that might need to be done. So that means we need to integrate it into various systems. But it also means that from a deployment perspective, you can easily stand up basis in your environment and put it to work without having to rip out existing software.

B (191:21)

And where like a lot of people have tracked the progress in software, most software, like when you say the doer of the work turns into the reviewer, that feels like what's happening in agent encoding right now? Where in accounting, how far along are we would you entrust a reviewer to not actually do any of the work on, I don't know, your personal tax return or something, But I don't know what the toy example is. But how close are we to this situation? Like what we're seeing with coding where people say I don't write any code anymore. When will we see accountants say I don't do any of the work, I just review the work.

E (192:04)

Yeah, it's a great question. I think it depends a little bit on the complexity of the task. For instance, there are some things that the basis has been able to do for quite a while where I think accountants have already sort of built up trust in basis ability to do these things. And so in those situations there may be spot checking. But they may not be reviewing quite as actively. And then there's some things. For instance, we just released yesterday an example of the first AI that's able to complete an entire, you know, corporate tax return workbook end to end by itself, which I think is a remarkable milestone in the, in the accounting world. And that's something where that's a capability that's only existed for the last couple of months. And so, you know, that capability, obviously you'd apply a different sort of level of review to those types of situations. The interesting thing is that accounting is actually built for these types of review. So accounting, obviously accuracy is extremely important. Humans themselves make lots of mistakes. And so accounting already has multiple levels of review that are built into processes. Which means that when accounting accountants are thinking about how to deploy basis, they can kind of apply that same review mentality to incorporating basis into their workflow. Think about what are the areas of highest risk, make sure that you can see it doing the work in the way that you think makes sense first. And then as you build up trust, you sort of, you know, I can adopt the way that you actually think about reviewing the work. And then to the point on coding, I think it's interesting because accounting is obviously not a text in, text out discipline. You know, it's not something where there's tons of data on the Internet about, you know, various accounting workflows. And so I think it doesn't quite, it's not quite as immediately obvious how to use AI to help with those workflows to the way, to the extent that it is in legal or in coding or other areas where elements out of the box are clearly very good. And so I think it has taken some time for AI to get good enough to the requisite level of accuracy and over sufficiently complex tasks for it to be extremely useful. But I think that's kind of flipped over the course of the last year. And I think what we've seen play out in, in sort of software engineering over the last year will play out in accounting in the company.

A (194:01)

What do various players in the accounting world, how are they processing AI over overall? Do they think that the fee model will have to change? I remember, I forget there was one auditor that was getting mad at their auditor.

B (194:16)

Oh yeah.

A (194:16)

For saying like, hey, you're using, we know you're using AI, you got to give us a better price. And they're like, wait, you're an auditor?

B (194:24)

Yeah, it was the Spider man meme.

A (194:26)

But yeah, how are people processing it?

E (194:29)

Yeah, it's a great question. I Think one of the interesting dynamics about accounting that I think is not true for a lot of other professions or areas of professional service is there is a very dramatic shortage of accountants in the US and so when we started out serving a number of our customers, they were not necessarily interested in doing AI for the sake of AI. I think over the course of the last year that's flipped and everyone's board says what are we doing from an AI perspective? And people want to have AI initiatives. When we started in 2023, that was not the case at all. What was the case though is that there was a huge shortage of accountants. And I think roughly 300,000 accountants left the profession over a two year period in and around Covid. You never know what to make of these demographic estimates, but they say roughly 75% of accountants will retire over the course of the next decade. And there are very few folks joining the profession to make up that gap. And so that means that if you are an accounting team or an accounting firm, one of the core things you have to solve for is how do we continue to do the work that we need to do with the people that, that we can hire. And, and so one of the things that basis enabled firms to do is to, to take on a lot of the work that otherwise they would not be able to get done because of the shortages that exist. And so I think that dynamic has made the adoption maybe happen quicker than it would have otherwise have if that shortage didn't exist.

A (195:44)

Yeah, it's fascinating. On the other side in law, from what we've seen, you have tons of people applying to law school simply because, maybe just because they don't have anything better to do, which implies like we may have, you know, a huge influx of lawyers, but accounting, not as prestigious, hasn't. Yeah, I can, I can see why it hasn't attracted the same influx. But yeah, I think even, even as I think about it, like all these

B (196:11)

kids just want to be asked.

A (196:11)

I've had, I've had, you know, working with a number of firms over the years. It's always like super annoying if the partner or associate more, more so on the associate side of an accounting firm leaves. And then there's all this like context that's lost and you really like, if, if an agent had been helping with that entire process, it would be a lot smoother. So I'm going to recommend basis to our firm.

B (196:37)

I have one more question. We'll let you get back to your busy day. How do you think about harness development and throwing the Context back to the human in the middle. Because it sounds great to be like, okay, I'll turn this agent loose and it'll come back to something I can review. But a lot of these things when, when I'm interacting with great accountants, it's not just sum up all the bills and you have your total revenue or something. It's like, okay, I found a bill, I don't know how to categorize it. Is this capex? Is this opex? Was this a business expense? How do we reclassifying this? And sometimes that data can just be agentically go and look at the receipt, figure out what happened, figure out how to classify it. But a lot of times it requires an interaction with a human either over slack or over an email or something. And so I imagine that deciding how to be persistent but not annoying, is that a big challenge that you're actually working on or do I have that kind of roadmap wrong?

E (197:37)

Yeah, 100% very important part of things, like part of doing accounting work is the world is this very messy, complicated place where all sorts of economic activity is happening and you need to figure out how to get that information in order to probably account for things when you were not present at the time of some kind of economic interaction. So very important. And I think it is one of the challenges of, you know, we spent a lot of time thinking about these long horizon agents because, you know, because of the nature of accounting work, you know, this sort of chat bot experience is not one that naturally works. And so we've been focused on developing these agents that can sort of work on more complex tasks over longer periods of time. And one of the challenges there is you have to, to figure out as the agent is going about doing its work, when to interrupt and ask human a question and also how to provide the human insight into what's going on. Let's say you're going to do a task and it's going to go off for the entire day. You know, the human can't find out at the end of the day that it did something that wasn't exactly aligned with what they wanted because then the whole day has been wasted and they need to deliver something to their client. So figuring out how to both give visibility and then also how to raise concerns to the human user is extremely important. And sometimes you even need to go directly to a different source, you need to send an email or take some other action in order to complete the workflow. So I think it's an essential part of

A (198:52)

super competitive hiring market. Right now give us your 60 second elevator pitch. If you're competing to hire somebody with the big labs or somebody that might join Harvey or, or Cognition or any of these other companies, how do you, how do you close them?

E (199:11)

Yeah, well, I mean, look, there's lots of exciting things going on and lots of great opportunities out there, but I think there are a couple of things that we think are critically important and it might not appeal to everyone, but certainly a certain type of person. I think the first is that, you know, we are pretty much solely focused on building the most capable, most accurate long horizon agents. And I think we're at the frontier of that work. We work very closely with the labs that are building the most sophisticated reasoning models and we have for quite a while. And so I think for folks who are interested in figuring out how can we build the most capable, most autonomous systems and apply them to actual, to real work in the real economy, that is part of the work that we do, or it's really the core part of the work we do that I think is methodologically extremely interesting. I think the second thing is that we are a fully in person team in New York and all of our research, all of our engineering takes place here. And I think there are very few other companies that are on the frontier of applied ML and are also fully in New York. Obviously there are limitations relative to the Bay Area in some ways, but I think there's amazing ML talent here. And so we are sort of hopefully becoming the home for applied ML talent in New York City. And then I think finally accounting is just a hugely important part of the economy and how we operate as a society. We sort of see it as the fundamental way in which we understand economic life, in which we sort of structure and systematize all the economic activity that happens. And then it has tremendous implications downstream in the organization for how people make decisions. And, and, and we think there's a huge opportunity to do way more counting than we're doing right now. And that could actually help organizations function in a much more effective manner. We actually think it has an interesting parallel to engineering in a certain sense, which is that accounting is about, you know, how do you systematize and abstract this very complicated world? In some sense that is like the practice of software engineering as well is like, how do we understand, you know, this very complex domain and reduce it to a piece of reliable software that we can use. And so I think it's much more important than people think. And I think it actually is very sort of compatible with the Ways that, you know, engineers and ML folks like, like to think as well. So those are the things that we sort of think are, are most important. And yeah, we're really lucky to have a great group of folks here at

B (201:24)

basis and you got some money to hire more people. Take us through the fundraising round. What happened?

E (201:30)

Yeah, so I think when we're thinking about raising money, one thing that never changes is the strategy and what we're trying to execute from business perspective. We have at actually at our first seed round and then at every successive round written a quick memo or cover letter. It just outlines the strategy and the strategy which is that, you know, we are trying to build these long horizon agents to solve the important, important problems. Accounting has not changed. Some of the core principles around how we make decisions have not changed. So we actually try to, to make sure that when we raise around, it doesn't actually change the strategy of what we're doing in a meaningful way. Like we think that you should be relatively consistent, you should obviously update as you go in that respect. What it does allow us to do is it allows us to keep growing the team in the ways that it needs to grow. And so, you know, that's mostly what we bought the additional capital on in order to do. There are so many different domains of accounting. There's so much ML and engineering work that needs to be done. No matter how much we use AI internally to make all those things more efficient, there's just endless things left to do. And so we just felt it was the right time to do it. And we're very thankful that we have a great set of investors around the table to allow us to keep making that.

B (202:33)

And how much did you raise in this most recent round?

E (202:36)

We raised 100 in this round.

C (202:42)

Massive.

B (202:42)

Thank you for coming on the show.

E (202:44)

Massive.

B (202:44)

How have a great rest of your day and we will talk to you soon.

A (202:48)

Yeah, great to meet you.

B (202:49)

Goodbye.

A (202:50)

Cheers.

B (202:51)

Well, I need to hop on with the English countryside soon, so are there any more news stories that we should cover before we plant the bomb? Figma director Andrew reed just bought $36.5 million worth of figma. The largest ever insider by a Figma. Very exciting for him. He's going Longma on the Figma Dreamweave. This is a stretch. People were debating back and forth with Also Capital founder Mike who came on the show earlier. I think he had a really good point. I like his point. I just think it's funny that Yimbieland ratioed us into the stratosphere by posting no you can't just vertically integrate like that. Meanwhile, in China, byd, I guess we do in ships now. I had no idea. I saw the BYD logo on a ship. I didn't realize that they made the ship. I guess they make everything. I guess they make cars and monorails too. Absolutely insane. But 10,000 likes on this. And although it's like, I don't know, it's a dunk ratio, whatever, it is an inspiring message. If they can do it, why can't we? So just do it. Just go build a supertanker, I guess. Figure it out. Anyway, anything else from the timeline that you'd like to talk about before we call it a day?

A (204:21)

There's a lot more.

B (204:22)

There's a ton more. Mark Zuckerberg is planning a stablecoin comeback. They also have a banger deal with AMD going on. And if you head to the bar this weekend and you drink too much, you should just say that you were the victim of a distillation attack. That's the correct turn of phrase. Anyway, thank you for watching Leave us. Five stars on Apple, Podcast and Spotify. Have a wonderful day. Nice work, brothers. I'll see you on the next one.